{
  "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": 0.07273956,
      "auxiliary_loss_mlp": 0.02777065,
      "balance_loss_clip": 2.56119728,
      "balance_loss_mlp": 2.41295195,
      "epoch": 0.00012024289063909097,
      "flos": 24932483919360.0,
      "grad_norm": 40.68027152993328,
      "language_loss": 2.58202171,
      "learning_rate": 0.0,
      "loss": 1.92230916,
      "num_input_tokens_seen": 20375,
      "step": 1,
      "time_per_iteration": 15.587866067886353
    },
    {
      "auxiliary_loss_clip": 0.0487144,
      "auxiliary_loss_mlp": 0.01920403,
      "balance_loss_clip": 1.6983521,
      "balance_loss_mlp": 1.66558146,
      "epoch": 0.00024048578127818193,
      "flos": 30664624377600.0,
      "grad_norm": 65.9475370987879,
      "language_loss": 1.89461303,
      "learning_rate": 5.021476677069823e-07,
      "loss": 1.96253145,
      "num_input_tokens_seen": 39035,
      "step": 2,
      "time_per_iteration": 2.608912229537964
    },
    {
      "auxiliary_loss_clip": 0.04823905,
      "auxiliary_loss_mlp": 0.01929298,
      "balance_loss_clip": 1.70028806,
      "balance_loss_mlp": 1.67027998,
      "epoch": 0.0003607286719172729,
      "flos": 19026227969280.0,
      "grad_norm": 41.39223363717724,
      "language_loss": 1.61846948,
      "learning_rate": 7.958852231401551e-07,
      "loss": 1.68600154,
      "num_input_tokens_seen": 57600,
      "step": 3,
      "time_per_iteration": 2.4822475910186768
    },
    {
      "auxiliary_loss_clip": 0.0485371,
      "auxiliary_loss_mlp": 0.0188784,
      "balance_loss_clip": 1.70755577,
      "balance_loss_mlp": 1.63034749,
      "epoch": 0.00048097156255636386,
      "flos": 19316314206720.0,
      "grad_norm": 37.95815721231649,
      "language_loss": 1.64579117,
      "learning_rate": 1.0042953354139647e-06,
      "loss": 1.71320677,
      "num_input_tokens_seen": 76465,
      "step": 4,
      "time_per_iteration": 2.5513994693756104
    },
    {
      "auxiliary_loss_clip": 0.04915157,
      "auxiliary_loss_mlp": 0.01871863,
      "balance_loss_clip": 1.70707476,
      "balance_loss_mlp": 1.6240983,
      "epoch": 0.0006012144531954548,
      "flos": 13991264893440.0,
      "grad_norm": 55.04188963521227,
      "language_loss": 1.93796301,
      "learning_rate": 1.1659507774310057e-06,
      "loss": 2.00583315,
      "num_input_tokens_seen": 94350,
      "step": 5,
      "time_per_iteration": 2.718616485595703
    },
    {
      "auxiliary_loss_clip": 0.04848668,
      "auxiliary_loss_mlp": 0.01821063,
      "balance_loss_clip": 1.69924033,
      "balance_loss_mlp": 1.57654071,
      "epoch": 0.0007214573438345458,
      "flos": 23148988225920.0,
      "grad_norm": 45.03501123790868,
      "language_loss": 1.61063862,
      "learning_rate": 1.2980328908471373e-06,
      "loss": 1.67733598,
      "num_input_tokens_seen": 114595,
      "step": 6,
      "time_per_iteration": 2.7510430812835693
    },
    {
      "auxiliary_loss_clip": 0.04894783,
      "auxiliary_loss_mlp": 0.02071045,
      "balance_loss_clip": 1.82885456,
      "balance_loss_mlp": 1.89938319,
      "epoch": 0.0008417002344736367,
      "flos": 67663246170240.0,
      "grad_norm": 4.667284297429525,
      "language_loss": 0.81488907,
      "learning_rate": 1.4097067265369432e-06,
      "loss": 0.88454735,
      "num_input_tokens_seen": 179590,
      "step": 7,
      "time_per_iteration": 3.2143805027008057
    },
    {
      "auxiliary_loss_clip": 0.04793639,
      "auxiliary_loss_mlp": 0.01850414,
      "balance_loss_clip": 1.70103884,
      "balance_loss_mlp": 1.61065984,
      "epoch": 0.0009619431251127277,
      "flos": 21281381504640.0,
      "grad_norm": 41.052100316451245,
      "language_loss": 1.58436847,
      "learning_rate": 1.506443003120947e-06,
      "loss": 1.65080905,
      "num_input_tokens_seen": 195090,
      "step": 8,
      "time_per_iteration": 2.764202833175659
    },
    {
      "auxiliary_loss_clip": 0.04793213,
      "auxiliary_loss_mlp": 0.01899782,
      "balance_loss_clip": 1.70406175,
      "balance_loss_mlp": 1.65564132,
      "epoch": 0.0010821860157518186,
      "flos": 23331342597120.0,
      "grad_norm": 18.63701827223438,
      "language_loss": 1.47908437,
      "learning_rate": 1.5917704462803102e-06,
      "loss": 1.54601431,
      "num_input_tokens_seen": 211635,
      "step": 9,
      "time_per_iteration": 2.888375759124756
    },
    {
      "auxiliary_loss_clip": 0.04824306,
      "auxiliary_loss_mlp": 0.01839617,
      "balance_loss_clip": 1.7075727,
      "balance_loss_mlp": 1.5809803,
      "epoch": 0.0012024289063909096,
      "flos": 17010166337280.0,
      "grad_norm": 13.292274512484996,
      "language_loss": 1.52923918,
      "learning_rate": 1.6680984451379884e-06,
      "loss": 1.59587848,
      "num_input_tokens_seen": 224705,
      "step": 10,
      "time_per_iteration": 2.6970810890197754
    },
    {
      "auxiliary_loss_clip": 0.04769979,
      "auxiliary_loss_mlp": 0.01857849,
      "balance_loss_clip": 1.69867611,
      "balance_loss_mlp": 1.60893989,
      "epoch": 0.0013226717970300007,
      "flos": 21288133261440.0,
      "grad_norm": 13.697798883493817,
      "language_loss": 1.32565725,
      "learning_rate": 1.7371455188905097e-06,
      "loss": 1.39193547,
      "num_input_tokens_seen": 244635,
      "step": 11,
      "time_per_iteration": 2.8143815994262695
    },
    {
      "auxiliary_loss_clip": 0.04806897,
      "auxiliary_loss_mlp": 0.0193564,
      "balance_loss_clip": 1.69805253,
      "balance_loss_mlp": 1.67089987,
      "epoch": 0.0014429146876690916,
      "flos": 27237884935680.0,
      "grad_norm": 13.075737069763093,
      "language_loss": 1.25268102,
      "learning_rate": 1.8001805585541196e-06,
      "loss": 1.32010639,
      "num_input_tokens_seen": 265765,
      "step": 12,
      "time_per_iteration": 2.804831027984619
    },
    {
      "auxiliary_loss_clip": 0.04762425,
      "auxiliary_loss_mlp": 0.01873502,
      "balance_loss_clip": 1.69492292,
      "balance_loss_mlp": 1.63050556,
      "epoch": 0.0015631575783081825,
      "flos": 19062174504960.0,
      "grad_norm": 6.558112953375323,
      "language_loss": 1.28807759,
      "learning_rate": 1.8581671739548328e-06,
      "loss": 1.35443687,
      "num_input_tokens_seen": 283500,
      "step": 13,
      "time_per_iteration": 2.8082141876220703
    },
    {
      "auxiliary_loss_clip": 0.04786546,
      "auxiliary_loss_mlp": 0.01834578,
      "balance_loss_clip": 1.69853282,
      "balance_loss_mlp": 1.59215355,
      "epoch": 0.0016834004689472734,
      "flos": 48139473985920.0,
      "grad_norm": 6.487125271079463,
      "language_loss": 1.13495255,
      "learning_rate": 1.9118543942439254e-06,
      "loss": 1.20116377,
      "num_input_tokens_seen": 305685,
      "step": 14,
      "time_per_iteration": 3.971452474594116
    },
    {
      "auxiliary_loss_clip": 0.04719846,
      "auxiliary_loss_mlp": 0.01836415,
      "balance_loss_clip": 1.69412673,
      "balance_loss_mlp": 1.59856868,
      "epoch": 0.0018036433595863645,
      "flos": 34970026314240.0,
      "grad_norm": 5.56351985062006,
      "language_loss": 1.12693787,
      "learning_rate": 1.961836000571161e-06,
      "loss": 1.19250035,
      "num_input_tokens_seen": 327340,
      "step": 15,
      "time_per_iteration": 3.8638408184051514
    },
    {
      "auxiliary_loss_clip": 0.04626504,
      "auxiliary_loss_mlp": 0.01834721,
      "balance_loss_clip": 1.81113029,
      "balance_loss_mlp": 1.67908132,
      "epoch": 0.0019238862502254555,
      "flos": 59768284440960.0,
      "grad_norm": 3.854003309700207,
      "language_loss": 0.64718378,
      "learning_rate": 2.0085906708279293e-06,
      "loss": 0.71179605,
      "num_input_tokens_seen": 382710,
      "step": 16,
      "time_per_iteration": 3.3347461223602295
    },
    {
      "auxiliary_loss_clip": 0.04716698,
      "auxiliary_loss_mlp": 0.01823396,
      "balance_loss_clip": 1.69599605,
      "balance_loss_mlp": 1.58917415,
      "epoch": 0.0020441291408645466,
      "flos": 20814543417600.0,
      "grad_norm": 4.272719600475656,
      "language_loss": 1.16116059,
      "learning_rate": 2.0525099325728135e-06,
      "loss": 1.22656155,
      "num_input_tokens_seen": 400890,
      "step": 17,
      "time_per_iteration": 3.0268287658691406
    },
    {
      "auxiliary_loss_clip": 0.04538409,
      "auxiliary_loss_mlp": 0.0172483,
      "balance_loss_clip": 1.8048954,
      "balance_loss_mlp": 1.57376802,
      "epoch": 0.0021643720315036373,
      "flos": 63857001582720.0,
      "grad_norm": 3.581057620466552,
      "language_loss": 0.72112948,
      "learning_rate": 2.0939181139872922e-06,
      "loss": 0.7837618,
      "num_input_tokens_seen": 462605,
      "step": 18,
      "time_per_iteration": 3.271836519241333
    },
    {
      "auxiliary_loss_clip": 0.04645232,
      "auxiliary_loss_mlp": 0.01749784,
      "balance_loss_clip": 1.6872946,
      "balance_loss_mlp": 1.51498926,
      "epoch": 0.0022846149221427284,
      "flos": 31284981192960.0,
      "grad_norm": 7.747780912795622,
      "language_loss": 1.01717639,
      "learning_rate": 2.1330868934640175e-06,
      "loss": 1.08112645,
      "num_input_tokens_seen": 483280,
      "step": 19,
      "time_per_iteration": 2.9813907146453857
    },
    {
      "auxiliary_loss_clip": 0.04408503,
      "auxiliary_loss_mlp": 0.01608801,
      "balance_loss_clip": 1.79448843,
      "balance_loss_mlp": 1.46460557,
      "epoch": 0.002404857812781819,
      "flos": 51083648161920.0,
      "grad_norm": 3.603895628493545,
      "language_loss": 0.76377952,
      "learning_rate": 2.170246112844971e-06,
      "loss": 0.8239525,
      "num_input_tokens_seen": 537620,
      "step": 20,
      "time_per_iteration": 3.248009443283081
    },
    {
      "auxiliary_loss_clip": 0.04552693,
      "auxiliary_loss_mlp": 0.01708598,
      "balance_loss_clip": 1.67583048,
      "balance_loss_mlp": 1.46960711,
      "epoch": 0.0025251007034209102,
      "flos": 15815347309440.0,
      "grad_norm": 4.243842615780336,
      "language_loss": 1.01518309,
      "learning_rate": 2.2055919496770983e-06,
      "loss": 1.07779598,
      "num_input_tokens_seen": 555760,
      "step": 21,
      "time_per_iteration": 2.8650996685028076
    },
    {
      "auxiliary_loss_clip": 0.0447053,
      "auxiliary_loss_mlp": 0.01673593,
      "balance_loss_clip": 1.67140961,
      "balance_loss_mlp": 1.4426136,
      "epoch": 0.0026453435940600014,
      "flos": 37851857458560.0,
      "grad_norm": 4.2638229527125135,
      "language_loss": 0.89662325,
      "learning_rate": 2.2392931865974923e-06,
      "loss": 0.95806456,
      "num_input_tokens_seen": 578450,
      "step": 22,
      "time_per_iteration": 3.021313428878784
    },
    {
      "auxiliary_loss_clip": 0.04412545,
      "auxiliary_loss_mlp": 0.0165232,
      "balance_loss_clip": 1.66456735,
      "balance_loss_mlp": 1.42963719,
      "epoch": 0.002765586484699092,
      "flos": 21141976821120.0,
      "grad_norm": 4.550578941567898,
      "language_loss": 1.01824236,
      "learning_rate": 2.271496085962064e-06,
      "loss": 1.07889104,
      "num_input_tokens_seen": 596145,
      "step": 23,
      "time_per_iteration": 2.85420560836792
    },
    {
      "auxiliary_loss_clip": 0.04373398,
      "auxiliary_loss_mlp": 0.01599677,
      "balance_loss_clip": 1.66505671,
      "balance_loss_mlp": 1.39025056,
      "epoch": 0.002885829375338183,
      "flos": 20667381396480.0,
      "grad_norm": 3.46239918107854,
      "language_loss": 1.02760446,
      "learning_rate": 2.3023282262611022e-06,
      "loss": 1.08733523,
      "num_input_tokens_seen": 614920,
      "step": 24,
      "time_per_iteration": 2.779857873916626
    },
    {
      "auxiliary_loss_clip": 0.04295626,
      "auxiliary_loss_mlp": 0.01572185,
      "balance_loss_clip": 1.65652156,
      "balance_loss_mlp": 1.35322118,
      "epoch": 0.003006072265977274,
      "flos": 34823869873920.0,
      "grad_norm": 3.1393108529683764,
      "language_loss": 0.92518288,
      "learning_rate": 2.3319015548620114e-06,
      "loss": 0.98386097,
      "num_input_tokens_seen": 636060,
      "step": 25,
      "time_per_iteration": 3.092501163482666
    },
    {
      "auxiliary_loss_clip": 0.04275288,
      "auxiliary_loss_mlp": 0.01553748,
      "balance_loss_clip": 1.65139651,
      "balance_loss_mlp": 1.3436532,
      "epoch": 0.003126315156616365,
      "flos": 24422021118720.0,
      "grad_norm": 2.3055251277723823,
      "language_loss": 0.93069053,
      "learning_rate": 2.3603148416618152e-06,
      "loss": 0.98898089,
      "num_input_tokens_seen": 655575,
      "step": 26,
      "time_per_iteration": 2.89554500579834
    },
    {
      "auxiliary_loss_clip": 0.04193484,
      "auxiliary_loss_mlp": 0.01498389,
      "balance_loss_clip": 1.64472055,
      "balance_loss_mlp": 1.29716349,
      "epoch": 0.003246558047255456,
      "flos": 23622326674560.0,
      "grad_norm": 2.8145370646028547,
      "language_loss": 1.01037955,
      "learning_rate": 2.3876556694204647e-06,
      "loss": 1.06729817,
      "num_input_tokens_seen": 675730,
      "step": 27,
      "time_per_iteration": 2.841012716293335
    },
    {
      "auxiliary_loss_clip": 0.04137793,
      "auxiliary_loss_mlp": 0.01520952,
      "balance_loss_clip": 1.63721514,
      "balance_loss_mlp": 1.31190646,
      "epoch": 0.003366800937894547,
      "flos": 17820275725440.0,
      "grad_norm": 2.474884664107992,
      "language_loss": 0.90733266,
      "learning_rate": 2.414002061950908e-06,
      "loss": 0.96392012,
      "num_input_tokens_seen": 694605,
      "step": 28,
      "time_per_iteration": 2.7712857723236084
    },
    {
      "auxiliary_loss_clip": 0.04079972,
      "auxiliary_loss_mlp": 0.01496019,
      "balance_loss_clip": 1.63299036,
      "balance_loss_mlp": 1.2940309,
      "epoch": 0.003487043828533638,
      "flos": 24426115269120.0,
      "grad_norm": 2.444227151207845,
      "language_loss": 0.99741185,
      "learning_rate": 2.4394238264681557e-06,
      "loss": 1.05317175,
      "num_input_tokens_seen": 714340,
      "step": 29,
      "time_per_iteration": 2.846684217453003
    },
    {
      "auxiliary_loss_clip": 0.04036798,
      "auxiliary_loss_mlp": 0.01471334,
      "balance_loss_clip": 1.62771642,
      "balance_loss_mlp": 1.26305151,
      "epoch": 0.003607286719172729,
      "flos": 26140311002880.0,
      "grad_norm": 2.11787354553362,
      "language_loss": 0.99490565,
      "learning_rate": 2.4639836682781433e-06,
      "loss": 1.04998684,
      "num_input_tokens_seen": 734470,
      "step": 30,
      "time_per_iteration": 2.7983009815216064
    },
    {
      "auxiliary_loss_clip": 0.0399409,
      "auxiliary_loss_mlp": 0.01499579,
      "balance_loss_clip": 1.63303304,
      "balance_loss_mlp": 1.30388558,
      "epoch": 0.00372752960981182,
      "flos": 20593082113920.0,
      "grad_norm": 2.800189347103699,
      "language_loss": 1.00112796,
      "learning_rate": 2.487738122623307e-06,
      "loss": 1.05606461,
      "num_input_tokens_seen": 753380,
      "step": 31,
      "time_per_iteration": 2.778055429458618
    },
    {
      "auxiliary_loss_clip": 0.038844,
      "auxiliary_loss_mlp": 0.01457766,
      "balance_loss_clip": 1.60530734,
      "balance_loss_mlp": 1.25549197,
      "epoch": 0.003847772500450911,
      "flos": 22674608282880.0,
      "grad_norm": 2.4650540079797754,
      "language_loss": 0.9886058,
      "learning_rate": 2.510738338534912e-06,
      "loss": 1.04202747,
      "num_input_tokens_seen": 772105,
      "step": 32,
      "time_per_iteration": 2.7730486392974854
    },
    {
      "auxiliary_loss_clip": 0.03744791,
      "auxiliary_loss_mlp": 0.014026,
      "balance_loss_clip": 1.58969796,
      "balance_loss_mlp": 1.21176982,
      "epoch": 0.003968015391090002,
      "flos": 17967796882560.0,
      "grad_norm": 6.519815717557976,
      "language_loss": 1.02502704,
      "learning_rate": 2.5330307420306648e-06,
      "loss": 1.07650089,
      "num_input_tokens_seen": 788955,
      "step": 33,
      "time_per_iteration": 2.768930673599243
    },
    {
      "auxiliary_loss_clip": 0.03636754,
      "auxiliary_loss_mlp": 0.01377121,
      "balance_loss_clip": 1.58272302,
      "balance_loss_mlp": 1.1967814,
      "epoch": 0.004088258281729093,
      "flos": 27304103658240.0,
      "grad_norm": 2.2426728418233433,
      "language_loss": 0.88056368,
      "learning_rate": 2.554657600279796e-06,
      "loss": 0.93070245,
      "num_input_tokens_seen": 810230,
      "step": 34,
      "time_per_iteration": 2.829650402069092
    },
    {
      "auxiliary_loss_clip": 0.0354379,
      "auxiliary_loss_mlp": 0.01401837,
      "balance_loss_clip": 1.57036376,
      "balance_loss_mlp": 1.21196055,
      "epoch": 0.004208501172368184,
      "flos": 23258587599360.0,
      "grad_norm": 2.4056574442763994,
      "language_loss": 1.03434634,
      "learning_rate": 2.5756575039679493e-06,
      "loss": 1.0838027,
      "num_input_tokens_seen": 829780,
      "step": 35,
      "time_per_iteration": 2.8085217475891113
    },
    {
      "auxiliary_loss_clip": 0.0347543,
      "auxiliary_loss_mlp": 0.01352325,
      "balance_loss_clip": 1.55909395,
      "balance_loss_mlp": 1.17465544,
      "epoch": 0.0043287440630072746,
      "flos": 17312104062720.0,
      "grad_norm": 2.0657679327712417,
      "language_loss": 0.95179415,
      "learning_rate": 2.5960657816942747e-06,
      "loss": 1.00007176,
      "num_input_tokens_seen": 848695,
      "step": 36,
      "time_per_iteration": 2.796475648880005
    },
    {
      "auxiliary_loss_clip": 0.02712981,
      "auxiliary_loss_mlp": 0.01257265,
      "balance_loss_clip": 1.55957174,
      "balance_loss_mlp": 1.15121651,
      "epoch": 0.004448986953646365,
      "flos": 53092491160320.0,
      "grad_norm": 1.398100459566507,
      "language_loss": 0.6094501,
      "learning_rate": 2.6159148575788668e-06,
      "loss": 0.64915258,
      "num_input_tokens_seen": 906730,
      "step": 37,
      "time_per_iteration": 3.263845920562744
    },
    {
      "auxiliary_loss_clip": 0.03324416,
      "auxiliary_loss_mlp": 0.01297949,
      "balance_loss_clip": 1.53723025,
      "balance_loss_mlp": 1.12972045,
      "epoch": 0.004569229844285457,
      "flos": 13444165866240.0,
      "grad_norm": 2.4402688603515634,
      "language_loss": 0.98738503,
      "learning_rate": 2.635234561171e-06,
      "loss": 1.03360868,
      "num_input_tokens_seen": 925125,
      "step": 38,
      "time_per_iteration": 2.854766607284546
    },
    {
      "auxiliary_loss_clip": 0.03211149,
      "auxiliary_loss_mlp": 0.01277196,
      "balance_loss_clip": 1.52883744,
      "balance_loss_mlp": 1.11106598,
      "epoch": 0.0046894727349245475,
      "flos": 16209609966720.0,
      "grad_norm": 2.555961364748435,
      "language_loss": 0.94091618,
      "learning_rate": 2.6540523970949877e-06,
      "loss": 0.98579961,
      "num_input_tokens_seen": 939970,
      "step": 39,
      "time_per_iteration": 2.82513427734375
    },
    {
      "auxiliary_loss_clip": 0.03167826,
      "auxiliary_loss_mlp": 0.01281906,
      "balance_loss_clip": 1.51552856,
      "balance_loss_mlp": 1.11940002,
      "epoch": 0.004809715625563638,
      "flos": 23914244505600.0,
      "grad_norm": 2.907335336352738,
      "language_loss": 0.92265552,
      "learning_rate": 2.6723937805519533e-06,
      "loss": 0.96715283,
      "num_input_tokens_seen": 957470,
      "step": 40,
      "time_per_iteration": 3.6738321781158447
    },
    {
      "auxiliary_loss_clip": 0.03130566,
      "auxiliary_loss_mlp": 0.01284811,
      "balance_loss_clip": 1.50870502,
      "balance_loss_mlp": 1.1187768,
      "epoch": 0.00492995851620273,
      "flos": 20773030273920.0,
      "grad_norm": 2.2055602961423872,
      "language_loss": 0.92955822,
      "learning_rate": 2.690282243737839e-06,
      "loss": 0.97371209,
      "num_input_tokens_seen": 976405,
      "step": 41,
      "time_per_iteration": 4.646177291870117
    },
    {
      "auxiliary_loss_clip": 0.03065901,
      "auxiliary_loss_mlp": 0.01274792,
      "balance_loss_clip": 1.50095725,
      "balance_loss_mlp": 1.11590958,
      "epoch": 0.0050502014068418205,
      "flos": 20338655103360.0,
      "grad_norm": 6.513918857093967,
      "language_loss": 0.99418163,
      "learning_rate": 2.7077396173840807e-06,
      "loss": 1.0375886,
      "num_input_tokens_seen": 994690,
      "step": 42,
      "time_per_iteration": 2.740849733352661
    },
    {
      "auxiliary_loss_clip": 0.03000081,
      "auxiliary_loss_mlp": 0.01270013,
      "balance_loss_clip": 1.48466992,
      "balance_loss_mlp": 1.10636294,
      "epoch": 0.005170444297480911,
      "flos": 25994872834560.0,
      "grad_norm": 3.3170561444957407,
      "language_loss": 0.92845654,
      "learning_rate": 2.7247861909342594e-06,
      "loss": 0.97115755,
      "num_input_tokens_seen": 1015615,
      "step": 43,
      "time_per_iteration": 2.7881457805633545
    },
    {
      "auxiliary_loss_clip": 0.02905116,
      "auxiliary_loss_mlp": 0.01230875,
      "balance_loss_clip": 1.47301531,
      "balance_loss_mlp": 1.07390022,
      "epoch": 0.005290687188120003,
      "flos": 20954055841920.0,
      "grad_norm": 2.478486970761795,
      "language_loss": 0.83110189,
      "learning_rate": 2.7414408543044743e-06,
      "loss": 0.8724618,
      "num_input_tokens_seen": 1031255,
      "step": 44,
      "time_per_iteration": 2.8284928798675537
    },
    {
      "auxiliary_loss_clip": 0.02895954,
      "auxiliary_loss_mlp": 0.01240639,
      "balance_loss_clip": 1.47105277,
      "balance_loss_mlp": 1.08147097,
      "epoch": 0.005410930078759093,
      "flos": 15851401585920.0,
      "grad_norm": 4.938522039258118,
      "language_loss": 0.79269403,
      "learning_rate": 2.7577212237113157e-06,
      "loss": 0.83405995,
      "num_input_tokens_seen": 1048295,
      "step": 45,
      "time_per_iteration": 2.7614693641662598
    },
    {
      "auxiliary_loss_clip": 0.02804271,
      "auxiliary_loss_mlp": 0.01229875,
      "balance_loss_clip": 1.45466638,
      "balance_loss_mlp": 1.08415341,
      "epoch": 0.005531172969398184,
      "flos": 21104988791040.0,
      "grad_norm": 3.128715021362949,
      "language_loss": 1.0423187,
      "learning_rate": 2.7736437536690466e-06,
      "loss": 1.08266008,
      "num_input_tokens_seen": 1067925,
      "step": 46,
      "time_per_iteration": 2.797236919403076
    },
    {
      "auxiliary_loss_clip": 0.02786208,
      "auxiliary_loss_mlp": 0.01229044,
      "balance_loss_clip": 1.45143819,
      "balance_loss_mlp": 1.0791266,
      "epoch": 0.005651415860037276,
      "flos": 20844887431680.0,
      "grad_norm": 2.16766247465609,
      "language_loss": 1.07930207,
      "learning_rate": 2.789223836941131e-06,
      "loss": 1.11945462,
      "num_input_tokens_seen": 1088060,
      "step": 47,
      "time_per_iteration": 2.754136800765991
    },
    {
      "auxiliary_loss_clip": 0.02711129,
      "auxiliary_loss_mlp": 0.01214267,
      "balance_loss_clip": 1.43714738,
      "balance_loss_mlp": 1.06978571,
      "epoch": 0.005771658750676366,
      "flos": 13260195383040.0,
      "grad_norm": 2.4704171073262673,
      "language_loss": 1.08733082,
      "learning_rate": 2.8044758939680847e-06,
      "loss": 1.12658477,
      "num_input_tokens_seen": 1104130,
      "step": 48,
      "time_per_iteration": 2.7690443992614746
    },
    {
      "auxiliary_loss_clip": 0.02664634,
      "auxiliary_loss_mlp": 0.01216606,
      "balance_loss_clip": 1.42337966,
      "balance_loss_mlp": 1.07756066,
      "epoch": 0.005891901641315457,
      "flos": 24425396997120.0,
      "grad_norm": 2.590883595183304,
      "language_loss": 1.02152681,
      "learning_rate": 2.8194134530738863e-06,
      "loss": 1.06033921,
      "num_input_tokens_seen": 1122900,
      "step": 49,
      "time_per_iteration": 2.8451554775238037
    },
    {
      "auxiliary_loss_clip": 0.02602167,
      "auxiliary_loss_mlp": 0.01195376,
      "balance_loss_clip": 1.41768551,
      "balance_loss_mlp": 1.06415009,
      "epoch": 0.006012144531954548,
      "flos": 23076197314560.0,
      "grad_norm": 2.800154995030968,
      "language_loss": 0.90446198,
      "learning_rate": 2.834049222568994e-06,
      "loss": 0.94243735,
      "num_input_tokens_seen": 1140250,
      "step": 50,
      "time_per_iteration": 2.7629048824310303
    },
    {
      "auxiliary_loss_clip": 0.0258097,
      "auxiliary_loss_mlp": 0.0121875,
      "balance_loss_clip": 1.4133811,
      "balance_loss_mlp": 1.07951343,
      "epoch": 0.006132387422593639,
      "flos": 22528775064960.0,
      "grad_norm": 2.136177130928648,
      "language_loss": 0.92719954,
      "learning_rate": 2.848395155712969e-06,
      "loss": 0.96519673,
      "num_input_tokens_seen": 1160470,
      "step": 51,
      "time_per_iteration": 2.8485653400421143
    },
    {
      "auxiliary_loss_clip": 0.02514725,
      "auxiliary_loss_mlp": 0.01211948,
      "balance_loss_clip": 1.39634442,
      "balance_loss_mlp": 1.08072293,
      "epoch": 0.00625263031323273,
      "flos": 27628340751360.0,
      "grad_norm": 2.563716381106121,
      "language_loss": 0.97788155,
      "learning_rate": 2.8624625093687977e-06,
      "loss": 1.0151484,
      "num_input_tokens_seen": 1177605,
      "step": 52,
      "time_per_iteration": 2.901848316192627
    },
    {
      "auxiliary_loss_clip": 0.02505774,
      "auxiliary_loss_mlp": 0.01206686,
      "balance_loss_clip": 1.39664388,
      "balance_loss_mlp": 1.07326722,
      "epoch": 0.006372873203871821,
      "flos": 23110671392640.0,
      "grad_norm": 2.264819244116928,
      "language_loss": 0.88972235,
      "learning_rate": 2.876261897070029e-06,
      "loss": 0.92684692,
      "num_input_tokens_seen": 1197735,
      "step": 53,
      "time_per_iteration": 2.8300676345825195
    },
    {
      "auxiliary_loss_clip": 0.02494719,
      "auxiliary_loss_mlp": 0.0118941,
      "balance_loss_clip": 1.38786495,
      "balance_loss_mlp": 1.05541921,
      "epoch": 0.006493116094510912,
      "flos": 22856028900480.0,
      "grad_norm": 2.1984807228297565,
      "language_loss": 0.92473441,
      "learning_rate": 2.889803337127447e-06,
      "loss": 0.96157563,
      "num_input_tokens_seen": 1216335,
      "step": 54,
      "time_per_iteration": 2.8134469985961914
    },
    {
      "auxiliary_loss_clip": 0.02443882,
      "auxiliary_loss_mlp": 0.01211901,
      "balance_loss_clip": 1.37708569,
      "balance_loss_mlp": 1.08153355,
      "epoch": 0.006613358985150003,
      "flos": 23071708114560.0,
      "grad_norm": 2.5152332861768985,
      "language_loss": 0.84783959,
      "learning_rate": 2.903096296321516e-06,
      "loss": 0.88439733,
      "num_input_tokens_seen": 1234480,
      "step": 55,
      "time_per_iteration": 2.738147497177124
    },
    {
      "auxiliary_loss_clip": 0.02391334,
      "auxiliary_loss_mlp": 0.01183289,
      "balance_loss_clip": 1.36923575,
      "balance_loss_mlp": 1.06017017,
      "epoch": 0.006733601875789094,
      "flos": 26537662229760.0,
      "grad_norm": 2.2973820784135297,
      "language_loss": 0.91556478,
      "learning_rate": 2.9161497296578907e-06,
      "loss": 0.95131099,
      "num_input_tokens_seen": 1253870,
      "step": 56,
      "time_per_iteration": 2.787972927093506
    },
    {
      "auxiliary_loss_clip": 0.02348511,
      "auxiliary_loss_mlp": 0.01183495,
      "balance_loss_clip": 1.35847557,
      "balance_loss_mlp": 1.06433296,
      "epoch": 0.006853844766428185,
      "flos": 15523178083200.0,
      "grad_norm": 2.3129510244085485,
      "language_loss": 0.86039537,
      "learning_rate": 2.928972116604173e-06,
      "loss": 0.89571548,
      "num_input_tokens_seen": 1270145,
      "step": 57,
      "time_per_iteration": 2.7782280445098877
    },
    {
      "auxiliary_loss_clip": 0.02347011,
      "auxiliary_loss_mlp": 0.01200212,
      "balance_loss_clip": 1.3570261,
      "balance_loss_mlp": 1.08271909,
      "epoch": 0.006974087657067276,
      "flos": 24243760897920.0,
      "grad_norm": 2.5923358001098524,
      "language_loss": 1.01897335,
      "learning_rate": 2.9415714941751377e-06,
      "loss": 1.05444551,
      "num_input_tokens_seen": 1291365,
      "step": 58,
      "time_per_iteration": 2.834991693496704
    },
    {
      "auxiliary_loss_clip": 0.02319447,
      "auxiliary_loss_mlp": 0.01176906,
      "balance_loss_clip": 1.35600996,
      "balance_loss_mlp": 1.06217885,
      "epoch": 0.007094330547706367,
      "flos": 25772513690880.0,
      "grad_norm": 3.467287956465026,
      "language_loss": 0.93718648,
      "learning_rate": 2.9539554871897396e-06,
      "loss": 0.97215003,
      "num_input_tokens_seen": 1311535,
      "step": 59,
      "time_per_iteration": 2.80979061126709
    },
    {
      "auxiliary_loss_clip": 0.02264673,
      "auxiliary_loss_mlp": 0.01156659,
      "balance_loss_clip": 1.3469466,
      "balance_loss_mlp": 1.04646206,
      "epoch": 0.007214573438345458,
      "flos": 21319015979520.0,
      "grad_norm": 2.062195976080685,
      "language_loss": 0.97550642,
      "learning_rate": 2.9661313359851253e-06,
      "loss": 1.00971973,
      "num_input_tokens_seen": 1329420,
      "step": 60,
      "time_per_iteration": 2.741609811782837
    },
    {
      "auxiliary_loss_clip": 0.02223253,
      "auxiliary_loss_mlp": 0.0118003,
      "balance_loss_clip": 1.33327842,
      "balance_loss_mlp": 1.06806922,
      "epoch": 0.007334816328984549,
      "flos": 24937088192640.0,
      "grad_norm": 2.0168639248592224,
      "language_loss": 0.93953395,
      "learning_rate": 2.978105921839922e-06,
      "loss": 0.97356683,
      "num_input_tokens_seen": 1349965,
      "step": 61,
      "time_per_iteration": 2.768526554107666
    },
    {
      "auxiliary_loss_clip": 0.02187746,
      "auxiliary_loss_mlp": 0.01163528,
      "balance_loss_clip": 1.32032764,
      "balance_loss_mlp": 1.05395103,
      "epoch": 0.00745505921962364,
      "flos": 18510586277760.0,
      "grad_norm": 2.1936242186930657,
      "language_loss": 0.72112602,
      "learning_rate": 2.9898857903302893e-06,
      "loss": 0.75463873,
      "num_input_tokens_seen": 1368915,
      "step": 62,
      "time_per_iteration": 2.6751015186309814
    },
    {
      "auxiliary_loss_clip": 0.02167141,
      "auxiliary_loss_mlp": 0.01178666,
      "balance_loss_clip": 1.31535447,
      "balance_loss_mlp": 1.07018542,
      "epoch": 0.007575302110262731,
      "flos": 18477656484480.0,
      "grad_norm": 2.535867711403972,
      "language_loss": 0.87774241,
      "learning_rate": 3.001477172817253e-06,
      "loss": 0.91120052,
      "num_input_tokens_seen": 1386805,
      "step": 63,
      "time_per_iteration": 2.6111745834350586
    },
    {
      "auxiliary_loss_clip": 0.02140367,
      "auxiliary_loss_mlp": 0.01154466,
      "balance_loss_clip": 1.31112742,
      "balance_loss_mlp": 1.05118251,
      "epoch": 0.007695545000901822,
      "flos": 24973178382720.0,
      "grad_norm": 3.4897943588201397,
      "language_loss": 0.96148896,
      "learning_rate": 3.012886006241894e-06,
      "loss": 0.99443734,
      "num_input_tokens_seen": 1406190,
      "step": 64,
      "time_per_iteration": 2.7253668308258057
    },
    {
      "auxiliary_loss_clip": 0.02120593,
      "auxiliary_loss_mlp": 0.01169841,
      "balance_loss_clip": 1.31187201,
      "balance_loss_mlp": 1.07094491,
      "epoch": 0.007815787891540913,
      "flos": 21324223451520.0,
      "grad_norm": 2.0888144716895463,
      "language_loss": 0.88160777,
      "learning_rate": 3.0241179513858383e-06,
      "loss": 0.91451216,
      "num_input_tokens_seen": 1425500,
      "step": 65,
      "time_per_iteration": 2.9570558071136475
    },
    {
      "auxiliary_loss_clip": 0.02104354,
      "auxiliary_loss_mlp": 0.01152048,
      "balance_loss_clip": 1.30326366,
      "balance_loss_mlp": 1.05148268,
      "epoch": 0.007936030782180003,
      "flos": 21575777374080.0,
      "grad_norm": 2.329740520358243,
      "language_loss": 0.8764829,
      "learning_rate": 3.035178409737647e-06,
      "loss": 0.90904689,
      "num_input_tokens_seen": 1442950,
      "step": 66,
      "time_per_iteration": 3.770076036453247
    },
    {
      "auxiliary_loss_clip": 0.02030131,
      "auxiliary_loss_mlp": 0.01145553,
      "balance_loss_clip": 1.29287148,
      "balance_loss_mlp": 1.053285,
      "epoch": 0.008056273672819095,
      "flos": 20120785159680.0,
      "grad_norm": 2.6827222443103795,
      "language_loss": 0.88649905,
      "learning_rate": 3.046072539090907e-06,
      "loss": 0.91825581,
      "num_input_tokens_seen": 1460915,
      "step": 67,
      "time_per_iteration": 3.8680179119110107
    },
    {
      "auxiliary_loss_clip": 0.02024112,
      "auxiliary_loss_mlp": 0.01158142,
      "balance_loss_clip": 1.28650498,
      "balance_loss_mlp": 1.06959343,
      "epoch": 0.008176516563458186,
      "flos": 18333116156160.0,
      "grad_norm": 2.650473262554964,
      "language_loss": 1.04769254,
      "learning_rate": 3.056805267986779e-06,
      "loss": 1.0795151,
      "num_input_tokens_seen": 1478385,
      "step": 68,
      "time_per_iteration": 3.649784564971924
    },
    {
      "auxiliary_loss_clip": 0.01987717,
      "auxiliary_loss_mlp": 0.01150064,
      "balance_loss_clip": 1.27811837,
      "balance_loss_mlp": 1.06227815,
      "epoch": 0.008296759454097276,
      "flos": 21872076664320.0,
      "grad_norm": 2.235463942775072,
      "language_loss": 0.95375091,
      "learning_rate": 3.0673813091022194e-06,
      "loss": 0.9851287,
      "num_input_tokens_seen": 1497605,
      "step": 69,
      "time_per_iteration": 2.780953884124756
    },
    {
      "auxiliary_loss_clip": 0.01643509,
      "auxiliary_loss_mlp": 0.01243579,
      "balance_loss_clip": 1.27731204,
      "balance_loss_mlp": 1.1999011,
      "epoch": 0.008417002344736368,
      "flos": 63408228036480.0,
      "grad_norm": 1.341759480450728,
      "language_loss": 0.62091112,
      "learning_rate": 3.0778051716749317e-06,
      "loss": 0.649782,
      "num_input_tokens_seen": 1561150,
      "step": 70,
      "time_per_iteration": 3.3333160877227783
    },
    {
      "auxiliary_loss_clip": 0.01965312,
      "auxiliary_loss_mlp": 0.01156451,
      "balance_loss_clip": 1.27226973,
      "balance_loss_mlp": 1.07052517,
      "epoch": 0.008537245235375458,
      "flos": 22966454286720.0,
      "grad_norm": 2.5309095659934844,
      "language_loss": 0.90660167,
      "learning_rate": 3.0880811730470094e-06,
      "loss": 0.9378193,
      "num_input_tokens_seen": 1580605,
      "step": 71,
      "time_per_iteration": 2.895136594772339
    },
    {
      "auxiliary_loss_clip": 0.0161845,
      "auxiliary_loss_mlp": 0.01206534,
      "balance_loss_clip": 1.26785111,
      "balance_loss_mlp": 1.16857743,
      "epoch": 0.008657488126014549,
      "flos": 61984046712960.0,
      "grad_norm": 1.1895129622032572,
      "language_loss": 0.58607066,
      "learning_rate": 3.098213449401257e-06,
      "loss": 0.61432052,
      "num_input_tokens_seen": 1647535,
      "step": 72,
      "time_per_iteration": 3.239751100540161
    },
    {
      "auxiliary_loss_clip": 0.01934355,
      "auxiliary_loss_mlp": 0.01130783,
      "balance_loss_clip": 1.26750875,
      "balance_loss_mlp": 1.05491781,
      "epoch": 0.00877773101665364,
      "flos": 30296791152000.0,
      "grad_norm": 2.232256158656011,
      "language_loss": 0.9878791,
      "learning_rate": 3.1082059657570015e-06,
      "loss": 1.01853037,
      "num_input_tokens_seen": 1666770,
      "step": 73,
      "time_per_iteration": 2.851335287094116
    },
    {
      "auxiliary_loss_clip": 0.01924768,
      "auxiliary_loss_mlp": 0.01158849,
      "balance_loss_clip": 1.26684916,
      "balance_loss_mlp": 1.08179212,
      "epoch": 0.00889797390729273,
      "flos": 23514056104320.0,
      "grad_norm": 2.048504211263131,
      "language_loss": 0.96835613,
      "learning_rate": 3.1180625252858496e-06,
      "loss": 0.9991923,
      "num_input_tokens_seen": 1685200,
      "step": 74,
      "time_per_iteration": 2.7568843364715576
    },
    {
      "auxiliary_loss_clip": 0.01885141,
      "auxiliary_loss_mlp": 0.01138854,
      "balance_loss_clip": 1.24832022,
      "balance_loss_mlp": 1.0633707,
      "epoch": 0.009018216797931822,
      "flos": 23075838178560.0,
      "grad_norm": 2.577971288282226,
      "language_loss": 0.80112499,
      "learning_rate": 3.1277867780021663e-06,
      "loss": 0.83136499,
      "num_input_tokens_seen": 1701835,
      "step": 75,
      "time_per_iteration": 2.7856545448303223
    },
    {
      "auxiliary_loss_clip": 0.01844935,
      "auxiliary_loss_mlp": 0.01136946,
      "balance_loss_clip": 1.24290299,
      "balance_loss_mlp": 1.06279755,
      "epoch": 0.009138459688570914,
      "flos": 15918877284480.0,
      "grad_norm": 2.1484092393885397,
      "language_loss": 0.95869827,
      "learning_rate": 3.1373822288779824e-06,
      "loss": 0.98851705,
      "num_input_tokens_seen": 1718415,
      "step": 76,
      "time_per_iteration": 2.7565224170684814
    },
    {
      "auxiliary_loss_clip": 0.01845295,
      "auxiliary_loss_mlp": 0.01150671,
      "balance_loss_clip": 1.24632525,
      "balance_loss_mlp": 1.07757187,
      "epoch": 0.009258702579210003,
      "flos": 27016531372800.0,
      "grad_norm": 1.8809145414819437,
      "language_loss": 0.79759175,
      "learning_rate": 3.1468522454274533e-06,
      "loss": 0.82755136,
      "num_input_tokens_seen": 1738770,
      "step": 77,
      "time_per_iteration": 2.8702969551086426
    },
    {
      "auxiliary_loss_clip": 0.01832947,
      "auxiliary_loss_mlp": 0.01130137,
      "balance_loss_clip": 1.23669004,
      "balance_loss_mlp": 1.05474889,
      "epoch": 0.009378945469849095,
      "flos": 26903196984960.0,
      "grad_norm": 2.56738546914615,
      "language_loss": 0.9164747,
      "learning_rate": 3.15620006480197e-06,
      "loss": 0.94610554,
      "num_input_tokens_seen": 1758040,
      "step": 78,
      "time_per_iteration": 2.8652310371398926
    },
    {
      "auxiliary_loss_clip": 0.01824199,
      "auxiliary_loss_mlp": 0.01126838,
      "balance_loss_clip": 1.23545003,
      "balance_loss_mlp": 1.05855536,
      "epoch": 0.009499188360488187,
      "flos": 35694236327040.0,
      "grad_norm": 2.591374617207664,
      "language_loss": 0.74869633,
      "learning_rate": 3.1654288004333087e-06,
      "loss": 0.77820671,
      "num_input_tokens_seen": 1776705,
      "step": 79,
      "time_per_iteration": 2.8695356845855713
    },
    {
      "auxiliary_loss_clip": 0.01785151,
      "auxiliary_loss_mlp": 0.01112771,
      "balance_loss_clip": 1.22042942,
      "balance_loss_mlp": 1.05125928,
      "epoch": 0.009619431251127276,
      "flos": 21503201944320.0,
      "grad_norm": 2.5951227508684873,
      "language_loss": 0.76107407,
      "learning_rate": 3.1745414482589353e-06,
      "loss": 0.79005325,
      "num_input_tokens_seen": 1795915,
      "step": 80,
      "time_per_iteration": 2.8591063022613525
    },
    {
      "auxiliary_loss_clip": 0.01773899,
      "auxiliary_loss_mlp": 0.01121752,
      "balance_loss_clip": 1.2230072,
      "balance_loss_mlp": 1.05819011,
      "epoch": 0.009739674141766368,
      "flos": 17421056991360.0,
      "grad_norm": 2.4447506894329445,
      "language_loss": 0.8707304,
      "learning_rate": 3.1835408925606204e-06,
      "loss": 0.89968681,
      "num_input_tokens_seen": 1814055,
      "step": 81,
      "time_per_iteration": 2.7691657543182373
    },
    {
      "auxiliary_loss_clip": 0.01755245,
      "auxiliary_loss_mlp": 0.01116468,
      "balance_loss_clip": 1.21605325,
      "balance_loss_mlp": 1.05376351,
      "epoch": 0.00985991703240546,
      "flos": 27527109246720.0,
      "grad_norm": 2.1900977133818444,
      "language_loss": 0.89185619,
      "learning_rate": 3.1924299114448214e-06,
      "loss": 0.92057335,
      "num_input_tokens_seen": 1834535,
      "step": 82,
      "time_per_iteration": 2.809569835662842
    },
    {
      "auxiliary_loss_clip": 0.01745594,
      "auxiliary_loss_mlp": 0.01131194,
      "balance_loss_clip": 1.21575546,
      "balance_loss_mlp": 1.06691599,
      "epoch": 0.00998015992304455,
      "flos": 13808084509440.0,
      "grad_norm": 2.0291243713255143,
      "language_loss": 0.83632255,
      "learning_rate": 3.2012111819909055e-06,
      "loss": 0.86509043,
      "num_input_tokens_seen": 1851865,
      "step": 83,
      "time_per_iteration": 2.782761335372925
    },
    {
      "auxiliary_loss_clip": 0.01730438,
      "auxiliary_loss_mlp": 0.01122958,
      "balance_loss_clip": 1.21008015,
      "balance_loss_mlp": 1.06151772,
      "epoch": 0.010100402813683641,
      "flos": 20191385341440.0,
      "grad_norm": 2.3287051743278853,
      "language_loss": 0.94855499,
      "learning_rate": 3.2098872850910627e-06,
      "loss": 0.97708893,
      "num_input_tokens_seen": 1868540,
      "step": 84,
      "time_per_iteration": 2.8347549438476562
    },
    {
      "auxiliary_loss_clip": 0.01717911,
      "auxiliary_loss_mlp": 0.0111999,
      "balance_loss_clip": 1.21120405,
      "balance_loss_mlp": 1.060004,
      "epoch": 0.010220645704322733,
      "flos": 17201642762880.0,
      "grad_norm": 2.195611358842131,
      "language_loss": 0.89236045,
      "learning_rate": 3.2184607100038194e-06,
      "loss": 0.92073953,
      "num_input_tokens_seen": 1887180,
      "step": 85,
      "time_per_iteration": 2.953371047973633
    },
    {
      "auxiliary_loss_clip": 0.01710079,
      "auxiliary_loss_mlp": 0.01107785,
      "balance_loss_clip": 1.20574975,
      "balance_loss_mlp": 1.04975426,
      "epoch": 0.010340888594961822,
      "flos": 21470415805440.0,
      "grad_norm": 2.2803132804969506,
      "language_loss": 0.93304276,
      "learning_rate": 3.2269338586412414e-06,
      "loss": 0.96122146,
      "num_input_tokens_seen": 1904765,
      "step": 86,
      "time_per_iteration": 2.92881178855896
    },
    {
      "auxiliary_loss_clip": 0.0170322,
      "auxiliary_loss_mlp": 0.01116354,
      "balance_loss_clip": 1.20604515,
      "balance_loss_mlp": 1.05796552,
      "epoch": 0.010461131485600914,
      "flos": 23002831785600.0,
      "grad_norm": 2.1319287769066677,
      "language_loss": 0.96499979,
      "learning_rate": 3.2353090496083106e-06,
      "loss": 0.99319553,
      "num_input_tokens_seen": 1922600,
      "step": 87,
      "time_per_iteration": 2.9226536750793457
    },
    {
      "auxiliary_loss_clip": 0.01660797,
      "auxiliary_loss_mlp": 0.0111233,
      "balance_loss_clip": 1.1932497,
      "balance_loss_mlp": 1.05715978,
      "epoch": 0.010581374376240005,
      "flos": 33546850571520.0,
      "grad_norm": 2.028649402364408,
      "language_loss": 0.81098443,
      "learning_rate": 3.2435885220114572e-06,
      "loss": 0.83871567,
      "num_input_tokens_seen": 1943950,
      "step": 88,
      "time_per_iteration": 2.8352644443511963
    },
    {
      "auxiliary_loss_clip": 0.01680936,
      "auxiliary_loss_mlp": 0.01117475,
      "balance_loss_clip": 1.20265746,
      "balance_loss_mlp": 1.06616735,
      "epoch": 0.010701617266879095,
      "flos": 21763087822080.0,
      "grad_norm": 6.957320169317885,
      "language_loss": 0.93812174,
      "learning_rate": 3.2517744390519113e-06,
      "loss": 0.96610582,
      "num_input_tokens_seen": 1962815,
      "step": 89,
      "time_per_iteration": 2.7989699840545654
    },
    {
      "auxiliary_loss_clip": 0.01657808,
      "auxiliary_loss_mlp": 0.01114101,
      "balance_loss_clip": 1.19245422,
      "balance_loss_mlp": 1.06386626,
      "epoch": 0.010821860157518187,
      "flos": 19060199256960.0,
      "grad_norm": 1.9824325174978124,
      "language_loss": 0.74970841,
      "learning_rate": 3.259868891418298e-06,
      "loss": 0.77742743,
      "num_input_tokens_seen": 1980580,
      "step": 90,
      "time_per_iteration": 2.799651861190796
    },
    {
      "auxiliary_loss_clip": 0.01663052,
      "auxiliary_loss_mlp": 0.01112002,
      "balance_loss_clip": 1.19538391,
      "balance_loss_mlp": 1.05823851,
      "epoch": 0.010942103048157278,
      "flos": 25447378757760.0,
      "grad_norm": 2.3007028281836526,
      "language_loss": 0.84954351,
      "learning_rate": 3.2678739004917757e-06,
      "loss": 0.87729406,
      "num_input_tokens_seen": 2000315,
      "step": 91,
      "time_per_iteration": 2.832686185836792
    },
    {
      "auxiliary_loss_clip": 0.01644726,
      "auxiliary_loss_mlp": 0.01108083,
      "balance_loss_clip": 1.19269633,
      "balance_loss_mlp": 1.05913532,
      "epoch": 0.011062345938796368,
      "flos": 27493928058240.0,
      "grad_norm": 1.6713736694172445,
      "language_loss": 0.92110264,
      "learning_rate": 3.275791421376029e-06,
      "loss": 0.94863069,
      "num_input_tokens_seen": 2023760,
      "step": 92,
      "time_per_iteration": 2.8843400478363037
    },
    {
      "auxiliary_loss_clip": 0.01630874,
      "auxiliary_loss_mlp": 0.0109708,
      "balance_loss_clip": 1.18641901,
      "balance_loss_mlp": 1.05049348,
      "epoch": 0.01118258882943546,
      "flos": 16071210864000.0,
      "grad_norm": 2.078020735138386,
      "language_loss": 0.96015704,
      "learning_rate": 3.2836233457634622e-06,
      "loss": 0.98743665,
      "num_input_tokens_seen": 2041895,
      "step": 93,
      "time_per_iteration": 4.8046112060546875
    },
    {
      "auxiliary_loss_clip": 0.01636639,
      "auxiliary_loss_mlp": 0.01098241,
      "balance_loss_clip": 1.18995714,
      "balance_loss_mlp": 1.0528934,
      "epoch": 0.011302831720074551,
      "flos": 20668602458880.0,
      "grad_norm": 1.9776767067830598,
      "language_loss": 0.85658115,
      "learning_rate": 3.2913715046481135e-06,
      "loss": 0.88392997,
      "num_input_tokens_seen": 2061640,
      "step": 94,
      "time_per_iteration": 5.202788352966309
    },
    {
      "auxiliary_loss_clip": 0.01624115,
      "auxiliary_loss_mlp": 0.01096097,
      "balance_loss_clip": 1.18583465,
      "balance_loss_mlp": 1.05249071,
      "epoch": 0.011423074610713641,
      "flos": 13072238490240.0,
      "grad_norm": 1.9784845195087397,
      "language_loss": 0.88755369,
      "learning_rate": 3.299037670895023e-06,
      "loss": 0.91475582,
      "num_input_tokens_seen": 2078255,
      "step": 95,
      "time_per_iteration": 2.762392520904541
    },
    {
      "auxiliary_loss_clip": 0.016202,
      "auxiliary_loss_mlp": 0.01099295,
      "balance_loss_clip": 1.18548727,
      "balance_loss_mlp": 1.05468702,
      "epoch": 0.011543317501352733,
      "flos": 30335646689280.0,
      "grad_norm": 1.731668438558827,
      "language_loss": 0.80089182,
      "learning_rate": 3.3066235616750667e-06,
      "loss": 0.82808673,
      "num_input_tokens_seen": 2099490,
      "step": 96,
      "time_per_iteration": 2.8771965503692627
    },
    {
      "auxiliary_loss_clip": 0.01594903,
      "auxiliary_loss_mlp": 0.01090875,
      "balance_loss_clip": 1.17956555,
      "balance_loss_mlp": 1.04950953,
      "epoch": 0.011663560391991824,
      "flos": 15522962601600.0,
      "grad_norm": 2.5235982423524437,
      "language_loss": 0.9245773,
      "learning_rate": 3.3141308407736276e-06,
      "loss": 0.95143503,
      "num_input_tokens_seen": 2116125,
      "step": 97,
      "time_per_iteration": 2.764641761779785
    },
    {
      "auxiliary_loss_clip": 0.01602236,
      "auxiliary_loss_mlp": 0.01079587,
      "balance_loss_clip": 1.17949677,
      "balance_loss_mlp": 1.04053414,
      "epoch": 0.011783803282630914,
      "flos": 19902125116800.0,
      "grad_norm": 2.063281036058668,
      "language_loss": 0.86665791,
      "learning_rate": 3.321561120780869e-06,
      "loss": 0.89347613,
      "num_input_tokens_seen": 2134835,
      "step": 98,
      "time_per_iteration": 2.846137523651123
    },
    {
      "auxiliary_loss_clip": 0.01599344,
      "auxiliary_loss_mlp": 0.01092327,
      "balance_loss_clip": 1.18627429,
      "balance_loss_mlp": 1.05084217,
      "epoch": 0.011904046173270006,
      "flos": 22340674517760.0,
      "grad_norm": 3.0092949355784437,
      "language_loss": 1.0157758,
      "learning_rate": 3.3289159651708192e-06,
      "loss": 1.04269242,
      "num_input_tokens_seen": 2152410,
      "step": 99,
      "time_per_iteration": 2.7981300354003906
    },
    {
      "auxiliary_loss_clip": 0.01572559,
      "auxiliary_loss_mlp": 0.01100576,
      "balance_loss_clip": 1.17492056,
      "balance_loss_mlp": 1.05940151,
      "epoch": 0.012024289063909096,
      "flos": 19100060375040.0,
      "grad_norm": 2.0621754410221826,
      "language_loss": 0.97771817,
      "learning_rate": 3.3361968902759768e-06,
      "loss": 1.00444949,
      "num_input_tokens_seen": 2172090,
      "step": 100,
      "time_per_iteration": 2.7650792598724365
    },
    {
      "auxiliary_loss_clip": 0.0155524,
      "auxiliary_loss_mlp": 0.01088165,
      "balance_loss_clip": 1.17003822,
      "balance_loss_mlp": 1.04935026,
      "epoch": 0.012144531954548187,
      "flos": 15012205159680.0,
      "grad_norm": 2.343354849719683,
      "language_loss": 0.93584847,
      "learning_rate": 3.343405367163663e-06,
      "loss": 0.96228266,
      "num_input_tokens_seen": 2189020,
      "step": 101,
      "time_per_iteration": 2.83579421043396
    },
    {
      "auxiliary_loss_clip": 0.0157256,
      "auxiliary_loss_mlp": 0.01098133,
      "balance_loss_clip": 1.1787045,
      "balance_loss_mlp": 1.05946159,
      "epoch": 0.012264774845187279,
      "flos": 15122020014720.0,
      "grad_norm": 9.116839843299696,
      "language_loss": 0.81270492,
      "learning_rate": 3.350542823419951e-06,
      "loss": 0.83941185,
      "num_input_tokens_seen": 2205620,
      "step": 102,
      "time_per_iteration": 2.7993578910827637
    },
    {
      "auxiliary_loss_clip": 0.01563564,
      "auxiliary_loss_mlp": 0.01100507,
      "balance_loss_clip": 1.17297792,
      "balance_loss_mlp": 1.06164503,
      "epoch": 0.012385017735826368,
      "flos": 13949248959360.0,
      "grad_norm": 4.582694231576832,
      "language_loss": 0.8787685,
      "learning_rate": 3.3576106448465615e-06,
      "loss": 0.90540928,
      "num_input_tokens_seen": 2219000,
      "step": 103,
      "time_per_iteration": 2.786637783050537
    },
    {
      "auxiliary_loss_clip": 0.01556815,
      "auxiliary_loss_mlp": 0.01089607,
      "balance_loss_clip": 1.16934812,
      "balance_loss_mlp": 1.05279493,
      "epoch": 0.01250526062646546,
      "flos": 23623260428160.0,
      "grad_norm": 2.0413618807897094,
      "language_loss": 0.8811847,
      "learning_rate": 3.3646101770757797e-06,
      "loss": 0.90764892,
      "num_input_tokens_seen": 2237790,
      "step": 104,
      "time_per_iteration": 2.8659443855285645
    },
    {
      "auxiliary_loss_clip": 0.01548834,
      "auxiliary_loss_mlp": 0.01084909,
      "balance_loss_clip": 1.16774249,
      "balance_loss_mlp": 1.05091083,
      "epoch": 0.012625503517104552,
      "flos": 34640078958720.0,
      "grad_norm": 1.7027478982017055,
      "language_loss": 0.85480714,
      "learning_rate": 3.371542727108104e-06,
      "loss": 0.88114464,
      "num_input_tokens_seen": 2259965,
      "step": 105,
      "time_per_iteration": 2.985774040222168
    },
    {
      "auxiliary_loss_clip": 0.01548797,
      "auxiliary_loss_mlp": 0.01108822,
      "balance_loss_clip": 1.16451538,
      "balance_loss_mlp": 1.06960237,
      "epoch": 0.012745746407743641,
      "flos": 17821891837440.0,
      "grad_norm": 2.516699186104519,
      "language_loss": 0.89945424,
      "learning_rate": 3.3784095647770114e-06,
      "loss": 0.92603046,
      "num_input_tokens_seen": 2278610,
      "step": 106,
      "time_per_iteration": 2.8466365337371826
    },
    {
      "auxiliary_loss_clip": 0.01551357,
      "auxiliary_loss_mlp": 0.01078363,
      "balance_loss_clip": 1.16814899,
      "balance_loss_mlp": 1.04600954,
      "epoch": 0.012865989298382733,
      "flos": 20595057361920.0,
      "grad_norm": 2.5523348530438783,
      "language_loss": 0.88534963,
      "learning_rate": 3.3852119241449547e-06,
      "loss": 0.91164684,
      "num_input_tokens_seen": 2297730,
      "step": 107,
      "time_per_iteration": 2.9068427085876465
    },
    {
      "auxiliary_loss_clip": 0.01539734,
      "auxiliary_loss_mlp": 0.01087736,
      "balance_loss_clip": 1.16477239,
      "balance_loss_mlp": 1.05247378,
      "epoch": 0.012986232189021825,
      "flos": 23948969978880.0,
      "grad_norm": 6.056361438684888,
      "language_loss": 0.96170348,
      "learning_rate": 3.3919510048344295e-06,
      "loss": 0.98797822,
      "num_input_tokens_seen": 2315740,
      "step": 108,
      "time_per_iteration": 2.910681962966919
    },
    {
      "auxiliary_loss_clip": 0.0152854,
      "auxiliary_loss_mlp": 0.01086855,
      "balance_loss_clip": 1.16209745,
      "balance_loss_mlp": 1.05497813,
      "epoch": 0.013106475079660914,
      "flos": 23725425686400.0,
      "grad_norm": 3.266021592298353,
      "language_loss": 0.86752617,
      "learning_rate": 3.3986279732976907e-06,
      "loss": 0.8936801,
      "num_input_tokens_seen": 2334215,
      "step": 109,
      "time_per_iteration": 2.8078019618988037
    },
    {
      "auxiliary_loss_clip": 0.01528404,
      "auxiliary_loss_mlp": 0.01081013,
      "balance_loss_clip": 1.16474009,
      "balance_loss_mlp": 1.05085301,
      "epoch": 0.013226717970300006,
      "flos": 21102438925440.0,
      "grad_norm": 1.9944893289246401,
      "language_loss": 0.95506442,
      "learning_rate": 3.4052439640284983e-06,
      "loss": 0.98115861,
      "num_input_tokens_seen": 2353130,
      "step": 110,
      "time_per_iteration": 2.846956491470337
    },
    {
      "auxiliary_loss_clip": 0.01530032,
      "auxiliary_loss_mlp": 0.01080789,
      "balance_loss_clip": 1.16352606,
      "balance_loss_mlp": 1.05065274,
      "epoch": 0.013346960860939098,
      "flos": 24863902231680.0,
      "grad_norm": 2.0016917695735503,
      "language_loss": 0.81261635,
      "learning_rate": 3.4118000807190217e-06,
      "loss": 0.83872455,
      "num_input_tokens_seen": 2374010,
      "step": 111,
      "time_per_iteration": 2.78352427482605
    },
    {
      "auxiliary_loss_clip": 0.01528775,
      "auxiliary_loss_mlp": 0.01071478,
      "balance_loss_clip": 1.16167212,
      "balance_loss_mlp": 1.04079378,
      "epoch": 0.013467203751578187,
      "flos": 28181940140160.0,
      "grad_norm": 1.7742805592110713,
      "language_loss": 0.75800318,
      "learning_rate": 3.4182973973648723e-06,
      "loss": 0.78400564,
      "num_input_tokens_seen": 2395220,
      "step": 112,
      "time_per_iteration": 2.9759156703948975
    },
    {
      "auxiliary_loss_clip": 0.01514226,
      "auxiliary_loss_mlp": 0.01097835,
      "balance_loss_clip": 1.16004682,
      "balance_loss_mlp": 1.06743622,
      "epoch": 0.013587446642217279,
      "flos": 18916233546240.0,
      "grad_norm": 2.602529591233348,
      "language_loss": 0.95265234,
      "learning_rate": 3.424736959321014e-06,
      "loss": 0.97877288,
      "num_input_tokens_seen": 2413025,
      "step": 113,
      "time_per_iteration": 2.8614418506622314
    },
    {
      "auxiliary_loss_clip": 0.01520047,
      "auxiliary_loss_mlp": 0.01091537,
      "balance_loss_clip": 1.15839767,
      "balance_loss_mlp": 1.05875492,
      "epoch": 0.01370768953285637,
      "flos": 23988615615360.0,
      "grad_norm": 2.1547496399024806,
      "language_loss": 0.88644689,
      "learning_rate": 3.431119784311155e-06,
      "loss": 0.91256273,
      "num_input_tokens_seen": 2432700,
      "step": 114,
      "time_per_iteration": 2.797990322113037
    },
    {
      "auxiliary_loss_clip": 0.01508017,
      "auxiliary_loss_mlp": 0.01091136,
      "balance_loss_clip": 1.15766716,
      "balance_loss_mlp": 1.06044006,
      "epoch": 0.01382793242349546,
      "flos": 39202565512320.0,
      "grad_norm": 1.7339134971736905,
      "language_loss": 0.77667272,
      "learning_rate": 3.43744686339307e-06,
      "loss": 0.80266422,
      "num_input_tokens_seen": 2455020,
      "step": 115,
      "time_per_iteration": 2.9621171951293945
    },
    {
      "auxiliary_loss_clip": 0.01497835,
      "auxiliary_loss_mlp": 0.01067453,
      "balance_loss_clip": 1.15251851,
      "balance_loss_mlp": 1.03947449,
      "epoch": 0.013948175314134552,
      "flos": 41353506714240.0,
      "grad_norm": 2.1547758033376687,
      "language_loss": 0.90842688,
      "learning_rate": 3.44371916188212e-06,
      "loss": 0.93407983,
      "num_input_tokens_seen": 2475775,
      "step": 116,
      "time_per_iteration": 2.8946290016174316
    },
    {
      "auxiliary_loss_clip": 0.01493234,
      "auxiliary_loss_mlp": 0.01090692,
      "balance_loss_clip": 1.14891481,
      "balance_loss_mlp": 1.06178355,
      "epoch": 0.014068418204773643,
      "flos": 22453542028800.0,
      "grad_norm": 2.782012181319536,
      "language_loss": 0.86166382,
      "learning_rate": 3.449937620235143e-06,
      "loss": 0.88750303,
      "num_input_tokens_seen": 2496370,
      "step": 117,
      "time_per_iteration": 2.7622833251953125
    },
    {
      "auxiliary_loss_clip": 0.01497608,
      "auxiliary_loss_mlp": 0.01075604,
      "balance_loss_clip": 1.15294349,
      "balance_loss_mlp": 1.04694593,
      "epoch": 0.014188661095412733,
      "flos": 23805147922560.0,
      "grad_norm": 1.7982760342484674,
      "language_loss": 0.89460754,
      "learning_rate": 3.456103154896722e-06,
      "loss": 0.92033958,
      "num_input_tokens_seen": 2517645,
      "step": 118,
      "time_per_iteration": 2.7412095069885254
    },
    {
      "auxiliary_loss_clip": 0.01489481,
      "auxiliary_loss_mlp": 0.01078194,
      "balance_loss_clip": 1.14973783,
      "balance_loss_mlp": 1.05215847,
      "epoch": 0.014308903986051825,
      "flos": 23660248458240.0,
      "grad_norm": 1.9699177898721085,
      "language_loss": 0.92343593,
      "learning_rate": 3.462216659109757e-06,
      "loss": 0.94911265,
      "num_input_tokens_seen": 2537825,
      "step": 119,
      "time_per_iteration": 4.601858615875244
    },
    {
      "auxiliary_loss_clip": 0.01507187,
      "auxiliary_loss_mlp": 0.01080168,
      "balance_loss_clip": 1.15565383,
      "balance_loss_mlp": 1.05277371,
      "epoch": 0.014429146876690916,
      "flos": 20667991927680.0,
      "grad_norm": 2.369660640377774,
      "language_loss": 0.8532933,
      "learning_rate": 3.4682790036921077e-06,
      "loss": 0.87916684,
      "num_input_tokens_seen": 2556485,
      "step": 120,
      "time_per_iteration": 4.709122180938721
    },
    {
      "auxiliary_loss_clip": 0.01474709,
      "auxiliary_loss_mlp": 0.01078605,
      "balance_loss_clip": 1.14706218,
      "balance_loss_mlp": 1.0529865,
      "epoch": 0.014549389767330006,
      "flos": 20229199384320.0,
      "grad_norm": 1.943612048652146,
      "language_loss": 0.8316592,
      "learning_rate": 3.4742910377810193e-06,
      "loss": 0.85719234,
      "num_input_tokens_seen": 2573945,
      "step": 121,
      "time_per_iteration": 2.846827983856201
    },
    {
      "auxiliary_loss_clip": 0.01478946,
      "auxiliary_loss_mlp": 0.01075063,
      "balance_loss_clip": 1.14681125,
      "balance_loss_mlp": 1.04992187,
      "epoch": 0.014669632657969098,
      "flos": 18004174381440.0,
      "grad_norm": 1.9949528070322342,
      "language_loss": 0.88655406,
      "learning_rate": 3.4802535895469042e-06,
      "loss": 0.91209412,
      "num_input_tokens_seen": 2592695,
      "step": 122,
      "time_per_iteration": 2.7874794006347656
    },
    {
      "auxiliary_loss_clip": 0.01479485,
      "auxiliary_loss_mlp": 0.01079413,
      "balance_loss_clip": 1.14765573,
      "balance_loss_mlp": 1.05388999,
      "epoch": 0.01478987554860819,
      "flos": 22741796672640.0,
      "grad_norm": 2.1202775329654764,
      "language_loss": 0.89679861,
      "learning_rate": 3.4861674668779934e-06,
      "loss": 0.9223876,
      "num_input_tokens_seen": 2610925,
      "step": 123,
      "time_per_iteration": 2.885566234588623
    },
    {
      "auxiliary_loss_clip": 0.01469684,
      "auxiliary_loss_mlp": 0.0107109,
      "balance_loss_clip": 1.14394712,
      "balance_loss_mlp": 1.04711723,
      "epoch": 0.01491011843924728,
      "flos": 17198590106880.0,
      "grad_norm": 1.9052288215273407,
      "language_loss": 0.84175205,
      "learning_rate": 3.492033458037272e-06,
      "loss": 0.86715978,
      "num_input_tokens_seen": 2629495,
      "step": 124,
      "time_per_iteration": 2.810899496078491
    },
    {
      "auxiliary_loss_clip": 0.01470372,
      "auxiliary_loss_mlp": 0.01070797,
      "balance_loss_clip": 1.14285707,
      "balance_loss_mlp": 1.04762268,
      "epoch": 0.01503036132988637,
      "flos": 17673867889920.0,
      "grad_norm": 2.398343924526405,
      "language_loss": 0.86740005,
      "learning_rate": 3.497852332293018e-06,
      "loss": 0.89281178,
      "num_input_tokens_seen": 2645070,
      "step": 125,
      "time_per_iteration": 2.813145875930786
    },
    {
      "auxiliary_loss_clip": 0.01473946,
      "auxiliary_loss_mlp": 0.01078166,
      "balance_loss_clip": 1.14581347,
      "balance_loss_mlp": 1.05537319,
      "epoch": 0.015150604220525462,
      "flos": 18878239935360.0,
      "grad_norm": 4.057477715060833,
      "language_loss": 0.96525335,
      "learning_rate": 3.5036248405242356e-06,
      "loss": 0.99077445,
      "num_input_tokens_seen": 2663825,
      "step": 126,
      "time_per_iteration": 2.7821781635284424
    },
    {
      "auxiliary_loss_clip": 0.01463577,
      "auxiliary_loss_mlp": 0.01083988,
      "balance_loss_clip": 1.13976955,
      "balance_loss_mlp": 1.06045604,
      "epoch": 0.015270847111164552,
      "flos": 39420184060800.0,
      "grad_norm": 2.0587146449584823,
      "language_loss": 0.82760453,
      "learning_rate": 3.509351715802146e-06,
      "loss": 0.85308015,
      "num_input_tokens_seen": 2684710,
      "step": 127,
      "time_per_iteration": 2.9877214431762695
    },
    {
      "auxiliary_loss_clip": 0.01465891,
      "auxiliary_loss_mlp": 0.01078648,
      "balance_loss_clip": 1.13728678,
      "balance_loss_mlp": 1.0551405,
      "epoch": 0.015391090001803644,
      "flos": 43762466286720.0,
      "grad_norm": 2.1694224815178624,
      "language_loss": 0.78385413,
      "learning_rate": 3.5150336739488763e-06,
      "loss": 0.80929959,
      "num_input_tokens_seen": 2706995,
      "step": 128,
      "time_per_iteration": 2.9644930362701416
    },
    {
      "auxiliary_loss_clip": 0.0145757,
      "auxiliary_loss_mlp": 0.01078615,
      "balance_loss_clip": 1.13832891,
      "balance_loss_mlp": 1.0593152,
      "epoch": 0.015511332892442733,
      "flos": 18916341287040.0,
      "grad_norm": 1.7437125991862497,
      "language_loss": 0.83822781,
      "learning_rate": 3.5206714140744143e-06,
      "loss": 0.86358964,
      "num_input_tokens_seen": 2727050,
      "step": 129,
      "time_per_iteration": 2.9021432399749756
    },
    {
      "auxiliary_loss_clip": 0.01470444,
      "auxiliary_loss_mlp": 0.01075014,
      "balance_loss_clip": 1.14761269,
      "balance_loss_mlp": 1.05298448,
      "epoch": 0.015631575783081827,
      "flos": 24535283679360.0,
      "grad_norm": 2.5020935259345984,
      "language_loss": 0.87468767,
      "learning_rate": 3.5262656190928208e-06,
      "loss": 0.90014219,
      "num_input_tokens_seen": 2745350,
      "step": 130,
      "time_per_iteration": 2.8288733959198
    },
    {
      "auxiliary_loss_clip": 0.01462186,
      "auxiliary_loss_mlp": 0.010452,
      "balance_loss_clip": 1.21475327,
      "balance_loss_mlp": 1.03985906,
      "epoch": 0.015751818673720917,
      "flos": 62328536098560.0,
      "grad_norm": 1.0508755128627707,
      "language_loss": 0.71512616,
      "learning_rate": 3.5318169562186737e-06,
      "loss": 0.74019998,
      "num_input_tokens_seen": 2814195,
      "step": 131,
      "time_per_iteration": 3.4526243209838867
    },
    {
      "auxiliary_loss_clip": 0.01459551,
      "auxiliary_loss_mlp": 0.01077592,
      "balance_loss_clip": 1.1421237,
      "balance_loss_mlp": 1.05582392,
      "epoch": 0.015872061564360006,
      "flos": 23878549365120.0,
      "grad_norm": 2.085814402451067,
      "language_loss": 0.82091594,
      "learning_rate": 3.5373260774446292e-06,
      "loss": 0.84628737,
      "num_input_tokens_seen": 2834645,
      "step": 132,
      "time_per_iteration": 2.815314292907715
    },
    {
      "auxiliary_loss_clip": 0.01448901,
      "auxiliary_loss_mlp": 0.01068681,
      "balance_loss_clip": 1.13513494,
      "balance_loss_mlp": 1.04783106,
      "epoch": 0.0159923044549991,
      "flos": 23367899664000.0,
      "grad_norm": 1.7192681825715221,
      "language_loss": 0.90388846,
      "learning_rate": 3.542793620000961e-06,
      "loss": 0.92906427,
      "num_input_tokens_seen": 2854120,
      "step": 133,
      "time_per_iteration": 2.8348300457000732
    },
    {
      "auxiliary_loss_clip": 0.01449391,
      "auxiliary_loss_mlp": 0.01074871,
      "balance_loss_clip": 1.13929522,
      "balance_loss_mlp": 1.05317521,
      "epoch": 0.01611254734563819,
      "flos": 17858305249920.0,
      "grad_norm": 2.128223210983026,
      "language_loss": 0.86716545,
      "learning_rate": 3.5482202067978894e-06,
      "loss": 0.89240807,
      "num_input_tokens_seen": 2871330,
      "step": 134,
      "time_per_iteration": 2.736873149871826
    },
    {
      "auxiliary_loss_clip": 0.01457087,
      "auxiliary_loss_mlp": 0.01068735,
      "balance_loss_clip": 1.13819087,
      "balance_loss_mlp": 1.04975641,
      "epoch": 0.01623279023627728,
      "flos": 20954774113920.0,
      "grad_norm": 2.0562355376697723,
      "language_loss": 0.75920033,
      "learning_rate": 3.553606446851471e-06,
      "loss": 0.78445858,
      "num_input_tokens_seen": 2888070,
      "step": 135,
      "time_per_iteration": 2.781238317489624
    },
    {
      "auxiliary_loss_clip": 0.01445996,
      "auxiliary_loss_mlp": 0.01069427,
      "balance_loss_clip": 1.13755643,
      "balance_loss_mlp": 1.04955482,
      "epoch": 0.016353033126916373,
      "flos": 15742412743680.0,
      "grad_norm": 2.3407852791183994,
      "language_loss": 0.83338398,
      "learning_rate": 3.5589529356937613e-06,
      "loss": 0.85853827,
      "num_input_tokens_seen": 2906465,
      "step": 136,
      "time_per_iteration": 2.6851398944854736
    },
    {
      "auxiliary_loss_clip": 0.0144159,
      "auxiliary_loss_mlp": 0.01068305,
      "balance_loss_clip": 1.13223243,
      "balance_loss_mlp": 1.04658532,
      "epoch": 0.016473276017555463,
      "flos": 18807280617600.0,
      "grad_norm": 1.8097695836599872,
      "language_loss": 0.76949793,
      "learning_rate": 3.5642602557679627e-06,
      "loss": 0.79459691,
      "num_input_tokens_seen": 2924915,
      "step": 137,
      "time_per_iteration": 2.8226726055145264
    },
    {
      "auxiliary_loss_clip": 0.01444402,
      "auxiliary_loss_mlp": 0.01081288,
      "balance_loss_clip": 1.13703656,
      "balance_loss_mlp": 1.06152368,
      "epoch": 0.016593518908194552,
      "flos": 24352641999360.0,
      "grad_norm": 2.2618611741412438,
      "language_loss": 0.84099442,
      "learning_rate": 3.569528976809202e-06,
      "loss": 0.86625129,
      "num_input_tokens_seen": 2942130,
      "step": 138,
      "time_per_iteration": 2.82380747795105
    },
    {
      "auxiliary_loss_clip": 0.01439213,
      "auxiliary_loss_mlp": 0.01068815,
      "balance_loss_clip": 1.13129473,
      "balance_loss_mlp": 1.04817986,
      "epoch": 0.016713761798833646,
      "flos": 22346133384960.0,
      "grad_norm": 1.92120249533572,
      "language_loss": 0.89787722,
      "learning_rate": 3.5747596562115522e-06,
      "loss": 0.92295754,
      "num_input_tokens_seen": 2962745,
      "step": 139,
      "time_per_iteration": 2.8132035732269287
    },
    {
      "auxiliary_loss_clip": 0.01448118,
      "auxiliary_loss_mlp": 0.01069869,
      "balance_loss_clip": 1.13341045,
      "balance_loss_mlp": 1.04786277,
      "epoch": 0.016834004689472735,
      "flos": 17821820010240.0,
      "grad_norm": 2.2276669399748785,
      "language_loss": 0.90866935,
      "learning_rate": 3.5799528393819138e-06,
      "loss": 0.93384928,
      "num_input_tokens_seen": 2981825,
      "step": 140,
      "time_per_iteration": 2.8224411010742188
    },
    {
      "auxiliary_loss_clip": 0.01424003,
      "auxiliary_loss_mlp": 0.01058729,
      "balance_loss_clip": 1.12578893,
      "balance_loss_mlp": 1.03984594,
      "epoch": 0.016954247580111825,
      "flos": 20519501103360.0,
      "grad_norm": 1.9284647642861006,
      "language_loss": 0.87934023,
      "learning_rate": 3.585109060081286e-06,
      "loss": 0.90416753,
      "num_input_tokens_seen": 3001625,
      "step": 141,
      "time_per_iteration": 2.8157191276550293
    },
    {
      "auxiliary_loss_clip": 0.0142897,
      "auxiliary_loss_mlp": 0.01071777,
      "balance_loss_clip": 1.12350273,
      "balance_loss_mlp": 1.05186939,
      "epoch": 0.017074490470750915,
      "flos": 22088869200000.0,
      "grad_norm": 2.6792858479647883,
      "language_loss": 0.78499019,
      "learning_rate": 3.590228840753992e-06,
      "loss": 0.80999768,
      "num_input_tokens_seen": 3022055,
      "step": 142,
      "time_per_iteration": 2.8258938789367676
    },
    {
      "auxiliary_loss_clip": 0.01427905,
      "auxiliary_loss_mlp": 0.01063849,
      "balance_loss_clip": 1.12618136,
      "balance_loss_mlp": 1.04553819,
      "epoch": 0.01719473336139001,
      "flos": 15997270717440.0,
      "grad_norm": 2.092391485090024,
      "language_loss": 0.87274098,
      "learning_rate": 3.5953126928453423e-06,
      "loss": 0.89765847,
      "num_input_tokens_seen": 3039605,
      "step": 143,
      "time_per_iteration": 2.6791129112243652
    },
    {
      "auxiliary_loss_clip": 0.01432818,
      "auxiliary_loss_mlp": 0.01059114,
      "balance_loss_clip": 1.13107371,
      "balance_loss_mlp": 1.04072011,
      "epoch": 0.017314976252029098,
      "flos": 22492038430080.0,
      "grad_norm": 2.116873403735208,
      "language_loss": 0.806862,
      "learning_rate": 3.600361117108239e-06,
      "loss": 0.83178139,
      "num_input_tokens_seen": 3059405,
      "step": 144,
      "time_per_iteration": 2.863239288330078
    },
    {
      "auxiliary_loss_clip": 0.01422679,
      "auxiliary_loss_mlp": 0.01057135,
      "balance_loss_clip": 1.12183857,
      "balance_loss_mlp": 1.03824019,
      "epoch": 0.017435219142668188,
      "flos": 22018053536640.0,
      "grad_norm": 2.052639115351349,
      "language_loss": 0.97150159,
      "learning_rate": 3.6053746038991616e-06,
      "loss": 0.99629974,
      "num_input_tokens_seen": 3078490,
      "step": 145,
      "time_per_iteration": 5.120325565338135
    },
    {
      "auxiliary_loss_clip": 0.01414625,
      "auxiliary_loss_mlp": 0.01009418,
      "balance_loss_clip": 1.17215109,
      "balance_loss_mlp": 1.00255108,
      "epoch": 0.01755546203330728,
      "flos": 72240526149120.0,
      "grad_norm": 1.0519147264420767,
      "language_loss": 0.58484626,
      "learning_rate": 3.6103536334639843e-06,
      "loss": 0.60908675,
      "num_input_tokens_seen": 3131755,
      "step": 146,
      "time_per_iteration": 3.461718797683716
    },
    {
      "auxiliary_loss_clip": 0.01418464,
      "auxiliary_loss_mlp": 0.01068471,
      "balance_loss_clip": 1.12316453,
      "balance_loss_mlp": 1.04943347,
      "epoch": 0.01767570492394637,
      "flos": 25337061112320.0,
      "grad_norm": 1.9974430043333418,
      "language_loss": 0.8527143,
      "learning_rate": 3.615298676214041e-06,
      "loss": 0.87758368,
      "num_input_tokens_seen": 3152035,
      "step": 147,
      "time_per_iteration": 3.813316822052002
    },
    {
      "auxiliary_loss_clip": 0.01415956,
      "auxiliary_loss_mlp": 0.01076494,
      "balance_loss_clip": 1.12090898,
      "balance_loss_mlp": 1.05737245,
      "epoch": 0.01779594781458546,
      "flos": 20449188230400.0,
      "grad_norm": 3.2874583308091134,
      "language_loss": 0.88807857,
      "learning_rate": 3.6202101929928317e-06,
      "loss": 0.91300309,
      "num_input_tokens_seen": 3170625,
      "step": 148,
      "time_per_iteration": 3.6785006523132324
    },
    {
      "auxiliary_loss_clip": 0.01411514,
      "auxiliary_loss_mlp": 0.01065662,
      "balance_loss_clip": 1.1206131,
      "balance_loss_mlp": 1.04875839,
      "epoch": 0.017916190705224554,
      "flos": 16253601148800.0,
      "grad_norm": 1.97392283640973,
      "language_loss": 0.88264978,
      "learning_rate": 3.6250886353337413e-06,
      "loss": 0.90742153,
      "num_input_tokens_seen": 3188155,
      "step": 149,
      "time_per_iteration": 2.8857944011688232
    },
    {
      "auxiliary_loss_clip": 0.01420296,
      "auxiliary_loss_mlp": 0.01060032,
      "balance_loss_clip": 1.12311161,
      "balance_loss_mlp": 1.04231763,
      "epoch": 0.018036433595863644,
      "flos": 23330588411520.0,
      "grad_norm": 4.512922033614906,
      "language_loss": 0.86243153,
      "learning_rate": 3.6299344457091488e-06,
      "loss": 0.88723481,
      "num_input_tokens_seen": 3209015,
      "step": 150,
      "time_per_iteration": 2.8048019409179688
    },
    {
      "auxiliary_loss_clip": 0.01416188,
      "auxiliary_loss_mlp": 0.01057314,
      "balance_loss_clip": 1.12134731,
      "balance_loss_mlp": 1.03936076,
      "epoch": 0.018156676486502734,
      "flos": 18588010043520.0,
      "grad_norm": 2.1290383131082047,
      "language_loss": 0.93703622,
      "learning_rate": 3.634748057771256e-06,
      "loss": 0.96177125,
      "num_input_tokens_seen": 3224955,
      "step": 151,
      "time_per_iteration": 2.8008227348327637
    },
    {
      "auxiliary_loss_clip": 0.01415145,
      "auxiliary_loss_mlp": 0.01072089,
      "balance_loss_clip": 1.12128651,
      "balance_loss_mlp": 1.05584061,
      "epoch": 0.018276919377141827,
      "flos": 25448707560960.0,
      "grad_norm": 2.581163696440333,
      "language_loss": 0.85484052,
      "learning_rate": 3.639529896584965e-06,
      "loss": 0.87971282,
      "num_input_tokens_seen": 3246330,
      "step": 152,
      "time_per_iteration": 2.779123544692993
    },
    {
      "auxiliary_loss_clip": 0.0141654,
      "auxiliary_loss_mlp": 0.01072071,
      "balance_loss_clip": 1.12209857,
      "balance_loss_mlp": 1.05451155,
      "epoch": 0.018397162267780917,
      "flos": 20047311889920.0,
      "grad_norm": 3.0912937841287222,
      "language_loss": 0.88584042,
      "learning_rate": 3.6442803788531233e-06,
      "loss": 0.91072649,
      "num_input_tokens_seen": 3264290,
      "step": 153,
      "time_per_iteration": 2.7715671062469482
    },
    {
      "auxiliary_loss_clip": 0.01412304,
      "auxiliary_loss_mlp": 0.01055409,
      "balance_loss_clip": 1.11962879,
      "balance_loss_mlp": 1.03649068,
      "epoch": 0.018517405158420007,
      "flos": 27565282425600.0,
      "grad_norm": 1.9518182589234072,
      "language_loss": 0.95964944,
      "learning_rate": 3.6489999131344357e-06,
      "loss": 0.98432654,
      "num_input_tokens_seen": 3287065,
      "step": 154,
      "time_per_iteration": 2.8485734462738037
    },
    {
      "auxiliary_loss_clip": 0.01399999,
      "auxiliary_loss_mlp": 0.01067466,
      "balance_loss_clip": 1.1170671,
      "balance_loss_mlp": 1.05179024,
      "epoch": 0.0186376480490591,
      "flos": 19354056422400.0,
      "grad_norm": 1.9879669453977855,
      "language_loss": 0.90595168,
      "learning_rate": 3.653688900054313e-06,
      "loss": 0.93062633,
      "num_input_tokens_seen": 3305595,
      "step": 155,
      "time_per_iteration": 2.7322568893432617
    },
    {
      "auxiliary_loss_clip": 0.01406807,
      "auxiliary_loss_mlp": 0.01064467,
      "balance_loss_clip": 1.11447883,
      "balance_loss_mlp": 1.04763508,
      "epoch": 0.01875789093969819,
      "flos": 26687840993280.0,
      "grad_norm": 1.860785764259434,
      "language_loss": 0.76076221,
      "learning_rate": 3.6583477325089526e-06,
      "loss": 0.78547502,
      "num_input_tokens_seen": 3326135,
      "step": 156,
      "time_per_iteration": 2.8410632610321045
    },
    {
      "auxiliary_loss_clip": 0.0140403,
      "auxiliary_loss_mlp": 0.01063176,
      "balance_loss_clip": 1.11799169,
      "balance_loss_mlp": 1.04662991,
      "epoch": 0.01887813383033728,
      "flos": 24353001135360.0,
      "grad_norm": 2.2422517963295303,
      "language_loss": 1.04473627,
      "learning_rate": 3.6629767958628916e-06,
      "loss": 1.06940818,
      "num_input_tokens_seen": 3343510,
      "step": 157,
      "time_per_iteration": 2.7347660064697266
    },
    {
      "auxiliary_loss_clip": 0.01405314,
      "auxiliary_loss_mlp": 0.01077657,
      "balance_loss_clip": 1.11686957,
      "balance_loss_mlp": 1.06080055,
      "epoch": 0.018998376720976373,
      "flos": 14647532330880.0,
      "grad_norm": 2.121453312419076,
      "language_loss": 0.85550094,
      "learning_rate": 3.667576468140291e-06,
      "loss": 0.88033068,
      "num_input_tokens_seen": 3361325,
      "step": 158,
      "time_per_iteration": 2.7819080352783203
    },
    {
      "auxiliary_loss_clip": 0.01394227,
      "auxiliary_loss_mlp": 0.01064341,
      "balance_loss_clip": 1.11028278,
      "balance_loss_mlp": 1.04774714,
      "epoch": 0.019118619611615463,
      "flos": 29305261146240.0,
      "grad_norm": 2.6766171486290165,
      "language_loss": 0.88976002,
      "learning_rate": 3.672147120210184e-06,
      "loss": 0.91434568,
      "num_input_tokens_seen": 3377925,
      "step": 159,
      "time_per_iteration": 2.7340800762176514
    },
    {
      "auxiliary_loss_clip": 0.013982,
      "auxiliary_loss_mlp": 0.01073857,
      "balance_loss_clip": 1.11681688,
      "balance_loss_mlp": 1.05714417,
      "epoch": 0.019238862502254553,
      "flos": 20886723797760.0,
      "grad_norm": 2.113298026802636,
      "language_loss": 0.86347437,
      "learning_rate": 3.6766891159659177e-06,
      "loss": 0.88819492,
      "num_input_tokens_seen": 3396335,
      "step": 160,
      "time_per_iteration": 2.7317209243774414
    },
    {
      "auxiliary_loss_clip": 0.01399305,
      "auxiliary_loss_mlp": 0.01057297,
      "balance_loss_clip": 1.11377871,
      "balance_loss_mlp": 1.04140055,
      "epoch": 0.019359105392893646,
      "flos": 21360672777600.0,
      "grad_norm": 2.7686993882953903,
      "language_loss": 0.87639689,
      "learning_rate": 3.6812028124990075e-06,
      "loss": 0.90096289,
      "num_input_tokens_seen": 3413605,
      "step": 161,
      "time_per_iteration": 2.7208690643310547
    },
    {
      "auxiliary_loss_clip": 0.01396022,
      "auxiliary_loss_mlp": 0.01073388,
      "balance_loss_clip": 1.11370099,
      "balance_loss_mlp": 1.05541146,
      "epoch": 0.019479348283532736,
      "flos": 16283729681280.0,
      "grad_norm": 5.7972919204508395,
      "language_loss": 0.8147496,
      "learning_rate": 3.6856885602676016e-06,
      "loss": 0.83944368,
      "num_input_tokens_seen": 3429640,
      "step": 162,
      "time_per_iteration": 2.7274043560028076
    },
    {
      "auxiliary_loss_clip": 0.01398433,
      "auxiliary_loss_mlp": 0.01070251,
      "balance_loss_clip": 1.11462641,
      "balance_loss_mlp": 1.0539546,
      "epoch": 0.019599591174171826,
      "flos": 22091239497600.0,
      "grad_norm": 2.357192159477666,
      "language_loss": 0.94264972,
      "learning_rate": 3.6901467032597733e-06,
      "loss": 0.9673366,
      "num_input_tokens_seen": 3448125,
      "step": 163,
      "time_per_iteration": 2.699155569076538
    },
    {
      "auxiliary_loss_clip": 0.01396837,
      "auxiliary_loss_mlp": 0.01063029,
      "balance_loss_clip": 1.11281908,
      "balance_loss_mlp": 1.04611313,
      "epoch": 0.01971983406481092,
      "flos": 19609668581760.0,
      "grad_norm": 2.539963040826809,
      "language_loss": 0.87456429,
      "learning_rate": 3.694577579151804e-06,
      "loss": 0.89916301,
      "num_input_tokens_seen": 3466535,
      "step": 164,
      "time_per_iteration": 2.7642979621887207
    },
    {
      "auxiliary_loss_clip": 0.01398641,
      "auxiliary_loss_mlp": 0.01065154,
      "balance_loss_clip": 1.1153537,
      "balance_loss_mlp": 1.04892969,
      "epoch": 0.01984007695545001,
      "flos": 19099342103040.0,
      "grad_norm": 2.8864695768556583,
      "language_loss": 0.73594671,
      "learning_rate": 3.6989815194616703e-06,
      "loss": 0.76058471,
      "num_input_tokens_seen": 3483730,
      "step": 165,
      "time_per_iteration": 2.7236971855163574
    },
    {
      "auxiliary_loss_clip": 0.01397984,
      "auxiliary_loss_mlp": 0.01067672,
      "balance_loss_clip": 1.11458433,
      "balance_loss_mlp": 1.05041063,
      "epoch": 0.0199603198460891,
      "flos": 20848406964480.0,
      "grad_norm": 2.5510549362314627,
      "language_loss": 0.80118227,
      "learning_rate": 3.703358849697888e-06,
      "loss": 0.82583886,
      "num_input_tokens_seen": 3503640,
      "step": 166,
      "time_per_iteration": 2.746067523956299
    },
    {
      "auxiliary_loss_clip": 0.01390423,
      "auxiliary_loss_mlp": 0.01070764,
      "balance_loss_clip": 1.11269021,
      "balance_loss_mlp": 1.05597067,
      "epoch": 0.020080562736728192,
      "flos": 21870747861120.0,
      "grad_norm": 1.8246953254952698,
      "language_loss": 0.82747579,
      "learning_rate": 3.7077098895038803e-06,
      "loss": 0.85208762,
      "num_input_tokens_seen": 3523010,
      "step": 167,
      "time_per_iteration": 2.726393699645996
    },
    {
      "auxiliary_loss_clip": 0.01394371,
      "auxiliary_loss_mlp": 0.01061185,
      "balance_loss_clip": 1.11159599,
      "balance_loss_mlp": 1.04468632,
      "epoch": 0.020200805627367282,
      "flos": 21688788539520.0,
      "grad_norm": 2.796741474940629,
      "language_loss": 0.96997058,
      "learning_rate": 3.712034952798045e-06,
      "loss": 0.99452615,
      "num_input_tokens_seen": 3541125,
      "step": 168,
      "time_per_iteration": 2.7018864154815674
    },
    {
      "auxiliary_loss_clip": 0.01393897,
      "auxiliary_loss_mlp": 0.01071239,
      "balance_loss_clip": 1.10993385,
      "balance_loss_mlp": 1.05222511,
      "epoch": 0.02032104851800637,
      "flos": 33543043729920.0,
      "grad_norm": 2.3131186635480496,
      "language_loss": 0.84632051,
      "learning_rate": 3.7163343479096656e-06,
      "loss": 0.87097192,
      "num_input_tokens_seen": 3562700,
      "step": 169,
      "time_per_iteration": 2.880140542984009
    },
    {
      "auxiliary_loss_clip": 0.01387796,
      "auxiliary_loss_mlp": 0.01061368,
      "balance_loss_clip": 1.11165226,
      "balance_loss_mlp": 1.04613256,
      "epoch": 0.020441291408645465,
      "flos": 31686965274240.0,
      "grad_norm": 2.041580037107391,
      "language_loss": 0.82826006,
      "learning_rate": 3.720608377710802e-06,
      "loss": 0.85275167,
      "num_input_tokens_seen": 3582790,
      "step": 170,
      "time_per_iteration": 2.7907161712646484
    },
    {
      "auxiliary_loss_clip": 0.01383946,
      "auxiliary_loss_mlp": 0.01057079,
      "balance_loss_clip": 1.10727096,
      "balance_loss_mlp": 1.04013944,
      "epoch": 0.020561534299284555,
      "flos": 20886687884160.0,
      "grad_norm": 2.692867177152872,
      "language_loss": 0.86521423,
      "learning_rate": 3.7248573397443277e-06,
      "loss": 0.88962448,
      "num_input_tokens_seen": 3601715,
      "step": 171,
      "time_per_iteration": 3.852539539337158
    },
    {
      "auxiliary_loss_clip": 0.01384792,
      "auxiliary_loss_mlp": 0.01071144,
      "balance_loss_clip": 1.11004758,
      "balance_loss_mlp": 1.05506313,
      "epoch": 0.020681777189923645,
      "flos": 20996610480000.0,
      "grad_norm": 3.020187005665384,
      "language_loss": 0.97540665,
      "learning_rate": 3.729081526348224e-06,
      "loss": 0.99996603,
      "num_input_tokens_seen": 3620245,
      "step": 172,
      "time_per_iteration": 3.786903142929077
    },
    {
      "auxiliary_loss_clip": 0.01381707,
      "auxiliary_loss_mlp": 0.01053099,
      "balance_loss_clip": 1.10604775,
      "balance_loss_mlp": 1.03521729,
      "epoch": 0.020802020080562738,
      "flos": 28257532312320.0,
      "grad_norm": 3.1170283324658583,
      "language_loss": 0.85064507,
      "learning_rate": 3.7332812247762777e-06,
      "loss": 0.87499309,
      "num_input_tokens_seen": 3641545,
      "step": 173,
      "time_per_iteration": 3.7448065280914307
    },
    {
      "auxiliary_loss_clip": 0.01385835,
      "auxiliary_loss_mlp": 0.01051962,
      "balance_loss_clip": 1.11195958,
      "balance_loss_mlp": 1.03623271,
      "epoch": 0.020922262971201828,
      "flos": 19681274344320.0,
      "grad_norm": 2.3116020980297836,
      "language_loss": 0.95834816,
      "learning_rate": 3.737456717315293e-06,
      "loss": 0.9827261,
      "num_input_tokens_seen": 3660510,
      "step": 174,
      "time_per_iteration": 3.7517755031585693
    },
    {
      "auxiliary_loss_clip": 0.01381008,
      "auxiliary_loss_mlp": 0.0106457,
      "balance_loss_clip": 1.11184216,
      "balance_loss_mlp": 1.04864323,
      "epoch": 0.021042505861840918,
      "flos": 15666353694720.0,
      "grad_norm": 1.7733400957192014,
      "language_loss": 0.90811592,
      "learning_rate": 3.7416082813989552e-06,
      "loss": 0.93257171,
      "num_input_tokens_seen": 3677505,
      "step": 175,
      "time_per_iteration": 2.7190804481506348
    },
    {
      "auxiliary_loss_clip": 0.01389405,
      "auxiliary_loss_mlp": 0.01054726,
      "balance_loss_clip": 1.10935295,
      "balance_loss_mlp": 1.03872848,
      "epoch": 0.02116274875248001,
      "flos": 21142012734720.0,
      "grad_norm": 2.467342364335822,
      "language_loss": 0.89228469,
      "learning_rate": 3.745736189718439e-06,
      "loss": 0.91672599,
      "num_input_tokens_seen": 3696760,
      "step": 176,
      "time_per_iteration": 2.738025665283203
    },
    {
      "auxiliary_loss_clip": 0.01373882,
      "auxiliary_loss_mlp": 0.01056231,
      "balance_loss_clip": 1.10465145,
      "balance_loss_mlp": 1.04113889,
      "epoch": 0.0212829916431191,
      "flos": 24715770543360.0,
      "grad_norm": 3.9061764850881984,
      "language_loss": 0.72664237,
      "learning_rate": 3.749840710329894e-06,
      "loss": 0.75094348,
      "num_input_tokens_seen": 3717465,
      "step": 177,
      "time_per_iteration": 2.741569757461548
    },
    {
      "auxiliary_loss_clip": 0.01397364,
      "auxiliary_loss_mlp": 0.01065142,
      "balance_loss_clip": 1.11235857,
      "balance_loss_mlp": 1.04655778,
      "epoch": 0.02140323453375819,
      "flos": 16645493508480.0,
      "grad_norm": 3.918375092947578,
      "language_loss": 0.97885656,
      "learning_rate": 3.7539221067588938e-06,
      "loss": 1.00348163,
      "num_input_tokens_seen": 3731440,
      "step": 178,
      "time_per_iteration": 2.675069570541382
    },
    {
      "auxiliary_loss_clip": 0.01387759,
      "auxiliary_loss_mlp": 0.01068711,
      "balance_loss_clip": 1.10893238,
      "balance_loss_mlp": 1.05124748,
      "epoch": 0.021523477424397284,
      "flos": 20299332689280.0,
      "grad_norm": 3.3247023749737092,
      "language_loss": 0.93913186,
      "learning_rate": 3.757980638101964e-06,
      "loss": 0.96369654,
      "num_input_tokens_seen": 3744935,
      "step": 179,
      "time_per_iteration": 2.7001724243164062
    },
    {
      "auxiliary_loss_clip": 0.01392645,
      "auxiliary_loss_mlp": 0.01059297,
      "balance_loss_clip": 1.11181569,
      "balance_loss_mlp": 1.04275036,
      "epoch": 0.021643720315036374,
      "flos": 26104005331200.0,
      "grad_norm": 2.073345825983855,
      "language_loss": 0.89317715,
      "learning_rate": 3.7620165591252806e-06,
      "loss": 0.91769648,
      "num_input_tokens_seen": 3763035,
      "step": 180,
      "time_per_iteration": 2.7686524391174316
    },
    {
      "auxiliary_loss_clip": 0.01374042,
      "auxiliary_loss_mlp": 0.01066175,
      "balance_loss_clip": 1.10637021,
      "balance_loss_mlp": 1.0494498,
      "epoch": 0.021763963205675464,
      "flos": 24787663614720.0,
      "grad_norm": 1.7737130172264786,
      "language_loss": 0.94448137,
      "learning_rate": 3.766030120360636e-06,
      "loss": 0.96888351,
      "num_input_tokens_seen": 3782665,
      "step": 181,
      "time_per_iteration": 2.7097997665405273
    },
    {
      "auxiliary_loss_clip": 0.01381451,
      "auxiliary_loss_mlp": 0.01065529,
      "balance_loss_clip": 1.10776222,
      "balance_loss_mlp": 1.04820776,
      "epoch": 0.021884206096314557,
      "flos": 25813559957760.0,
      "grad_norm": 2.245012251124866,
      "language_loss": 0.90298784,
      "learning_rate": 3.7700215681987578e-06,
      "loss": 0.92745763,
      "num_input_tokens_seen": 3802435,
      "step": 182,
      "time_per_iteration": 2.818112373352051
    },
    {
      "auxiliary_loss_clip": 0.01378682,
      "auxiliary_loss_mlp": 0.01062576,
      "balance_loss_clip": 1.10545254,
      "balance_loss_mlp": 1.04536176,
      "epoch": 0.022004448986953647,
      "flos": 20082719721600.0,
      "grad_norm": 1.859995095833018,
      "language_loss": 0.82321972,
      "learning_rate": 3.7739911449800767e-06,
      "loss": 0.84763229,
      "num_input_tokens_seen": 3822490,
      "step": 183,
      "time_per_iteration": 2.74141526222229
    },
    {
      "auxiliary_loss_clip": 0.01381243,
      "auxiliary_loss_mlp": 0.01053697,
      "balance_loss_clip": 1.10621166,
      "balance_loss_mlp": 1.0374012,
      "epoch": 0.022124691877592736,
      "flos": 20480609652480.0,
      "grad_norm": 1.9082293728855013,
      "language_loss": 0.80553555,
      "learning_rate": 3.7779390890830114e-06,
      "loss": 0.82988501,
      "num_input_tokens_seen": 3841140,
      "step": 184,
      "time_per_iteration": 2.6918177604675293
    },
    {
      "auxiliary_loss_clip": 0.01379911,
      "auxiliary_loss_mlp": 0.01057986,
      "balance_loss_clip": 1.10476851,
      "balance_loss_mlp": 1.04176188,
      "epoch": 0.02224493476823183,
      "flos": 23586847015680.0,
      "grad_norm": 1.7371612011106197,
      "language_loss": 0.85874104,
      "learning_rate": 3.7818656350098723e-06,
      "loss": 0.88311994,
      "num_input_tokens_seen": 3862090,
      "step": 185,
      "time_per_iteration": 2.778953790664673
    },
    {
      "auxiliary_loss_clip": 0.01380741,
      "auxiliary_loss_mlp": 0.01056001,
      "balance_loss_clip": 1.10670125,
      "balance_loss_mlp": 1.03912091,
      "epoch": 0.02236517765887092,
      "flos": 16909940413440.0,
      "grad_norm": 2.409680354111259,
      "language_loss": 0.77365446,
      "learning_rate": 3.7857710134704447e-06,
      "loss": 0.79802191,
      "num_input_tokens_seen": 3881025,
      "step": 186,
      "time_per_iteration": 2.806182622909546
    },
    {
      "auxiliary_loss_clip": 0.01374866,
      "auxiliary_loss_mlp": 0.01080654,
      "balance_loss_clip": 1.10657775,
      "balance_loss_mlp": 1.06454897,
      "epoch": 0.02248542054951001,
      "flos": 43508182930560.0,
      "grad_norm": 1.9057256571478942,
      "language_loss": 0.7919991,
      "learning_rate": 3.7896554514633234e-06,
      "loss": 0.81655431,
      "num_input_tokens_seen": 3905310,
      "step": 187,
      "time_per_iteration": 2.887655735015869
    },
    {
      "auxiliary_loss_clip": 0.01366278,
      "auxiliary_loss_mlp": 0.01065949,
      "balance_loss_clip": 1.10226941,
      "balance_loss_mlp": 1.04866338,
      "epoch": 0.022605663440149103,
      "flos": 23367648268800.0,
      "grad_norm": 1.9272245310281728,
      "language_loss": 0.84168959,
      "learning_rate": 3.7935191723550955e-06,
      "loss": 0.86601186,
      "num_input_tokens_seen": 3924265,
      "step": 188,
      "time_per_iteration": 2.734398365020752
    },
    {
      "auxiliary_loss_clip": 0.01370288,
      "auxiliary_loss_mlp": 0.01058897,
      "balance_loss_clip": 1.10495234,
      "balance_loss_mlp": 1.04347134,
      "epoch": 0.022725906330788193,
      "flos": 29019915504000.0,
      "grad_norm": 2.2157329964628762,
      "language_loss": 0.8835389,
      "learning_rate": 3.797362395957408e-06,
      "loss": 0.90783077,
      "num_input_tokens_seen": 3944830,
      "step": 189,
      "time_per_iteration": 2.83530592918396
    },
    {
      "auxiliary_loss_clip": 0.01381865,
      "auxiliary_loss_mlp": 0.01060809,
      "balance_loss_clip": 1.10783255,
      "balance_loss_mlp": 1.04408407,
      "epoch": 0.022846149221427282,
      "flos": 24496176746880.0,
      "grad_norm": 2.1258099117445886,
      "language_loss": 0.78331459,
      "learning_rate": 3.8011853386020055e-06,
      "loss": 0.80774134,
      "num_input_tokens_seen": 3965735,
      "step": 190,
      "time_per_iteration": 2.7453553676605225
    },
    {
      "auxiliary_loss_clip": 0.0137327,
      "auxiliary_loss_mlp": 0.01056506,
      "balance_loss_clip": 1.10553849,
      "balance_loss_mlp": 1.03928065,
      "epoch": 0.022966392112066376,
      "flos": 15523537219200.0,
      "grad_norm": 2.6933432709788057,
      "language_loss": 0.89586079,
      "learning_rate": 3.804988213213804e-06,
      "loss": 0.92015857,
      "num_input_tokens_seen": 3983975,
      "step": 191,
      "time_per_iteration": 2.7105584144592285
    },
    {
      "auxiliary_loss_clip": 0.01362617,
      "auxiliary_loss_mlp": 0.01010753,
      "balance_loss_clip": 1.14914536,
      "balance_loss_mlp": 1.00407696,
      "epoch": 0.023086635002705466,
      "flos": 55650408433920.0,
      "grad_norm": 1.0224292171404512,
      "language_loss": 0.63202894,
      "learning_rate": 3.808771229382049e-06,
      "loss": 0.65576261,
      "num_input_tokens_seen": 4043440,
      "step": 192,
      "time_per_iteration": 3.1564300060272217
    },
    {
      "auxiliary_loss_clip": 0.01369367,
      "auxiliary_loss_mlp": 0.01061952,
      "balance_loss_clip": 1.10354054,
      "balance_loss_mlp": 1.04566824,
      "epoch": 0.023206877893344555,
      "flos": 19313441118720.0,
      "grad_norm": 1.8597086328313224,
      "language_loss": 0.84496146,
      "learning_rate": 3.8125345934296324e-06,
      "loss": 0.86927462,
      "num_input_tokens_seen": 4061750,
      "step": 193,
      "time_per_iteration": 2.652597188949585
    },
    {
      "auxiliary_loss_clip": 0.01370796,
      "auxiliary_loss_mlp": 0.01055918,
      "balance_loss_clip": 1.10159993,
      "balance_loss_mlp": 1.0393362,
      "epoch": 0.02332712078398365,
      "flos": 23072965090560.0,
      "grad_norm": 2.271024951190071,
      "language_loss": 0.87875211,
      "learning_rate": 3.81627850848061e-06,
      "loss": 0.90301931,
      "num_input_tokens_seen": 4082345,
      "step": 194,
      "time_per_iteration": 2.786513328552246
    },
    {
      "auxiliary_loss_clip": 0.01365816,
      "auxiliary_loss_mlp": 0.01058275,
      "balance_loss_clip": 1.10169458,
      "balance_loss_mlp": 1.04326677,
      "epoch": 0.02344736367462274,
      "flos": 24425971614720.0,
      "grad_norm": 5.002127193720737,
      "language_loss": 0.86228883,
      "learning_rate": 3.820003174525994e-06,
      "loss": 0.8865298,
      "num_input_tokens_seen": 4101770,
      "step": 195,
      "time_per_iteration": 2.7558159828186035
    },
    {
      "auxiliary_loss_clip": 0.01366917,
      "auxiliary_loss_mlp": 0.01059319,
      "balance_loss_clip": 1.10314548,
      "balance_loss_mlp": 1.04433453,
      "epoch": 0.02356760656526183,
      "flos": 21579799697280.0,
      "grad_norm": 3.0944655480158585,
      "language_loss": 0.8266834,
      "learning_rate": 3.823708788487851e-06,
      "loss": 0.85094577,
      "num_input_tokens_seen": 4118770,
      "step": 196,
      "time_per_iteration": 2.6900830268859863
    },
    {
      "auxiliary_loss_clip": 0.01366301,
      "auxiliary_loss_mlp": 0.01055583,
      "balance_loss_clip": 1.10309482,
      "balance_loss_mlp": 1.04060459,
      "epoch": 0.02368784945590092,
      "flos": 25193598192000.0,
      "grad_norm": 1.897406913428512,
      "language_loss": 0.84598553,
      "learning_rate": 3.827395544281781e-06,
      "loss": 0.87020445,
      "num_input_tokens_seen": 4141110,
      "step": 197,
      "time_per_iteration": 3.7845380306243896
    },
    {
      "auxiliary_loss_clip": 0.01367016,
      "auxiliary_loss_mlp": 0.0107154,
      "balance_loss_clip": 1.10250759,
      "balance_loss_mlp": 1.05617356,
      "epoch": 0.02380809234654001,
      "flos": 27562481164800.0,
      "grad_norm": 1.9490419038039482,
      "language_loss": 0.79054427,
      "learning_rate": 3.831063632877802e-06,
      "loss": 0.81492984,
      "num_input_tokens_seen": 4161430,
      "step": 198,
      "time_per_iteration": 3.7029826641082764
    },
    {
      "auxiliary_loss_clip": 0.01363238,
      "auxiliary_loss_mlp": 0.01046093,
      "balance_loss_clip": 1.105937,
      "balance_loss_mlp": 1.0323658,
      "epoch": 0.0239283352371791,
      "flos": 18259786540800.0,
      "grad_norm": 2.251082673081909,
      "language_loss": 0.7592361,
      "learning_rate": 3.834713242359712e-06,
      "loss": 0.78332937,
      "num_input_tokens_seen": 4179260,
      "step": 199,
      "time_per_iteration": 2.879580497741699
    },
    {
      "auxiliary_loss_clip": 0.01363877,
      "auxiliary_loss_mlp": 0.01056573,
      "balance_loss_clip": 1.09886312,
      "balance_loss_mlp": 1.04009223,
      "epoch": 0.02404857812781819,
      "flos": 21395110942080.0,
      "grad_norm": 2.0129237518509773,
      "language_loss": 0.87339389,
      "learning_rate": 3.838344557982959e-06,
      "loss": 0.89759839,
      "num_input_tokens_seen": 4200640,
      "step": 200,
      "time_per_iteration": 3.6486713886260986
    },
    {
      "auxiliary_loss_clip": 0.01363367,
      "auxiliary_loss_mlp": 0.01056072,
      "balance_loss_clip": 1.10000587,
      "balance_loss_mlp": 1.04107606,
      "epoch": 0.024168821018457284,
      "flos": 16654256426880.0,
      "grad_norm": 2.559392847768232,
      "language_loss": 0.84745347,
      "learning_rate": 3.841957762231063e-06,
      "loss": 0.87164783,
      "num_input_tokens_seen": 4218170,
      "step": 201,
      "time_per_iteration": 3.7164454460144043
    },
    {
      "auxiliary_loss_clip": 0.01355563,
      "auxiliary_loss_mlp": 0.01057461,
      "balance_loss_clip": 1.09565878,
      "balance_loss_mlp": 1.04397225,
      "epoch": 0.024289063909096374,
      "flos": 22820872464000.0,
      "grad_norm": 1.9222946232583975,
      "language_loss": 0.87612146,
      "learning_rate": 3.8455530348706454e-06,
      "loss": 0.90025163,
      "num_input_tokens_seen": 4237770,
      "step": 202,
      "time_per_iteration": 2.7543201446533203
    },
    {
      "auxiliary_loss_clip": 0.01361388,
      "auxiliary_loss_mlp": 0.01059417,
      "balance_loss_clip": 1.10074067,
      "balance_loss_mlp": 1.04520762,
      "epoch": 0.024409306799735464,
      "flos": 17748598135680.0,
      "grad_norm": 2.0918257377529965,
      "language_loss": 0.77200997,
      "learning_rate": 3.849130553005099e-06,
      "loss": 0.79621804,
      "num_input_tokens_seen": 4255985,
      "step": 203,
      "time_per_iteration": 2.7765746116638184
    },
    {
      "auxiliary_loss_clip": 0.01361517,
      "auxiliary_loss_mlp": 0.01065857,
      "balance_loss_clip": 1.09796476,
      "balance_loss_mlp": 1.05040765,
      "epoch": 0.024529549690374557,
      "flos": 21616213109760.0,
      "grad_norm": 3.4129737789659336,
      "language_loss": 0.83534062,
      "learning_rate": 3.852690491126933e-06,
      "loss": 0.85961437,
      "num_input_tokens_seen": 4276035,
      "step": 204,
      "time_per_iteration": 2.779125213623047
    },
    {
      "auxiliary_loss_clip": 0.01359006,
      "auxiliary_loss_mlp": 0.01050221,
      "balance_loss_clip": 1.09902787,
      "balance_loss_mlp": 1.03427124,
      "epoch": 0.024649792581013647,
      "flos": 25551662918400.0,
      "grad_norm": 2.280573438348697,
      "language_loss": 0.91003799,
      "learning_rate": 3.856233021168845e-06,
      "loss": 0.93413031,
      "num_input_tokens_seen": 4295730,
      "step": 205,
      "time_per_iteration": 2.751268148422241
    },
    {
      "auxiliary_loss_clip": 0.01348472,
      "auxiliary_loss_mlp": 0.0105677,
      "balance_loss_clip": 1.09499216,
      "balance_loss_mlp": 1.04247701,
      "epoch": 0.024770035471652737,
      "flos": 34495574544000.0,
      "grad_norm": 2.2233270794306637,
      "language_loss": 0.90980506,
      "learning_rate": 3.859758312553544e-06,
      "loss": 0.9338575,
      "num_input_tokens_seen": 4317950,
      "step": 206,
      "time_per_iteration": 2.864678382873535
    },
    {
      "auxiliary_loss_clip": 0.01358278,
      "auxiliary_loss_mlp": 0.01061915,
      "balance_loss_clip": 1.09941065,
      "balance_loss_mlp": 1.04607248,
      "epoch": 0.02489027836229183,
      "flos": 21505428587520.0,
      "grad_norm": 3.270535012580431,
      "language_loss": 0.91780508,
      "learning_rate": 3.8632665322423735e-06,
      "loss": 0.94200701,
      "num_input_tokens_seen": 4337605,
      "step": 207,
      "time_per_iteration": 2.7139346599578857
    },
    {
      "auxiliary_loss_clip": 0.01353422,
      "auxiliary_loss_mlp": 0.01048144,
      "balance_loss_clip": 1.09936523,
      "balance_loss_mlp": 1.03333247,
      "epoch": 0.02501052125293092,
      "flos": 23219013790080.0,
      "grad_norm": 1.9731327193872572,
      "language_loss": 0.86004245,
      "learning_rate": 3.866757844782762e-06,
      "loss": 0.88405812,
      "num_input_tokens_seen": 4358110,
      "step": 208,
      "time_per_iteration": 2.740046501159668
    },
    {
      "auxiliary_loss_clip": 0.01359644,
      "auxiliary_loss_mlp": 0.01053752,
      "balance_loss_clip": 1.09965503,
      "balance_loss_mlp": 1.03867221,
      "epoch": 0.02513076414357001,
      "flos": 26388920010240.0,
      "grad_norm": 2.4833604316039075,
      "language_loss": 0.910519,
      "learning_rate": 3.870232412354527e-06,
      "loss": 0.93465304,
      "num_input_tokens_seen": 4374955,
      "step": 209,
      "time_per_iteration": 2.743403673171997
    },
    {
      "auxiliary_loss_clip": 0.01350562,
      "auxiliary_loss_mlp": 0.01058939,
      "balance_loss_clip": 1.09460163,
      "balance_loss_mlp": 1.04330468,
      "epoch": 0.025251007034209103,
      "flos": 13590430047360.0,
      "grad_norm": 2.913138721724304,
      "language_loss": 0.92403734,
      "learning_rate": 3.873690394815086e-06,
      "loss": 0.94813228,
      "num_input_tokens_seen": 4391535,
      "step": 210,
      "time_per_iteration": 2.746443033218384
    },
    {
      "auxiliary_loss_clip": 0.01356681,
      "auxiliary_loss_mlp": 0.0105647,
      "balance_loss_clip": 1.09646106,
      "balance_loss_mlp": 1.04174757,
      "epoch": 0.025371249924848193,
      "flos": 15049229103360.0,
      "grad_norm": 2.3153514320887094,
      "language_loss": 0.91014028,
      "learning_rate": 3.877131949743587e-06,
      "loss": 0.93427175,
      "num_input_tokens_seen": 4408400,
      "step": 211,
      "time_per_iteration": 2.718679904937744
    },
    {
      "auxiliary_loss_clip": 0.0135578,
      "auxiliary_loss_mlp": 0.01060467,
      "balance_loss_clip": 1.09789538,
      "balance_loss_mlp": 1.0440762,
      "epoch": 0.025491492815487283,
      "flos": 25553853648000.0,
      "grad_norm": 2.4770952869765237,
      "language_loss": 0.7833975,
      "learning_rate": 3.880557232483993e-06,
      "loss": 0.80756003,
      "num_input_tokens_seen": 4427840,
      "step": 212,
      "time_per_iteration": 2.7565579414367676
    },
    {
      "auxiliary_loss_clip": 0.01349074,
      "auxiliary_loss_mlp": 0.01056336,
      "balance_loss_clip": 1.09432817,
      "balance_loss_mlp": 1.04065967,
      "epoch": 0.025611735706126376,
      "flos": 20630752502400.0,
      "grad_norm": 2.543593856958179,
      "language_loss": 0.86977309,
      "learning_rate": 3.883966396187164e-06,
      "loss": 0.8938272,
      "num_input_tokens_seen": 4447110,
      "step": 213,
      "time_per_iteration": 2.7266104221343994
    },
    {
      "auxiliary_loss_clip": 0.01354647,
      "auxiliary_loss_mlp": 0.01056647,
      "balance_loss_clip": 1.09811723,
      "balance_loss_mlp": 1.04261649,
      "epoch": 0.025731978596765466,
      "flos": 19062282245760.0,
      "grad_norm": 1.9248190065479132,
      "language_loss": 0.8993811,
      "learning_rate": 3.887359591851937e-06,
      "loss": 0.9234941,
      "num_input_tokens_seen": 4464715,
      "step": 214,
      "time_per_iteration": 2.6999728679656982
    },
    {
      "auxiliary_loss_clip": 0.01348035,
      "auxiliary_loss_mlp": 0.01060083,
      "balance_loss_clip": 1.09617507,
      "balance_loss_mlp": 1.04545605,
      "epoch": 0.025852221487404556,
      "flos": 22163814927360.0,
      "grad_norm": 1.7273725823777355,
      "language_loss": 0.92429721,
      "learning_rate": 3.890736968365265e-06,
      "loss": 0.94837844,
      "num_input_tokens_seen": 4485030,
      "step": 215,
      "time_per_iteration": 2.793107748031616
    },
    {
      "auxiliary_loss_clip": 0.01346832,
      "auxiliary_loss_mlp": 0.01049209,
      "balance_loss_clip": 1.09401822,
      "balance_loss_mlp": 1.03492212,
      "epoch": 0.02597246437804365,
      "flos": 26541971861760.0,
      "grad_norm": 1.9343186633154572,
      "language_loss": 0.85073602,
      "learning_rate": 3.894098672541412e-06,
      "loss": 0.87469649,
      "num_input_tokens_seen": 4505935,
      "step": 216,
      "time_per_iteration": 2.794757843017578
    },
    {
      "auxiliary_loss_clip": 0.01353829,
      "auxiliary_loss_mlp": 0.01058706,
      "balance_loss_clip": 1.09663582,
      "balance_loss_mlp": 1.04401326,
      "epoch": 0.02609270726868274,
      "flos": 32671671696000.0,
      "grad_norm": 1.678053155824288,
      "language_loss": 0.75257039,
      "learning_rate": 3.89744484916025e-06,
      "loss": 0.77669573,
      "num_input_tokens_seen": 4527045,
      "step": 217,
      "time_per_iteration": 2.8602938652038574
    },
    {
      "auxiliary_loss_clip": 0.01359429,
      "auxiliary_loss_mlp": 0.01051557,
      "balance_loss_clip": 1.09836435,
      "balance_loss_mlp": 1.03557134,
      "epoch": 0.02621295015932183,
      "flos": 26243553669120.0,
      "grad_norm": 1.9041915577010193,
      "language_loss": 0.87196612,
      "learning_rate": 3.900775641004673e-06,
      "loss": 0.89607596,
      "num_input_tokens_seen": 4546360,
      "step": 218,
      "time_per_iteration": 2.8233447074890137
    },
    {
      "auxiliary_loss_clip": 0.01369004,
      "auxiliary_loss_mlp": 0.01050181,
      "balance_loss_clip": 1.10509157,
      "balance_loss_mlp": 1.03390884,
      "epoch": 0.026333193049960922,
      "flos": 42921402353280.0,
      "grad_norm": 3.6578004256237713,
      "language_loss": 0.74148369,
      "learning_rate": 3.904091188897156e-06,
      "loss": 0.76567554,
      "num_input_tokens_seen": 4565495,
      "step": 219,
      "time_per_iteration": 2.963517189025879
    },
    {
      "auxiliary_loss_clip": 0.01355625,
      "auxiliary_loss_mlp": 0.01058261,
      "balance_loss_clip": 1.09891236,
      "balance_loss_mlp": 1.04260266,
      "epoch": 0.026453435940600012,
      "flos": 17963846386560.0,
      "grad_norm": 2.148971543186778,
      "language_loss": 0.81864071,
      "learning_rate": 3.90739163173548e-06,
      "loss": 0.84277952,
      "num_input_tokens_seen": 4583330,
      "step": 220,
      "time_per_iteration": 2.738960027694702
    },
    {
      "auxiliary_loss_clip": 0.01347726,
      "auxiliary_loss_mlp": 0.01068348,
      "balance_loss_clip": 1.09313965,
      "balance_loss_mlp": 1.05348277,
      "epoch": 0.026573678831239102,
      "flos": 18984319776000.0,
      "grad_norm": 2.653144461308966,
      "language_loss": 0.88172376,
      "learning_rate": 3.910677106527646e-06,
      "loss": 0.9058845,
      "num_input_tokens_seen": 4600520,
      "step": 221,
      "time_per_iteration": 2.684863805770874
    },
    {
      "auxiliary_loss_clip": 0.01345338,
      "auxiliary_loss_mlp": 0.01062846,
      "balance_loss_clip": 1.09442759,
      "balance_loss_mlp": 1.04808807,
      "epoch": 0.026693921721878195,
      "flos": 29241448634880.0,
      "grad_norm": 2.2241472143852152,
      "language_loss": 0.84362555,
      "learning_rate": 3.913947748426004e-06,
      "loss": 0.86770737,
      "num_input_tokens_seen": 4617340,
      "step": 222,
      "time_per_iteration": 2.8048698902130127
    },
    {
      "auxiliary_loss_clip": 0.01352753,
      "auxiliary_loss_mlp": 0.01062436,
      "balance_loss_clip": 1.09823895,
      "balance_loss_mlp": 1.04738033,
      "epoch": 0.026814164612517285,
      "flos": 14128083797760.0,
      "grad_norm": 3.3134504059770604,
      "language_loss": 0.7641083,
      "learning_rate": 3.9172036907606136e-06,
      "loss": 0.78826022,
      "num_input_tokens_seen": 4630820,
      "step": 223,
      "time_per_iteration": 3.7074942588806152
    },
    {
      "auxiliary_loss_clip": 0.01343368,
      "auxiliary_loss_mlp": 0.01057805,
      "balance_loss_clip": 1.09160352,
      "balance_loss_mlp": 1.04335666,
      "epoch": 0.026934407503156375,
      "flos": 23511973115520.0,
      "grad_norm": 2.033041144490643,
      "language_loss": 0.94969189,
      "learning_rate": 3.920445065071855e-06,
      "loss": 0.97370368,
      "num_input_tokens_seen": 4651985,
      "step": 224,
      "time_per_iteration": 3.8566625118255615
    },
    {
      "auxiliary_loss_clip": 0.01347478,
      "auxiliary_loss_mlp": 0.01053151,
      "balance_loss_clip": 1.09408665,
      "balance_loss_mlp": 1.03885818,
      "epoch": 0.027054650393795468,
      "flos": 28950356816640.0,
      "grad_norm": 2.1950200439875767,
      "language_loss": 0.80038989,
      "learning_rate": 3.923672001142322e-06,
      "loss": 0.82439619,
      "num_input_tokens_seen": 4672295,
      "step": 225,
      "time_per_iteration": 3.866769552230835
    },
    {
      "auxiliary_loss_clip": 0.01342047,
      "auxiliary_loss_mlp": 0.01055638,
      "balance_loss_clip": 1.09187973,
      "balance_loss_mlp": 1.04016423,
      "epoch": 0.027174893284434558,
      "flos": 31431568596480.0,
      "grad_norm": 1.8053170322484289,
      "language_loss": 0.84419972,
      "learning_rate": 3.926884627027996e-06,
      "loss": 0.86817658,
      "num_input_tokens_seen": 4696065,
      "step": 226,
      "time_per_iteration": 2.932894468307495
    },
    {
      "auxiliary_loss_clip": 0.01342689,
      "auxiliary_loss_mlp": 0.01056637,
      "balance_loss_clip": 1.09350777,
      "balance_loss_mlp": 1.04120564,
      "epoch": 0.027295136175073648,
      "flos": 22054466949120.0,
      "grad_norm": 1.9661839636262635,
      "language_loss": 0.77484894,
      "learning_rate": 3.930083069088744e-06,
      "loss": 0.79884219,
      "num_input_tokens_seen": 4716065,
      "step": 227,
      "time_per_iteration": 3.807560920715332
    },
    {
      "auxiliary_loss_clip": 0.01336388,
      "auxiliary_loss_mlp": 0.01007908,
      "balance_loss_clip": 1.14233971,
      "balance_loss_mlp": 1.00128031,
      "epoch": 0.02741537906571274,
      "flos": 60800752972800.0,
      "grad_norm": 0.9858299995306204,
      "language_loss": 0.59343934,
      "learning_rate": 3.933267452018137e-06,
      "loss": 0.61688232,
      "num_input_tokens_seen": 4775860,
      "step": 228,
      "time_per_iteration": 3.3442225456237793
    },
    {
      "auxiliary_loss_clip": 0.01345798,
      "auxiliary_loss_mlp": 0.01062389,
      "balance_loss_clip": 1.0962317,
      "balance_loss_mlp": 1.04754138,
      "epoch": 0.02753562195635183,
      "flos": 24606278910720.0,
      "grad_norm": 2.0198896477153974,
      "language_loss": 0.84197628,
      "learning_rate": 3.936437898872622e-06,
      "loss": 0.86605811,
      "num_input_tokens_seen": 4795835,
      "step": 229,
      "time_per_iteration": 2.8526601791381836
    },
    {
      "auxiliary_loss_clip": 0.01342009,
      "auxiliary_loss_mlp": 0.01048271,
      "balance_loss_clip": 1.09204793,
      "balance_loss_mlp": 1.03308332,
      "epoch": 0.02765586484699092,
      "flos": 34094236907520.0,
      "grad_norm": 2.6489710160430007,
      "language_loss": 0.79394042,
      "learning_rate": 3.9395945311000525e-06,
      "loss": 0.81784332,
      "num_input_tokens_seen": 4817460,
      "step": 230,
      "time_per_iteration": 2.8598990440368652
    },
    {
      "auxiliary_loss_clip": 0.01345253,
      "auxiliary_loss_mlp": 0.01061856,
      "balance_loss_clip": 1.09289682,
      "balance_loss_mlp": 1.04814732,
      "epoch": 0.027776107737630014,
      "flos": 14829922615680.0,
      "grad_norm": 2.4653966623615897,
      "language_loss": 0.91016328,
      "learning_rate": 3.942737468567608e-06,
      "loss": 0.93423438,
      "num_input_tokens_seen": 4835475,
      "step": 231,
      "time_per_iteration": 2.9433236122131348
    },
    {
      "auxiliary_loss_clip": 0.01345698,
      "auxiliary_loss_mlp": 0.01051994,
      "balance_loss_clip": 1.09421277,
      "balance_loss_mlp": 1.03728414,
      "epoch": 0.027896350628269104,
      "flos": 47920347066240.0,
      "grad_norm": 1.9904979373677474,
      "language_loss": 0.85963786,
      "learning_rate": 3.9458668295891026e-06,
      "loss": 0.88361472,
      "num_input_tokens_seen": 4857760,
      "step": 232,
      "time_per_iteration": 3.049485921859741
    },
    {
      "auxiliary_loss_clip": 0.01344262,
      "auxiliary_loss_mlp": 0.01057487,
      "balance_loss_clip": 1.09264684,
      "balance_loss_mlp": 1.04261613,
      "epoch": 0.028016593518908194,
      "flos": 21684550734720.0,
      "grad_norm": 3.116153035655552,
      "language_loss": 0.86829925,
      "learning_rate": 3.948982730951712e-06,
      "loss": 0.89231676,
      "num_input_tokens_seen": 4875855,
      "step": 233,
      "time_per_iteration": 2.804206609725952
    },
    {
      "auxiliary_loss_clip": 0.01351143,
      "auxiliary_loss_mlp": 0.0105262,
      "balance_loss_clip": 1.09483147,
      "balance_loss_mlp": 1.03681266,
      "epoch": 0.028136836409547287,
      "flos": 18439483305600.0,
      "grad_norm": 2.3656718827734107,
      "language_loss": 0.81888855,
      "learning_rate": 3.9520852879421254e-06,
      "loss": 0.84292614,
      "num_input_tokens_seen": 4893200,
      "step": 234,
      "time_per_iteration": 2.791130542755127
    },
    {
      "auxiliary_loss_clip": 0.01341275,
      "auxiliary_loss_mlp": 0.0104547,
      "balance_loss_clip": 1.09475338,
      "balance_loss_mlp": 1.03104544,
      "epoch": 0.028257079300186377,
      "flos": 31576934937600.0,
      "grad_norm": 3.701227848067122,
      "language_loss": 0.8151238,
      "learning_rate": 3.955174614372137e-06,
      "loss": 0.83899128,
      "num_input_tokens_seen": 4912965,
      "step": 235,
      "time_per_iteration": 2.838330030441284
    },
    {
      "auxiliary_loss_clip": 0.01345919,
      "auxiliary_loss_mlp": 0.01059904,
      "balance_loss_clip": 1.09593964,
      "balance_loss_mlp": 1.04501462,
      "epoch": 0.028377322190825467,
      "flos": 23513337832320.0,
      "grad_norm": 2.452374591253405,
      "language_loss": 0.84526265,
      "learning_rate": 3.9582508226037045e-06,
      "loss": 0.86932087,
      "num_input_tokens_seen": 4933105,
      "step": 236,
      "time_per_iteration": 2.8852930068969727
    },
    {
      "auxiliary_loss_clip": 0.01348392,
      "auxiliary_loss_mlp": 0.01055256,
      "balance_loss_clip": 1.09231043,
      "balance_loss_mlp": 1.04010487,
      "epoch": 0.02849756508146456,
      "flos": 20479604071680.0,
      "grad_norm": 2.3549950552687213,
      "language_loss": 0.94107628,
      "learning_rate": 3.9613140235734636e-06,
      "loss": 0.96511281,
      "num_input_tokens_seen": 4950085,
      "step": 237,
      "time_per_iteration": 2.8479321002960205
    },
    {
      "auxiliary_loss_clip": 0.01346181,
      "auxiliary_loss_mlp": 0.01054971,
      "balance_loss_clip": 1.09455585,
      "balance_loss_mlp": 1.04008746,
      "epoch": 0.02861780797210365,
      "flos": 14283362292480.0,
      "grad_norm": 1.9104713524299644,
      "language_loss": 0.81138903,
      "learning_rate": 3.96436432681674e-06,
      "loss": 0.83540058,
      "num_input_tokens_seen": 4968075,
      "step": 238,
      "time_per_iteration": 2.7985689640045166
    },
    {
      "auxiliary_loss_clip": 0.01348065,
      "auxiliary_loss_mlp": 0.01062854,
      "balance_loss_clip": 1.09623289,
      "balance_loss_mlp": 1.04789901,
      "epoch": 0.02873805086274274,
      "flos": 25808532053760.0,
      "grad_norm": 3.5182623419634633,
      "language_loss": 0.89101464,
      "learning_rate": 3.967401840491044e-06,
      "loss": 0.91512382,
      "num_input_tokens_seen": 4987355,
      "step": 239,
      "time_per_iteration": 2.8351426124572754
    },
    {
      "auxiliary_loss_clip": 0.01334451,
      "auxiliary_loss_mlp": 0.01063471,
      "balance_loss_clip": 1.08902168,
      "balance_loss_mlp": 1.04949927,
      "epoch": 0.028858293753381833,
      "flos": 17304238984320.0,
      "grad_norm": 2.4103461983579773,
      "language_loss": 0.87659043,
      "learning_rate": 3.97042667139909e-06,
      "loss": 0.90056968,
      "num_input_tokens_seen": 5004680,
      "step": 240,
      "time_per_iteration": 2.8291516304016113
    },
    {
      "auxiliary_loss_clip": 0.01339646,
      "auxiliary_loss_mlp": 0.01057946,
      "balance_loss_clip": 1.09262002,
      "balance_loss_mlp": 1.04256821,
      "epoch": 0.028978536644020923,
      "flos": 23038347358080.0,
      "grad_norm": 1.9990557897503582,
      "language_loss": 0.87514293,
      "learning_rate": 3.973438925011327e-06,
      "loss": 0.8991189,
      "num_input_tokens_seen": 5022965,
      "step": 241,
      "time_per_iteration": 2.706932783126831
    },
    {
      "auxiliary_loss_clip": 0.013398,
      "auxiliary_loss_mlp": 0.01056693,
      "balance_loss_clip": 1.09149015,
      "balance_loss_mlp": 1.0406239,
      "epoch": 0.029098779534660012,
      "flos": 28329712692480.0,
      "grad_norm": 2.171833556622135,
      "language_loss": 0.91297573,
      "learning_rate": 3.976438705488002e-06,
      "loss": 0.93694073,
      "num_input_tokens_seen": 5042625,
      "step": 242,
      "time_per_iteration": 2.76932692527771
    },
    {
      "auxiliary_loss_clip": 0.01339514,
      "auxiliary_loss_mlp": 0.01051107,
      "balance_loss_clip": 1.09307182,
      "balance_loss_mlp": 1.03650403,
      "epoch": 0.029219022425299106,
      "flos": 13881665520000.0,
      "grad_norm": 2.3707959982132594,
      "language_loss": 0.93041319,
      "learning_rate": 3.9794261157007744e-06,
      "loss": 0.95431936,
      "num_input_tokens_seen": 5060380,
      "step": 243,
      "time_per_iteration": 2.724710464477539
    },
    {
      "auxiliary_loss_clip": 0.0135628,
      "auxiliary_loss_mlp": 0.01071932,
      "balance_loss_clip": 1.09969532,
      "balance_loss_mlp": 1.05542135,
      "epoch": 0.029339265315938196,
      "flos": 19422501788160.0,
      "grad_norm": 2.163669360193518,
      "language_loss": 0.84865451,
      "learning_rate": 3.982401257253887e-06,
      "loss": 0.87293661,
      "num_input_tokens_seen": 5078720,
      "step": 244,
      "time_per_iteration": 2.73980712890625
    },
    {
      "auxiliary_loss_clip": 0.01334811,
      "auxiliary_loss_mlp": 0.01064634,
      "balance_loss_clip": 1.08953285,
      "balance_loss_mlp": 1.0494113,
      "epoch": 0.029459508206577285,
      "flos": 15669550005120.0,
      "grad_norm": 2.5320201674335703,
      "language_loss": 0.89676952,
      "learning_rate": 3.985364230504893e-06,
      "loss": 0.92076397,
      "num_input_tokens_seen": 5096605,
      "step": 245,
      "time_per_iteration": 2.717994451522827
    },
    {
      "auxiliary_loss_clip": 0.01344071,
      "auxiliary_loss_mlp": 0.01053984,
      "balance_loss_clip": 1.09444046,
      "balance_loss_mlp": 1.03953552,
      "epoch": 0.02957975109721638,
      "flos": 28220975245440.0,
      "grad_norm": 2.6332482719181103,
      "language_loss": 0.8433398,
      "learning_rate": 3.988315134584976e-06,
      "loss": 0.8673203,
      "num_input_tokens_seen": 5116285,
      "step": 246,
      "time_per_iteration": 2.7317516803741455
    },
    {
      "auxiliary_loss_clip": 0.01340057,
      "auxiliary_loss_mlp": 0.01062563,
      "balance_loss_clip": 1.09108543,
      "balance_loss_mlp": 1.04594564,
      "epoch": 0.02969999398785547,
      "flos": 24315869450880.0,
      "grad_norm": 1.7311477770830181,
      "language_loss": 0.80111498,
      "learning_rate": 3.991254067418851e-06,
      "loss": 0.82514119,
      "num_input_tokens_seen": 5136825,
      "step": 247,
      "time_per_iteration": 2.7159109115600586
    },
    {
      "auxiliary_loss_clip": 0.01338209,
      "auxiliary_loss_mlp": 0.01051928,
      "balance_loss_clip": 1.09374213,
      "balance_loss_mlp": 1.03732514,
      "epoch": 0.02982023687849456,
      "flos": 35078584193280.0,
      "grad_norm": 1.9763612333899638,
      "language_loss": 0.82641578,
      "learning_rate": 3.994181125744254e-06,
      "loss": 0.85031712,
      "num_input_tokens_seen": 5158630,
      "step": 248,
      "time_per_iteration": 2.8393144607543945
    },
    {
      "auxiliary_loss_clip": 0.01333599,
      "auxiliary_loss_mlp": 0.01050241,
      "balance_loss_clip": 1.09050083,
      "balance_loss_mlp": 1.03536332,
      "epoch": 0.02994047976913365,
      "flos": 26177155378560.0,
      "grad_norm": 2.2976624677506665,
      "language_loss": 0.7388373,
      "learning_rate": 3.99709640513106e-06,
      "loss": 0.76267564,
      "num_input_tokens_seen": 5179510,
      "step": 249,
      "time_per_iteration": 2.735832452774048
    },
    {
      "auxiliary_loss_clip": 0.013442,
      "auxiliary_loss_mlp": 0.01054715,
      "balance_loss_clip": 1.09295702,
      "balance_loss_mlp": 1.03809714,
      "epoch": 0.03006072265977274,
      "flos": 25625028447360.0,
      "grad_norm": 3.1908854775199207,
      "language_loss": 0.85586947,
      "learning_rate": 4e-06,
      "loss": 0.87985861,
      "num_input_tokens_seen": 5199345,
      "step": 250,
      "time_per_iteration": 4.628480672836304
    },
    {
      "auxiliary_loss_clip": 0.01335833,
      "auxiliary_loss_mlp": 0.01068095,
      "balance_loss_clip": 1.09261644,
      "balance_loss_mlp": 1.05326581,
      "epoch": 0.03018096555041183,
      "flos": 22127078292480.0,
      "grad_norm": 10.027365180336401,
      "language_loss": 0.8845253,
      "learning_rate": 3.999999848300794e-06,
      "loss": 0.90856457,
      "num_input_tokens_seen": 5218330,
      "step": 251,
      "time_per_iteration": 2.6487646102905273
    },
    {
      "auxiliary_loss_clip": 0.01330894,
      "auxiliary_loss_mlp": 0.01045217,
      "balance_loss_clip": 1.08842683,
      "balance_loss_mlp": 1.03210974,
      "epoch": 0.030301208441050925,
      "flos": 30188197359360.0,
      "grad_norm": 1.6711365845553159,
      "language_loss": 0.8905766,
      "learning_rate": 3.999999393203203e-06,
      "loss": 0.91433764,
      "num_input_tokens_seen": 5240740,
      "step": 252,
      "time_per_iteration": 3.7815980911254883
    },
    {
      "auxiliary_loss_clip": 0.01334164,
      "auxiliary_loss_mlp": 0.01054722,
      "balance_loss_clip": 1.08876538,
      "balance_loss_mlp": 1.04112577,
      "epoch": 0.030421451331690014,
      "flos": 23621392920960.0,
      "grad_norm": 2.4781605844193653,
      "language_loss": 0.84854364,
      "learning_rate": 3.999998634707293e-06,
      "loss": 0.87243247,
      "num_input_tokens_seen": 5260290,
      "step": 253,
      "time_per_iteration": 2.778583526611328
    },
    {
      "auxiliary_loss_clip": 0.01344991,
      "auxiliary_loss_mlp": 0.01066708,
      "balance_loss_clip": 1.09341764,
      "balance_loss_mlp": 1.05171776,
      "epoch": 0.030541694222329104,
      "flos": 27928446883200.0,
      "grad_norm": 2.214662109956874,
      "language_loss": 0.96184695,
      "learning_rate": 3.999997572813182e-06,
      "loss": 0.98596394,
      "num_input_tokens_seen": 5278100,
      "step": 254,
      "time_per_iteration": 3.7728610038757324
    },
    {
      "auxiliary_loss_clip": 0.01333185,
      "auxiliary_loss_mlp": 0.01060658,
      "balance_loss_clip": 1.09071434,
      "balance_loss_mlp": 1.04587579,
      "epoch": 0.030661937112968194,
      "flos": 18588441006720.0,
      "grad_norm": 1.9789267813380245,
      "language_loss": 0.87527734,
      "learning_rate": 3.999996207521028e-06,
      "loss": 0.89921576,
      "num_input_tokens_seen": 5296810,
      "step": 255,
      "time_per_iteration": 2.7499802112579346
    },
    {
      "auxiliary_loss_clip": 0.01338081,
      "auxiliary_loss_mlp": 0.01052864,
      "balance_loss_clip": 1.09033275,
      "balance_loss_mlp": 1.03720558,
      "epoch": 0.030782180003607287,
      "flos": 12969139478400.0,
      "grad_norm": 2.5403706987980423,
      "language_loss": 0.81981158,
      "learning_rate": 3.999994538831039e-06,
      "loss": 0.84372091,
      "num_input_tokens_seen": 5313395,
      "step": 256,
      "time_per_iteration": 2.7065606117248535
    },
    {
      "auxiliary_loss_clip": 0.01335236,
      "auxiliary_loss_mlp": 0.01058021,
      "balance_loss_clip": 1.0887835,
      "balance_loss_mlp": 1.04338169,
      "epoch": 0.030902422894246377,
      "flos": 23335364920320.0,
      "grad_norm": 2.5300696418310635,
      "language_loss": 0.85618865,
      "learning_rate": 3.99999256674347e-06,
      "loss": 0.88012123,
      "num_input_tokens_seen": 5333545,
      "step": 257,
      "time_per_iteration": 2.7716550827026367
    },
    {
      "auxiliary_loss_clip": 0.01320813,
      "auxiliary_loss_mlp": 0.01008325,
      "balance_loss_clip": 1.13664043,
      "balance_loss_mlp": 1.00222194,
      "epoch": 0.031022665784885467,
      "flos": 55094151438720.0,
      "grad_norm": 1.0131449418675342,
      "language_loss": 0.53498912,
      "learning_rate": 3.999990291258618e-06,
      "loss": 0.55828059,
      "num_input_tokens_seen": 5392235,
      "step": 258,
      "time_per_iteration": 3.2839112281799316
    },
    {
      "auxiliary_loss_clip": 0.01335585,
      "auxiliary_loss_mlp": 0.01054884,
      "balance_loss_clip": 1.09051704,
      "balance_loss_mlp": 1.03951836,
      "epoch": 0.03114290867552456,
      "flos": 19317786664320.0,
      "grad_norm": 2.260949000352887,
      "language_loss": 0.86876005,
      "learning_rate": 3.999987712376829e-06,
      "loss": 0.89266479,
      "num_input_tokens_seen": 5410555,
      "step": 259,
      "time_per_iteration": 2.7221620082855225
    },
    {
      "auxiliary_loss_clip": 0.01335568,
      "auxiliary_loss_mlp": 0.01056645,
      "balance_loss_clip": 1.09074676,
      "balance_loss_mlp": 1.0416069,
      "epoch": 0.031263151566163654,
      "flos": 20959442881920.0,
      "grad_norm": 1.9522031956883243,
      "language_loss": 0.81970823,
      "learning_rate": 3.999984830098494e-06,
      "loss": 0.84363037,
      "num_input_tokens_seen": 5430135,
      "step": 260,
      "time_per_iteration": 2.673962354660034
    },
    {
      "auxiliary_loss_clip": 0.01335023,
      "auxiliary_loss_mlp": 0.0105813,
      "balance_loss_clip": 1.08920598,
      "balance_loss_mlp": 1.04233432,
      "epoch": 0.03138339445680274,
      "flos": 14793006412800.0,
      "grad_norm": 2.8946892708997063,
      "language_loss": 0.97895324,
      "learning_rate": 3.999981644424051e-06,
      "loss": 1.00288475,
      "num_input_tokens_seen": 5444935,
      "step": 261,
      "time_per_iteration": 2.600484609603882
    },
    {
      "auxiliary_loss_clip": 0.01338459,
      "auxiliary_loss_mlp": 0.01071632,
      "balance_loss_clip": 1.09132063,
      "balance_loss_mlp": 1.05516863,
      "epoch": 0.03150363734744183,
      "flos": 11655599022720.0,
      "grad_norm": 2.2966120521314135,
      "language_loss": 0.85973322,
      "learning_rate": 3.999978155353982e-06,
      "loss": 0.88383406,
      "num_input_tokens_seen": 5462080,
      "step": 262,
      "time_per_iteration": 2.656310796737671
    },
    {
      "auxiliary_loss_clip": 0.01331543,
      "auxiliary_loss_mlp": 0.01049125,
      "balance_loss_clip": 1.08633566,
      "balance_loss_mlp": 1.03456974,
      "epoch": 0.03162388023808092,
      "flos": 33727732485120.0,
      "grad_norm": 2.1825309326403435,
      "language_loss": 0.80292708,
      "learning_rate": 3.9999743628888186e-06,
      "loss": 0.82673383,
      "num_input_tokens_seen": 5483870,
      "step": 263,
      "time_per_iteration": 2.7442522048950195
    },
    {
      "auxiliary_loss_clip": 0.01337745,
      "auxiliary_loss_mlp": 0.01058261,
      "balance_loss_clip": 1.08931971,
      "balance_loss_mlp": 1.04329467,
      "epoch": 0.03174412312872001,
      "flos": 20810952057600.0,
      "grad_norm": 2.2143672036215807,
      "language_loss": 0.89498675,
      "learning_rate": 3.999970267029133e-06,
      "loss": 0.91894686,
      "num_input_tokens_seen": 5502830,
      "step": 264,
      "time_per_iteration": 2.7064802646636963
    },
    {
      "auxiliary_loss_clip": 0.01334145,
      "auxiliary_loss_mlp": 0.01059202,
      "balance_loss_clip": 1.09049654,
      "balance_loss_mlp": 1.04326367,
      "epoch": 0.0318643660193591,
      "flos": 23727939638400.0,
      "grad_norm": 1.7877398507581546,
      "language_loss": 0.80008018,
      "learning_rate": 3.999965867775548e-06,
      "loss": 0.82401371,
      "num_input_tokens_seen": 5523225,
      "step": 265,
      "time_per_iteration": 2.750957489013672
    },
    {
      "auxiliary_loss_clip": 0.01334308,
      "auxiliary_loss_mlp": 0.01055298,
      "balance_loss_clip": 1.08746576,
      "balance_loss_mlp": 1.04098654,
      "epoch": 0.0319846089099982,
      "flos": 13917863450880.0,
      "grad_norm": 2.6866332283993573,
      "language_loss": 0.86840475,
      "learning_rate": 3.9999611651287315e-06,
      "loss": 0.89230078,
      "num_input_tokens_seen": 5541380,
      "step": 266,
      "time_per_iteration": 2.644536256790161
    },
    {
      "auxiliary_loss_clip": 0.01335217,
      "auxiliary_loss_mlp": 0.01055623,
      "balance_loss_clip": 1.089903,
      "balance_loss_mlp": 1.03951752,
      "epoch": 0.03210485180063729,
      "flos": 14753253035520.0,
      "grad_norm": 2.9488747141350573,
      "language_loss": 0.78737748,
      "learning_rate": 3.999956159089396e-06,
      "loss": 0.81128585,
      "num_input_tokens_seen": 5558830,
      "step": 267,
      "time_per_iteration": 2.606074810028076
    },
    {
      "auxiliary_loss_clip": 0.01332425,
      "auxiliary_loss_mlp": 0.01053633,
      "balance_loss_clip": 1.09030938,
      "balance_loss_mlp": 1.03910708,
      "epoch": 0.03222509469127638,
      "flos": 28913153304960.0,
      "grad_norm": 2.349967558920265,
      "language_loss": 0.79592538,
      "learning_rate": 3.999950849658302e-06,
      "loss": 0.81978595,
      "num_input_tokens_seen": 5577750,
      "step": 268,
      "time_per_iteration": 2.7208755016326904
    },
    {
      "auxiliary_loss_clip": 0.01335547,
      "auxiliary_loss_mlp": 0.01062284,
      "balance_loss_clip": 1.0900259,
      "balance_loss_mlp": 1.04632187,
      "epoch": 0.03234533758191547,
      "flos": 16946389739520.0,
      "grad_norm": 2.1479788111965794,
      "language_loss": 0.84174955,
      "learning_rate": 3.999945236836254e-06,
      "loss": 0.8657279,
      "num_input_tokens_seen": 5596715,
      "step": 269,
      "time_per_iteration": 2.6877505779266357
    },
    {
      "auxiliary_loss_clip": 0.01344785,
      "auxiliary_loss_mlp": 0.01063695,
      "balance_loss_clip": 1.09565926,
      "balance_loss_mlp": 1.04726839,
      "epoch": 0.03246558047255456,
      "flos": 18989096284800.0,
      "grad_norm": 2.653849097308841,
      "language_loss": 0.9457196,
      "learning_rate": 3.999939320624103e-06,
      "loss": 0.96980447,
      "num_input_tokens_seen": 5611865,
      "step": 270,
      "time_per_iteration": 2.7283596992492676
    },
    {
      "auxiliary_loss_clip": 0.0133801,
      "auxiliary_loss_mlp": 0.01065889,
      "balance_loss_clip": 1.09172523,
      "balance_loss_mlp": 1.05111909,
      "epoch": 0.03258582336319365,
      "flos": 23728334688000.0,
      "grad_norm": 2.895736350791968,
      "language_loss": 0.89780611,
      "learning_rate": 3.999933101022749e-06,
      "loss": 0.92184508,
      "num_input_tokens_seen": 5632270,
      "step": 271,
      "time_per_iteration": 2.7464873790740967
    },
    {
      "auxiliary_loss_clip": 0.01337948,
      "auxiliary_loss_mlp": 0.01064164,
      "balance_loss_clip": 1.0917455,
      "balance_loss_mlp": 1.04910803,
      "epoch": 0.032706066253832745,
      "flos": 27670823562240.0,
      "grad_norm": 1.889998856525756,
      "language_loss": 0.86772782,
      "learning_rate": 3.999926578033132e-06,
      "loss": 0.89174891,
      "num_input_tokens_seen": 5652085,
      "step": 272,
      "time_per_iteration": 2.8822593688964844
    },
    {
      "auxiliary_loss_clip": 0.01341405,
      "auxiliary_loss_mlp": 0.01058323,
      "balance_loss_clip": 1.08839989,
      "balance_loss_mlp": 1.04155004,
      "epoch": 0.032826309144471835,
      "flos": 45624685968000.0,
      "grad_norm": 2.0175636067862412,
      "language_loss": 0.63195097,
      "learning_rate": 3.999919751656244e-06,
      "loss": 0.65594828,
      "num_input_tokens_seen": 5678985,
      "step": 273,
      "time_per_iteration": 2.9773874282836914
    },
    {
      "auxiliary_loss_clip": 0.01337581,
      "auxiliary_loss_mlp": 0.01062239,
      "balance_loss_clip": 1.09088624,
      "balance_loss_mlp": 1.04705191,
      "epoch": 0.032946552035110925,
      "flos": 25812374808960.0,
      "grad_norm": 2.6077895063632854,
      "language_loss": 0.75929302,
      "learning_rate": 3.9999126218931195e-06,
      "loss": 0.78329122,
      "num_input_tokens_seen": 5697020,
      "step": 274,
      "time_per_iteration": 2.7138047218322754
    },
    {
      "auxiliary_loss_clip": 0.0133854,
      "auxiliary_loss_mlp": 0.01058599,
      "balance_loss_clip": 1.0921663,
      "balance_loss_mlp": 1.04392481,
      "epoch": 0.033066794925750015,
      "flos": 15121984101120.0,
      "grad_norm": 2.2820656061607694,
      "language_loss": 0.89460123,
      "learning_rate": 3.99990518874484e-06,
      "loss": 0.91857255,
      "num_input_tokens_seen": 5713460,
      "step": 275,
      "time_per_iteration": 2.733167886734009
    },
    {
      "auxiliary_loss_clip": 0.01337959,
      "auxiliary_loss_mlp": 0.01055851,
      "balance_loss_clip": 1.09284866,
      "balance_loss_mlp": 1.03985357,
      "epoch": 0.033187037816389105,
      "flos": 22776593973120.0,
      "grad_norm": 2.3153532887415658,
      "language_loss": 0.92714316,
      "learning_rate": 3.999897452212534e-06,
      "loss": 0.95108116,
      "num_input_tokens_seen": 5730790,
      "step": 276,
      "time_per_iteration": 3.619053363800049
    },
    {
      "auxiliary_loss_clip": 0.01331471,
      "auxiliary_loss_mlp": 0.01066662,
      "balance_loss_clip": 1.08743811,
      "balance_loss_mlp": 1.04935241,
      "epoch": 0.033307280707028195,
      "flos": 23331414424320.0,
      "grad_norm": 2.0080440573723473,
      "language_loss": 1.0015732,
      "learning_rate": 3.999889412297374e-06,
      "loss": 1.02555454,
      "num_input_tokens_seen": 5750215,
      "step": 277,
      "time_per_iteration": 3.6231579780578613
    },
    {
      "auxiliary_loss_clip": 0.01330888,
      "auxiliary_loss_mlp": 0.01053405,
      "balance_loss_clip": 1.08723164,
      "balance_loss_mlp": 1.03712142,
      "epoch": 0.03342752359766729,
      "flos": 28840290566400.0,
      "grad_norm": 2.1740102521813873,
      "language_loss": 0.78921986,
      "learning_rate": 3.999881069000581e-06,
      "loss": 0.81306279,
      "num_input_tokens_seen": 5769945,
      "step": 278,
      "time_per_iteration": 3.650899648666382
    },
    {
      "auxiliary_loss_clip": 0.01335989,
      "auxiliary_loss_mlp": 0.01063159,
      "balance_loss_clip": 1.08828437,
      "balance_loss_mlp": 1.04736328,
      "epoch": 0.03354776648830638,
      "flos": 19384544090880.0,
      "grad_norm": 2.3948982271436643,
      "language_loss": 0.86758232,
      "learning_rate": 3.99987242232342e-06,
      "loss": 0.89157379,
      "num_input_tokens_seen": 5784950,
      "step": 279,
      "time_per_iteration": 2.6596972942352295
    },
    {
      "auxiliary_loss_clip": 0.01337941,
      "auxiliary_loss_mlp": 0.01060439,
      "balance_loss_clip": 1.09350348,
      "balance_loss_mlp": 1.04450119,
      "epoch": 0.03366800937894547,
      "flos": 17858628472320.0,
      "grad_norm": 1.9144071175893862,
      "language_loss": 0.79588437,
      "learning_rate": 3.9998634722672026e-06,
      "loss": 0.81986815,
      "num_input_tokens_seen": 5805005,
      "step": 280,
      "time_per_iteration": 3.561228036880493
    },
    {
      "auxiliary_loss_clip": 0.0134066,
      "auxiliary_loss_mlp": 0.0107078,
      "balance_loss_clip": 1.09462404,
      "balance_loss_mlp": 1.05518746,
      "epoch": 0.03378825226958456,
      "flos": 35951033635200.0,
      "grad_norm": 2.2196885891662395,
      "language_loss": 0.78491569,
      "learning_rate": 3.999854218833286e-06,
      "loss": 0.80903006,
      "num_input_tokens_seen": 5825825,
      "step": 281,
      "time_per_iteration": 2.767331123352051
    },
    {
      "auxiliary_loss_clip": 0.01333864,
      "auxiliary_loss_mlp": 0.01049074,
      "balance_loss_clip": 1.0913558,
      "balance_loss_mlp": 1.03281331,
      "epoch": 0.03390849516022365,
      "flos": 25702488126720.0,
      "grad_norm": 1.9929799604455207,
      "language_loss": 0.81851161,
      "learning_rate": 3.999844662023075e-06,
      "loss": 0.84234101,
      "num_input_tokens_seen": 5845700,
      "step": 282,
      "time_per_iteration": 2.68393874168396
    },
    {
      "auxiliary_loss_clip": 0.01325199,
      "auxiliary_loss_mlp": 0.01049912,
      "balance_loss_clip": 1.08729422,
      "balance_loss_mlp": 1.0363574,
      "epoch": 0.03402873805086274,
      "flos": 21284505987840.0,
      "grad_norm": 1.8436334062901851,
      "language_loss": 0.92173672,
      "learning_rate": 3.999834801838018e-06,
      "loss": 0.9454878,
      "num_input_tokens_seen": 5864680,
      "step": 283,
      "time_per_iteration": 2.6988792419433594
    },
    {
      "auxiliary_loss_clip": 0.01329179,
      "auxiliary_loss_mlp": 0.01059004,
      "balance_loss_clip": 1.08808446,
      "balance_loss_mlp": 1.04387057,
      "epoch": 0.03414898094150183,
      "flos": 22710913954560.0,
      "grad_norm": 1.9337302776827028,
      "language_loss": 0.73931074,
      "learning_rate": 3.9998246382796115e-06,
      "loss": 0.76319265,
      "num_input_tokens_seen": 5884260,
      "step": 284,
      "time_per_iteration": 2.671165704727173
    },
    {
      "auxiliary_loss_clip": 0.01334267,
      "auxiliary_loss_mlp": 0.01063773,
      "balance_loss_clip": 1.08848119,
      "balance_loss_mlp": 1.04821646,
      "epoch": 0.03426922383214093,
      "flos": 18879927874560.0,
      "grad_norm": 2.0642202228814317,
      "language_loss": 0.90554523,
      "learning_rate": 3.999814171349399e-06,
      "loss": 0.92952561,
      "num_input_tokens_seen": 5902120,
      "step": 285,
      "time_per_iteration": 2.6445631980895996
    },
    {
      "auxiliary_loss_clip": 0.0132793,
      "auxiliary_loss_mlp": 0.01060621,
      "balance_loss_clip": 1.08746052,
      "balance_loss_mlp": 1.04530299,
      "epoch": 0.03438946672278002,
      "flos": 34752012716160.0,
      "grad_norm": 2.235819451352049,
      "language_loss": 0.73543012,
      "learning_rate": 3.9998034010489655e-06,
      "loss": 0.75931567,
      "num_input_tokens_seen": 5925810,
      "step": 286,
      "time_per_iteration": 2.8113138675689697
    },
    {
      "auxiliary_loss_clip": 0.0133639,
      "auxiliary_loss_mlp": 0.01058687,
      "balance_loss_clip": 1.09505486,
      "balance_loss_mlp": 1.04347563,
      "epoch": 0.03450970961341911,
      "flos": 22164102236160.0,
      "grad_norm": 2.4354484691343927,
      "language_loss": 0.75910318,
      "learning_rate": 3.999792327379946e-06,
      "loss": 0.78305387,
      "num_input_tokens_seen": 5945185,
      "step": 287,
      "time_per_iteration": 2.6789889335632324
    },
    {
      "auxiliary_loss_clip": 0.01335634,
      "auxiliary_loss_mlp": 0.01052782,
      "balance_loss_clip": 1.09318697,
      "balance_loss_mlp": 1.03748751,
      "epoch": 0.034629952504058197,
      "flos": 21725740656000.0,
      "grad_norm": 2.2778081205466574,
      "language_loss": 0.96258342,
      "learning_rate": 3.999780950344021e-06,
      "loss": 0.9864676,
      "num_input_tokens_seen": 5963375,
      "step": 288,
      "time_per_iteration": 2.7528510093688965
    },
    {
      "auxiliary_loss_clip": 0.01337043,
      "auxiliary_loss_mlp": 0.01057051,
      "balance_loss_clip": 1.09135699,
      "balance_loss_mlp": 1.04176283,
      "epoch": 0.034750195394697286,
      "flos": 20047994248320.0,
      "grad_norm": 1.9816994147285223,
      "language_loss": 0.82785642,
      "learning_rate": 3.999769269942916e-06,
      "loss": 0.8517974,
      "num_input_tokens_seen": 5983415,
      "step": 289,
      "time_per_iteration": 2.807454824447632
    },
    {
      "auxiliary_loss_clip": 0.01326651,
      "auxiliary_loss_mlp": 0.01048594,
      "balance_loss_clip": 1.08768094,
      "balance_loss_mlp": 1.03451514,
      "epoch": 0.034870438285336376,
      "flos": 27965865876480.0,
      "grad_norm": 1.8954701218870351,
      "language_loss": 0.81099749,
      "learning_rate": 3.999757286178402e-06,
      "loss": 0.83474994,
      "num_input_tokens_seen": 6005850,
      "step": 290,
      "time_per_iteration": 2.788499593734741
    },
    {
      "auxiliary_loss_clip": 0.01330669,
      "auxiliary_loss_mlp": 0.0105001,
      "balance_loss_clip": 1.08988369,
      "balance_loss_mlp": 1.03478718,
      "epoch": 0.03499068117597547,
      "flos": 22017514832640.0,
      "grad_norm": 4.299885121585119,
      "language_loss": 0.90792263,
      "learning_rate": 3.999744999052299e-06,
      "loss": 0.93172944,
      "num_input_tokens_seen": 6027240,
      "step": 291,
      "time_per_iteration": 2.7443437576293945
    },
    {
      "auxiliary_loss_clip": 0.01301724,
      "auxiliary_loss_mlp": 0.01010008,
      "balance_loss_clip": 1.12528706,
      "balance_loss_mlp": 1.00473928,
      "epoch": 0.03511092406661456,
      "flos": 57242147725440.0,
      "grad_norm": 0.9603902455439782,
      "language_loss": 0.6121186,
      "learning_rate": 3.9997324085664675e-06,
      "loss": 0.63523591,
      "num_input_tokens_seen": 6087470,
      "step": 292,
      "time_per_iteration": 3.236111640930176
    },
    {
      "auxiliary_loss_clip": 0.01327111,
      "auxiliary_loss_mlp": 0.01056604,
      "balance_loss_clip": 1.08605778,
      "balance_loss_mlp": 1.04153585,
      "epoch": 0.03523116695725365,
      "flos": 22928065626240.0,
      "grad_norm": 2.014457862468084,
      "language_loss": 0.91757083,
      "learning_rate": 3.999719514722821e-06,
      "loss": 0.94140804,
      "num_input_tokens_seen": 6107600,
      "step": 293,
      "time_per_iteration": 2.686160087585449
    },
    {
      "auxiliary_loss_clip": 0.01322523,
      "auxiliary_loss_mlp": 0.01061974,
      "balance_loss_clip": 1.08794928,
      "balance_loss_mlp": 1.04647696,
      "epoch": 0.03535140984789274,
      "flos": 36903241226880.0,
      "grad_norm": 2.521196896542567,
      "language_loss": 0.74759579,
      "learning_rate": 3.999706317523314e-06,
      "loss": 0.77144074,
      "num_input_tokens_seen": 6126160,
      "step": 294,
      "time_per_iteration": 2.850278377532959
    },
    {
      "auxiliary_loss_clip": 0.01318164,
      "auxiliary_loss_mlp": 0.01065942,
      "balance_loss_clip": 1.08438814,
      "balance_loss_mlp": 1.05059338,
      "epoch": 0.03547165273853183,
      "flos": 20449152316800.0,
      "grad_norm": 2.06749000659813,
      "language_loss": 0.86005014,
      "learning_rate": 3.999692816969948e-06,
      "loss": 0.88389117,
      "num_input_tokens_seen": 6145695,
      "step": 295,
      "time_per_iteration": 2.7040791511535645
    },
    {
      "auxiliary_loss_clip": 0.01296157,
      "auxiliary_loss_mlp": 0.01011983,
      "balance_loss_clip": 1.12172794,
      "balance_loss_mlp": 1.00661898,
      "epoch": 0.03559189562917092,
      "flos": 69850564871040.0,
      "grad_norm": 1.0744841039231894,
      "language_loss": 0.69467986,
      "learning_rate": 3.999679013064772e-06,
      "loss": 0.71776128,
      "num_input_tokens_seen": 6212440,
      "step": 296,
      "time_per_iteration": 3.2436583042144775
    },
    {
      "auxiliary_loss_clip": 0.0132202,
      "auxiliary_loss_mlp": 0.01053051,
      "balance_loss_clip": 1.0857693,
      "balance_loss_mlp": 1.03849578,
      "epoch": 0.03571213851981002,
      "flos": 21651944163840.0,
      "grad_norm": 2.3907382923984337,
      "language_loss": 0.85401487,
      "learning_rate": 3.99966490580988e-06,
      "loss": 0.87776566,
      "num_input_tokens_seen": 6229800,
      "step": 297,
      "time_per_iteration": 2.8155014514923096
    },
    {
      "auxiliary_loss_clip": 0.01330573,
      "auxiliary_loss_mlp": 0.01056448,
      "balance_loss_clip": 1.08965671,
      "balance_loss_mlp": 1.04074836,
      "epoch": 0.03583238141044911,
      "flos": 43945610757120.0,
      "grad_norm": 2.414783915620299,
      "language_loss": 0.65311241,
      "learning_rate": 3.999650495207411e-06,
      "loss": 0.6769827,
      "num_input_tokens_seen": 6255825,
      "step": 298,
      "time_per_iteration": 2.8225913047790527
    },
    {
      "auxiliary_loss_clip": 0.01325438,
      "auxiliary_loss_mlp": 0.01066677,
      "balance_loss_clip": 1.0889864,
      "balance_loss_mlp": 1.05014312,
      "epoch": 0.0359526243010882,
      "flos": 18910810592640.0,
      "grad_norm": 3.33261504060784,
      "language_loss": 0.90449864,
      "learning_rate": 3.999635781259553e-06,
      "loss": 0.92841977,
      "num_input_tokens_seen": 6271090,
      "step": 299,
      "time_per_iteration": 2.6684176921844482
    },
    {
      "auxiliary_loss_clip": 0.01286338,
      "auxiliary_loss_mlp": 0.01006292,
      "balance_loss_clip": 1.11458969,
      "balance_loss_mlp": 1.00102317,
      "epoch": 0.03607286719172729,
      "flos": 61668892782720.0,
      "grad_norm": 0.919055904246368,
      "language_loss": 0.5230794,
      "learning_rate": 3.999620763968535e-06,
      "loss": 0.54600573,
      "num_input_tokens_seen": 6329965,
      "step": 300,
      "time_per_iteration": 3.124018430709839
    },
    {
      "auxiliary_loss_clip": 0.01319916,
      "auxiliary_loss_mlp": 0.01058105,
      "balance_loss_clip": 1.08736467,
      "balance_loss_mlp": 1.04192841,
      "epoch": 0.03619311008236638,
      "flos": 27819062991360.0,
      "grad_norm": 1.590917043859873,
      "language_loss": 0.86298877,
      "learning_rate": 3.999605443336638e-06,
      "loss": 0.88676894,
      "num_input_tokens_seen": 6352095,
      "step": 301,
      "time_per_iteration": 2.6808321475982666
    },
    {
      "auxiliary_loss_clip": 0.01325507,
      "auxiliary_loss_mlp": 0.01058653,
      "balance_loss_clip": 1.08916736,
      "balance_loss_mlp": 1.04276228,
      "epoch": 0.03631335297300547,
      "flos": 13621133197440.0,
      "grad_norm": 2.3546212537050994,
      "language_loss": 0.89269519,
      "learning_rate": 3.999589819366185e-06,
      "loss": 0.91653681,
      "num_input_tokens_seen": 6365885,
      "step": 302,
      "time_per_iteration": 3.558337450027466
    },
    {
      "auxiliary_loss_clip": 0.01329058,
      "auxiliary_loss_mlp": 0.01057567,
      "balance_loss_clip": 1.09008455,
      "balance_loss_mlp": 1.04287457,
      "epoch": 0.036433595863644565,
      "flos": 27631788456960.0,
      "grad_norm": 1.9937965683336478,
      "language_loss": 0.84928674,
      "learning_rate": 3.999573892059547e-06,
      "loss": 0.87315291,
      "num_input_tokens_seen": 6385015,
      "step": 303,
      "time_per_iteration": 2.7173068523406982
    },
    {
      "auxiliary_loss_clip": 0.01331956,
      "auxiliary_loss_mlp": 0.01058083,
      "balance_loss_clip": 1.08982635,
      "balance_loss_mlp": 1.04082131,
      "epoch": 0.036553838754283655,
      "flos": 24572020314240.0,
      "grad_norm": 2.284656249997884,
      "language_loss": 0.8128171,
      "learning_rate": 3.999557661419138e-06,
      "loss": 0.83671749,
      "num_input_tokens_seen": 6405165,
      "step": 304,
      "time_per_iteration": 3.6380724906921387
    },
    {
      "auxiliary_loss_clip": 0.01328224,
      "auxiliary_loss_mlp": 0.01057721,
      "balance_loss_clip": 1.09160089,
      "balance_loss_mlp": 1.04233158,
      "epoch": 0.036674081644922744,
      "flos": 23404313076480.0,
      "grad_norm": 5.643400318077628,
      "language_loss": 0.81392622,
      "learning_rate": 3.9995411274474225e-06,
      "loss": 0.83778572,
      "num_input_tokens_seen": 6424445,
      "step": 305,
      "time_per_iteration": 2.701444149017334
    },
    {
      "auxiliary_loss_clip": 0.0132416,
      "auxiliary_loss_mlp": 0.01060231,
      "balance_loss_clip": 1.08916938,
      "balance_loss_mlp": 1.04470968,
      "epoch": 0.036794324535561834,
      "flos": 27489690253440.0,
      "grad_norm": 3.302506328480613,
      "language_loss": 0.81833971,
      "learning_rate": 3.999524290146908e-06,
      "loss": 0.84218359,
      "num_input_tokens_seen": 6444650,
      "step": 306,
      "time_per_iteration": 3.647221565246582
    },
    {
      "auxiliary_loss_clip": 0.01324385,
      "auxiliary_loss_mlp": 0.01063006,
      "balance_loss_clip": 1.08805275,
      "balance_loss_mlp": 1.04713893,
      "epoch": 0.036914567426200924,
      "flos": 19463476227840.0,
      "grad_norm": 2.2081985729499656,
      "language_loss": 0.9279958,
      "learning_rate": 3.9995071495201485e-06,
      "loss": 0.95186973,
      "num_input_tokens_seen": 6461755,
      "step": 307,
      "time_per_iteration": 2.728444814682007
    },
    {
      "auxiliary_loss_clip": 0.01325007,
      "auxiliary_loss_mlp": 0.01060594,
      "balance_loss_clip": 1.08908033,
      "balance_loss_mlp": 1.04583585,
      "epoch": 0.037034810316840014,
      "flos": 22309324922880.0,
      "grad_norm": 2.399741777494334,
      "language_loss": 0.9749068,
      "learning_rate": 3.999489705569744e-06,
      "loss": 0.99876273,
      "num_input_tokens_seen": 6479455,
      "step": 308,
      "time_per_iteration": 3.707104206085205
    },
    {
      "auxiliary_loss_clip": 0.01324022,
      "auxiliary_loss_mlp": 0.01051751,
      "balance_loss_clip": 1.08898151,
      "balance_loss_mlp": 1.03605103,
      "epoch": 0.03715505320747911,
      "flos": 18588333265920.0,
      "grad_norm": 1.9901587164688666,
      "language_loss": 0.86613011,
      "learning_rate": 3.999471958298341e-06,
      "loss": 0.88988781,
      "num_input_tokens_seen": 6498365,
      "step": 309,
      "time_per_iteration": 2.6418161392211914
    },
    {
      "auxiliary_loss_clip": 0.01329347,
      "auxiliary_loss_mlp": 0.01073235,
      "balance_loss_clip": 1.09132409,
      "balance_loss_mlp": 1.0576067,
      "epoch": 0.0372752960981182,
      "flos": 35955343267200.0,
      "grad_norm": 1.947793603280558,
      "language_loss": 0.75979769,
      "learning_rate": 3.999453907708631e-06,
      "loss": 0.78382349,
      "num_input_tokens_seen": 6520770,
      "step": 310,
      "time_per_iteration": 2.8056466579437256
    },
    {
      "auxiliary_loss_clip": 0.01317958,
      "auxiliary_loss_mlp": 0.01055529,
      "balance_loss_clip": 1.08717191,
      "balance_loss_mlp": 1.04128313,
      "epoch": 0.03739553898875729,
      "flos": 20814040627200.0,
      "grad_norm": 1.7987909884756,
      "language_loss": 0.81380582,
      "learning_rate": 3.999435553803353e-06,
      "loss": 0.83754075,
      "num_input_tokens_seen": 6540170,
      "step": 311,
      "time_per_iteration": 2.663356065750122
    },
    {
      "auxiliary_loss_clip": 0.01320201,
      "auxiliary_loss_mlp": 0.0105171,
      "balance_loss_clip": 1.08660769,
      "balance_loss_mlp": 1.03691018,
      "epoch": 0.03751578187939638,
      "flos": 20264140339200.0,
      "grad_norm": 2.40939944727452,
      "language_loss": 0.83275223,
      "learning_rate": 3.999416896585292e-06,
      "loss": 0.85647136,
      "num_input_tokens_seen": 6557200,
      "step": 312,
      "time_per_iteration": 2.673175096511841
    },
    {
      "auxiliary_loss_clip": 0.01320732,
      "auxiliary_loss_mlp": 0.01052655,
      "balance_loss_clip": 1.08813667,
      "balance_loss_mlp": 1.03663349,
      "epoch": 0.03763602477003547,
      "flos": 20668063754880.0,
      "grad_norm": 3.569079998876569,
      "language_loss": 0.85956919,
      "learning_rate": 3.9993979360572775e-06,
      "loss": 0.88330305,
      "num_input_tokens_seen": 6577340,
      "step": 313,
      "time_per_iteration": 2.6363887786865234
    },
    {
      "auxiliary_loss_clip": 0.01326326,
      "auxiliary_loss_mlp": 0.01062789,
      "balance_loss_clip": 1.09021068,
      "balance_loss_mlp": 1.04670739,
      "epoch": 0.03775626766067456,
      "flos": 16691352197760.0,
      "grad_norm": 2.9476296965622444,
      "language_loss": 0.83368045,
      "learning_rate": 3.999378672222185e-06,
      "loss": 0.8575716,
      "num_input_tokens_seen": 6595125,
      "step": 314,
      "time_per_iteration": 2.685351848602295
    },
    {
      "auxiliary_loss_clip": 0.01325003,
      "auxiliary_loss_mlp": 0.01070999,
      "balance_loss_clip": 1.0887661,
      "balance_loss_mlp": 1.0560863,
      "epoch": 0.03787651055131366,
      "flos": 21141797253120.0,
      "grad_norm": 1.958543097100174,
      "language_loss": 0.83029556,
      "learning_rate": 3.9993591050829385e-06,
      "loss": 0.85425556,
      "num_input_tokens_seen": 6612990,
      "step": 315,
      "time_per_iteration": 2.661104679107666
    },
    {
      "auxiliary_loss_clip": 0.01319984,
      "auxiliary_loss_mlp": 0.01059043,
      "balance_loss_clip": 1.0874722,
      "balance_loss_mlp": 1.04500616,
      "epoch": 0.037996753441952746,
      "flos": 22018089450240.0,
      "grad_norm": 1.9533518967973622,
      "language_loss": 0.79295075,
      "learning_rate": 3.999339234642506e-06,
      "loss": 0.81674105,
      "num_input_tokens_seen": 6632740,
      "step": 316,
      "time_per_iteration": 2.663609504699707
    },
    {
      "auxiliary_loss_clip": 0.01325908,
      "auxiliary_loss_mlp": 0.01062061,
      "balance_loss_clip": 1.08830214,
      "balance_loss_mlp": 1.04521668,
      "epoch": 0.038116996332591836,
      "flos": 27709391790720.0,
      "grad_norm": 1.8778840358951758,
      "language_loss": 0.83731687,
      "learning_rate": 3.9993190609038994e-06,
      "loss": 0.86119652,
      "num_input_tokens_seen": 6651505,
      "step": 317,
      "time_per_iteration": 2.708087205886841
    },
    {
      "auxiliary_loss_clip": 0.01322239,
      "auxiliary_loss_mlp": 0.01055123,
      "balance_loss_clip": 1.08887744,
      "balance_loss_mlp": 1.04046023,
      "epoch": 0.038237239223230926,
      "flos": 21178067011200.0,
      "grad_norm": 2.1005862019832544,
      "language_loss": 0.83239818,
      "learning_rate": 3.999298583870182e-06,
      "loss": 0.85617179,
      "num_input_tokens_seen": 6671090,
      "step": 318,
      "time_per_iteration": 2.685556650161743
    },
    {
      "auxiliary_loss_clip": 0.01319767,
      "auxiliary_loss_mlp": 0.01053146,
      "balance_loss_clip": 1.08678842,
      "balance_loss_mlp": 1.03783953,
      "epoch": 0.038357482113870016,
      "flos": 25556618995200.0,
      "grad_norm": 2.6848076717811984,
      "language_loss": 0.77350283,
      "learning_rate": 3.999277803544458e-06,
      "loss": 0.79723191,
      "num_input_tokens_seen": 6691245,
      "step": 319,
      "time_per_iteration": 2.7490956783294678
    },
    {
      "auxiliary_loss_clip": 0.01263059,
      "auxiliary_loss_mlp": 0.01013824,
      "balance_loss_clip": 1.09847927,
      "balance_loss_mlp": 1.00891292,
      "epoch": 0.038477725004509106,
      "flos": 59227578034560.0,
      "grad_norm": 0.957623578733668,
      "language_loss": 0.6241945,
      "learning_rate": 3.999256719929882e-06,
      "loss": 0.64696336,
      "num_input_tokens_seen": 6752520,
      "step": 320,
      "time_per_iteration": 3.200852394104004
    },
    {
      "auxiliary_loss_clip": 0.01261223,
      "auxiliary_loss_mlp": 0.01007665,
      "balance_loss_clip": 1.09707046,
      "balance_loss_mlp": 1.00292063,
      "epoch": 0.0385979678951482,
      "flos": 67317676398720.0,
      "grad_norm": 1.2314174498361476,
      "language_loss": 0.67169464,
      "learning_rate": 3.999235333029651e-06,
      "loss": 0.6943835,
      "num_input_tokens_seen": 6806460,
      "step": 321,
      "time_per_iteration": 3.1301326751708984
    },
    {
      "auxiliary_loss_clip": 0.01317695,
      "auxiliary_loss_mlp": 0.01068063,
      "balance_loss_clip": 1.08765244,
      "balance_loss_mlp": 1.05235696,
      "epoch": 0.03871821078578729,
      "flos": 22746752749440.0,
      "grad_norm": 2.1480441434304156,
      "language_loss": 0.8188405,
      "learning_rate": 3.999213642847009e-06,
      "loss": 0.8426981,
      "num_input_tokens_seen": 6827045,
      "step": 322,
      "time_per_iteration": 2.727092742919922
    },
    {
      "auxiliary_loss_clip": 0.01318731,
      "auxiliary_loss_mlp": 0.01053232,
      "balance_loss_clip": 1.08692503,
      "balance_loss_mlp": 1.0371145,
      "epoch": 0.03883845367642638,
      "flos": 26280613526400.0,
      "grad_norm": 1.867365758171466,
      "language_loss": 0.91263676,
      "learning_rate": 3.999191649385247e-06,
      "loss": 0.93635631,
      "num_input_tokens_seen": 6848220,
      "step": 323,
      "time_per_iteration": 2.723374605178833
    },
    {
      "auxiliary_loss_clip": 0.01254009,
      "auxiliary_loss_mlp": 0.01003291,
      "balance_loss_clip": 1.09243894,
      "balance_loss_mlp": 0.99878484,
      "epoch": 0.03895869656706547,
      "flos": 56962835568000.0,
      "grad_norm": 0.9105747183693512,
      "language_loss": 0.59850091,
      "learning_rate": 3.999169352647702e-06,
      "loss": 0.6210739,
      "num_input_tokens_seen": 6909400,
      "step": 324,
      "time_per_iteration": 3.2467825412750244
    },
    {
      "auxiliary_loss_clip": 0.01324138,
      "auxiliary_loss_mlp": 0.01064228,
      "balance_loss_clip": 1.08696461,
      "balance_loss_mlp": 1.04648328,
      "epoch": 0.03907893945770456,
      "flos": 24863363527680.0,
      "grad_norm": 2.1367008261696174,
      "language_loss": 0.8299309,
      "learning_rate": 3.999146752637755e-06,
      "loss": 0.85381454,
      "num_input_tokens_seen": 6930445,
      "step": 325,
      "time_per_iteration": 2.8821394443511963
    },
    {
      "auxiliary_loss_clip": 0.01317885,
      "auxiliary_loss_mlp": 0.01059235,
      "balance_loss_clip": 1.08483148,
      "balance_loss_mlp": 1.04340363,
      "epoch": 0.03919918234834365,
      "flos": 18368595815040.0,
      "grad_norm": 2.3643817315765094,
      "language_loss": 0.89487243,
      "learning_rate": 3.999123849358836e-06,
      "loss": 0.91864365,
      "num_input_tokens_seen": 6948110,
      "step": 326,
      "time_per_iteration": 2.6631593704223633
    },
    {
      "auxiliary_loss_clip": 0.01316921,
      "auxiliary_loss_mlp": 0.01048432,
      "balance_loss_clip": 1.08562589,
      "balance_loss_mlp": 1.03341162,
      "epoch": 0.03931942523898275,
      "flos": 25225414663680.0,
      "grad_norm": 1.873372532507237,
      "language_loss": 0.74638951,
      "learning_rate": 3.999100642814418e-06,
      "loss": 0.77004302,
      "num_input_tokens_seen": 6968550,
      "step": 327,
      "time_per_iteration": 2.662123203277588
    },
    {
      "auxiliary_loss_clip": 0.01319467,
      "auxiliary_loss_mlp": 0.01054427,
      "balance_loss_clip": 1.08762598,
      "balance_loss_mlp": 1.03872728,
      "epoch": 0.03943966812962184,
      "flos": 23257905240960.0,
      "grad_norm": 5.5660570479650255,
      "language_loss": 0.88391262,
      "learning_rate": 3.999077133008022e-06,
      "loss": 0.90765154,
      "num_input_tokens_seen": 6987135,
      "step": 328,
      "time_per_iteration": 2.7567856311798096
    },
    {
      "auxiliary_loss_clip": 0.01324966,
      "auxiliary_loss_mlp": 0.01056223,
      "balance_loss_clip": 1.08920717,
      "balance_loss_mlp": 1.03977227,
      "epoch": 0.03955991102026093,
      "flos": 29168837291520.0,
      "grad_norm": 1.8224433606844714,
      "language_loss": 0.90631604,
      "learning_rate": 3.9990533199432145e-06,
      "loss": 0.93012798,
      "num_input_tokens_seen": 7008630,
      "step": 329,
      "time_per_iteration": 3.7631208896636963
    },
    {
      "auxiliary_loss_clip": 0.01323739,
      "auxiliary_loss_mlp": 0.01057285,
      "balance_loss_clip": 1.09028649,
      "balance_loss_mlp": 1.04192519,
      "epoch": 0.03968015391090002,
      "flos": 17602441695360.0,
      "grad_norm": 2.301101558784639,
      "language_loss": 0.75612521,
      "learning_rate": 3.999029203623608e-06,
      "loss": 0.77993548,
      "num_input_tokens_seen": 7026350,
      "step": 330,
      "time_per_iteration": 3.7259390354156494
    },
    {
      "auxiliary_loss_clip": 0.01314509,
      "auxiliary_loss_mlp": 0.01057776,
      "balance_loss_clip": 1.08644581,
      "balance_loss_mlp": 1.04326808,
      "epoch": 0.03980039680153911,
      "flos": 21799285752960.0,
      "grad_norm": 1.9021873755462995,
      "language_loss": 0.87010121,
      "learning_rate": 3.99900478405286e-06,
      "loss": 0.89382404,
      "num_input_tokens_seen": 7045660,
      "step": 331,
      "time_per_iteration": 2.7362570762634277
    },
    {
      "auxiliary_loss_clip": 0.01311889,
      "auxiliary_loss_mlp": 0.0105565,
      "balance_loss_clip": 1.08637559,
      "balance_loss_mlp": 1.04039156,
      "epoch": 0.0399206396921782,
      "flos": 15195134148480.0,
      "grad_norm": 2.899467198381227,
      "language_loss": 0.82355404,
      "learning_rate": 3.998980061234676e-06,
      "loss": 0.84722948,
      "num_input_tokens_seen": 7063575,
      "step": 332,
      "time_per_iteration": 2.564887046813965
    },
    {
      "auxiliary_loss_clip": 0.01317159,
      "auxiliary_loss_mlp": 0.01051539,
      "balance_loss_clip": 1.0855602,
      "balance_loss_mlp": 1.03706706,
      "epoch": 0.040040882582817294,
      "flos": 14422910630400.0,
      "grad_norm": 2.359886785578573,
      "language_loss": 0.75743723,
      "learning_rate": 3.9989550351728055e-06,
      "loss": 0.78112423,
      "num_input_tokens_seen": 7080505,
      "step": 333,
      "time_per_iteration": 3.622380495071411
    },
    {
      "auxiliary_loss_clip": 0.01312425,
      "auxiliary_loss_mlp": 0.0107005,
      "balance_loss_clip": 1.08413112,
      "balance_loss_mlp": 1.05492222,
      "epoch": 0.040161125473456384,
      "flos": 19280906375040.0,
      "grad_norm": 2.24184282021934,
      "language_loss": 0.8412016,
      "learning_rate": 3.998929705871046e-06,
      "loss": 0.86502635,
      "num_input_tokens_seen": 7097860,
      "step": 334,
      "time_per_iteration": 2.7787773609161377
    },
    {
      "auxiliary_loss_clip": 0.01312057,
      "auxiliary_loss_mlp": 0.01056945,
      "balance_loss_clip": 1.08507097,
      "balance_loss_mlp": 1.04195476,
      "epoch": 0.040281368364095474,
      "flos": 17821101738240.0,
      "grad_norm": 3.225907738546793,
      "language_loss": 0.89054215,
      "learning_rate": 3.99890407333324e-06,
      "loss": 0.91423213,
      "num_input_tokens_seen": 7116390,
      "step": 335,
      "time_per_iteration": 3.6070199012756348
    },
    {
      "auxiliary_loss_clip": 0.01311989,
      "auxiliary_loss_mlp": 0.01051851,
      "balance_loss_clip": 1.08430457,
      "balance_loss_mlp": 1.03719401,
      "epoch": 0.040401611254734564,
      "flos": 19573757959680.0,
      "grad_norm": 1.716287407534517,
      "language_loss": 0.87066573,
      "learning_rate": 3.998878137563275e-06,
      "loss": 0.89430416,
      "num_input_tokens_seen": 7135940,
      "step": 336,
      "time_per_iteration": 2.634819984436035
    },
    {
      "auxiliary_loss_clip": 0.01314514,
      "auxiliary_loss_mlp": 0.0105435,
      "balance_loss_clip": 1.08620882,
      "balance_loss_mlp": 1.04007506,
      "epoch": 0.040521854145373654,
      "flos": 22054466949120.0,
      "grad_norm": 2.350390769492453,
      "language_loss": 0.85484165,
      "learning_rate": 3.998851898565085e-06,
      "loss": 0.87853032,
      "num_input_tokens_seen": 7155745,
      "step": 337,
      "time_per_iteration": 2.640092611312866
    },
    {
      "auxiliary_loss_clip": 0.01314337,
      "auxiliary_loss_mlp": 0.01047638,
      "balance_loss_clip": 1.08517027,
      "balance_loss_mlp": 1.03297544,
      "epoch": 0.04064209703601274,
      "flos": 22674644196480.0,
      "grad_norm": 2.44763968826883,
      "language_loss": 0.83110791,
      "learning_rate": 3.998825356342653e-06,
      "loss": 0.85472769,
      "num_input_tokens_seen": 7175920,
      "step": 338,
      "time_per_iteration": 2.6961305141448975
    },
    {
      "auxiliary_loss_clip": 0.01309037,
      "auxiliary_loss_mlp": 0.01049529,
      "balance_loss_clip": 1.08350503,
      "balance_loss_mlp": 1.03375793,
      "epoch": 0.04076233992665183,
      "flos": 38582172783360.0,
      "grad_norm": 2.182998875612784,
      "language_loss": 0.73315269,
      "learning_rate": 3.998798510900003e-06,
      "loss": 0.75673831,
      "num_input_tokens_seen": 7198720,
      "step": 339,
      "time_per_iteration": 2.7684221267700195
    },
    {
      "auxiliary_loss_clip": 0.01312091,
      "auxiliary_loss_mlp": 0.01048511,
      "balance_loss_clip": 1.08282232,
      "balance_loss_mlp": 1.03495717,
      "epoch": 0.04088258281729093,
      "flos": 25885309374720.0,
      "grad_norm": 2.3487336517645097,
      "language_loss": 0.84002692,
      "learning_rate": 3.998771362241207e-06,
      "loss": 0.86363292,
      "num_input_tokens_seen": 7219125,
      "step": 340,
      "time_per_iteration": 2.706491231918335
    },
    {
      "auxiliary_loss_clip": 0.01304505,
      "auxiliary_loss_mlp": 0.01054368,
      "balance_loss_clip": 1.08076298,
      "balance_loss_mlp": 1.04044986,
      "epoch": 0.04100282570793002,
      "flos": 19789832223360.0,
      "grad_norm": 2.398612581975192,
      "language_loss": 0.88087493,
      "learning_rate": 3.998743910370385e-06,
      "loss": 0.90446371,
      "num_input_tokens_seen": 7237985,
      "step": 341,
      "time_per_iteration": 2.722095251083374
    },
    {
      "auxiliary_loss_clip": 0.01316921,
      "auxiliary_loss_mlp": 0.01060516,
      "balance_loss_clip": 1.09072542,
      "balance_loss_mlp": 1.04503036,
      "epoch": 0.04112306859856911,
      "flos": 22565152563840.0,
      "grad_norm": 2.3193774457078784,
      "language_loss": 0.73678493,
      "learning_rate": 3.998716155291702e-06,
      "loss": 0.76055932,
      "num_input_tokens_seen": 7255825,
      "step": 342,
      "time_per_iteration": 2.7793984413146973
    },
    {
      "auxiliary_loss_clip": 0.01316895,
      "auxiliary_loss_mlp": 0.01052746,
      "balance_loss_clip": 1.08914804,
      "balance_loss_mlp": 1.03761816,
      "epoch": 0.0412433114892082,
      "flos": 25040654081280.0,
      "grad_norm": 1.8237885681159773,
      "language_loss": 0.90536803,
      "learning_rate": 3.998688097009366e-06,
      "loss": 0.92906445,
      "num_input_tokens_seen": 7276590,
      "step": 343,
      "time_per_iteration": 2.7760610580444336
    },
    {
      "auxiliary_loss_clip": 0.01313532,
      "auxiliary_loss_mlp": 0.01050231,
      "balance_loss_clip": 1.0865972,
      "balance_loss_mlp": 1.03581238,
      "epoch": 0.04136355437984729,
      "flos": 25191371548800.0,
      "grad_norm": 2.175402597736529,
      "language_loss": 0.79973251,
      "learning_rate": 3.998659735527636e-06,
      "loss": 0.8233701,
      "num_input_tokens_seen": 7295680,
      "step": 344,
      "time_per_iteration": 2.665452003479004
    },
    {
      "auxiliary_loss_clip": 0.01310503,
      "auxiliary_loss_mlp": 0.01047006,
      "balance_loss_clip": 1.08570528,
      "balance_loss_mlp": 1.03248584,
      "epoch": 0.04148379727048638,
      "flos": 22966777509120.0,
      "grad_norm": 1.9996322392604697,
      "language_loss": 0.77578318,
      "learning_rate": 3.998631070850813e-06,
      "loss": 0.79935831,
      "num_input_tokens_seen": 7316300,
      "step": 345,
      "time_per_iteration": 2.8552448749542236
    },
    {
      "auxiliary_loss_clip": 0.0130888,
      "auxiliary_loss_mlp": 0.0104767,
      "balance_loss_clip": 1.08635271,
      "balance_loss_mlp": 1.03348458,
      "epoch": 0.041604040161125476,
      "flos": 14063481187200.0,
      "grad_norm": 2.100132638673303,
      "language_loss": 0.83257282,
      "learning_rate": 3.9986021029832455e-06,
      "loss": 0.85613823,
      "num_input_tokens_seen": 7333615,
      "step": 346,
      "time_per_iteration": 2.690383195877075
    },
    {
      "auxiliary_loss_clip": 0.0131672,
      "auxiliary_loss_mlp": 0.01054446,
      "balance_loss_clip": 1.08467698,
      "balance_loss_mlp": 1.03927028,
      "epoch": 0.041724283051764566,
      "flos": 12091877614080.0,
      "grad_norm": 2.5105221738479435,
      "language_loss": 0.91407454,
      "learning_rate": 3.9985728319293285e-06,
      "loss": 0.93778616,
      "num_input_tokens_seen": 7347590,
      "step": 347,
      "time_per_iteration": 2.6803200244903564
    },
    {
      "auxiliary_loss_clip": 0.01317013,
      "auxiliary_loss_mlp": 0.0105791,
      "balance_loss_clip": 1.084777,
      "balance_loss_mlp": 1.0419476,
      "epoch": 0.041844525942403656,
      "flos": 12385303816320.0,
      "grad_norm": 2.6293732250845934,
      "language_loss": 0.85233259,
      "learning_rate": 3.998543257693501e-06,
      "loss": 0.87608182,
      "num_input_tokens_seen": 7364345,
      "step": 348,
      "time_per_iteration": 2.735825538635254
    },
    {
      "auxiliary_loss_clip": 0.01313549,
      "auxiliary_loss_mlp": 0.01052083,
      "balance_loss_clip": 1.08754873,
      "balance_loss_mlp": 1.03774261,
      "epoch": 0.041964768833042745,
      "flos": 23769345041280.0,
      "grad_norm": 1.827060606087551,
      "language_loss": 0.87758636,
      "learning_rate": 3.998513380280251e-06,
      "loss": 0.90124273,
      "num_input_tokens_seen": 7384625,
      "step": 349,
      "time_per_iteration": 2.695889949798584
    },
    {
      "auxiliary_loss_clip": 0.01319207,
      "auxiliary_loss_mlp": 0.01059718,
      "balance_loss_clip": 1.08752012,
      "balance_loss_mlp": 1.04425693,
      "epoch": 0.042085011723681835,
      "flos": 11875336473600.0,
      "grad_norm": 2.4443329349921328,
      "language_loss": 0.94926071,
      "learning_rate": 3.99848319969411e-06,
      "loss": 0.97305,
      "num_input_tokens_seen": 7402225,
      "step": 350,
      "time_per_iteration": 2.683741807937622
    },
    {
      "auxiliary_loss_clip": 0.01323332,
      "auxiliary_loss_mlp": 0.01079803,
      "balance_loss_clip": 1.09286213,
      "balance_loss_mlp": 1.06361449,
      "epoch": 0.042205254614320925,
      "flos": 16873957964160.0,
      "grad_norm": 2.6414075301769855,
      "language_loss": 0.79003799,
      "learning_rate": 3.9984527159396564e-06,
      "loss": 0.81406933,
      "num_input_tokens_seen": 7420865,
      "step": 351,
      "time_per_iteration": 2.6703805923461914
    },
    {
      "auxiliary_loss_clip": 0.0131477,
      "auxiliary_loss_mlp": 0.01067143,
      "balance_loss_clip": 1.08513737,
      "balance_loss_mlp": 1.051682,
      "epoch": 0.04232549750496002,
      "flos": 25118508810240.0,
      "grad_norm": 2.0466854113709108,
      "language_loss": 0.84430909,
      "learning_rate": 3.9984219290215154e-06,
      "loss": 0.86812824,
      "num_input_tokens_seen": 7441040,
      "step": 352,
      "time_per_iteration": 2.7543489933013916
    },
    {
      "auxiliary_loss_clip": 0.01306988,
      "auxiliary_loss_mlp": 0.01050534,
      "balance_loss_clip": 1.08384001,
      "balance_loss_mlp": 1.03694367,
      "epoch": 0.04244574039559911,
      "flos": 26724541714560.0,
      "grad_norm": 1.589469652422911,
      "language_loss": 0.89233238,
      "learning_rate": 3.998390838944356e-06,
      "loss": 0.91590762,
      "num_input_tokens_seen": 7462545,
      "step": 353,
      "time_per_iteration": 2.7222673892974854
    },
    {
      "auxiliary_loss_clip": 0.01313797,
      "auxiliary_loss_mlp": 0.01047112,
      "balance_loss_clip": 1.08504236,
      "balance_loss_mlp": 1.03281307,
      "epoch": 0.0425659832862382,
      "flos": 20923244951040.0,
      "grad_norm": 2.290136035254513,
      "language_loss": 0.90161014,
      "learning_rate": 3.998359445712895e-06,
      "loss": 0.9252193,
      "num_input_tokens_seen": 7481650,
      "step": 354,
      "time_per_iteration": 2.666365623474121
    },
    {
      "auxiliary_loss_clip": 0.01309824,
      "auxiliary_loss_mlp": 0.01051617,
      "balance_loss_clip": 1.08214974,
      "balance_loss_mlp": 1.03621507,
      "epoch": 0.04268622617687729,
      "flos": 23331127115520.0,
      "grad_norm": 2.0803201349018314,
      "language_loss": 0.81446379,
      "learning_rate": 3.9983277493318955e-06,
      "loss": 0.8380782,
      "num_input_tokens_seen": 7500945,
      "step": 355,
      "time_per_iteration": 2.6575381755828857
    },
    {
      "auxiliary_loss_clip": 0.01310679,
      "auxiliary_loss_mlp": 0.01052951,
      "balance_loss_clip": 1.08243203,
      "balance_loss_mlp": 1.0388484,
      "epoch": 0.04280646906751638,
      "flos": 25994010908160.0,
      "grad_norm": 1.8752142704642512,
      "language_loss": 0.81173491,
      "learning_rate": 3.998295749806165e-06,
      "loss": 0.83537114,
      "num_input_tokens_seen": 7522170,
      "step": 356,
      "time_per_iteration": 3.774496555328369
    },
    {
      "auxiliary_loss_clip": 0.01316253,
      "auxiliary_loss_mlp": 0.01057671,
      "balance_loss_clip": 1.08813977,
      "balance_loss_mlp": 1.04320502,
      "epoch": 0.04292671195815547,
      "flos": 26906824258560.0,
      "grad_norm": 1.8827040325997253,
      "language_loss": 0.83420122,
      "learning_rate": 3.998263447140558e-06,
      "loss": 0.85794044,
      "num_input_tokens_seen": 7542370,
      "step": 357,
      "time_per_iteration": 3.636260986328125
    },
    {
      "auxiliary_loss_clip": 0.0130865,
      "auxiliary_loss_mlp": 0.01056036,
      "balance_loss_clip": 1.08235526,
      "balance_loss_mlp": 1.04220796,
      "epoch": 0.04304695484879457,
      "flos": 39457315745280.0,
      "grad_norm": 2.266123965498067,
      "language_loss": 0.81879109,
      "learning_rate": 3.998230841339976e-06,
      "loss": 0.84243798,
      "num_input_tokens_seen": 7564380,
      "step": 358,
      "time_per_iteration": 2.850451946258545
    },
    {
      "auxiliary_loss_clip": 0.01310614,
      "auxiliary_loss_mlp": 0.01060436,
      "balance_loss_clip": 1.08772337,
      "balance_loss_mlp": 1.04640555,
      "epoch": 0.04316719773943366,
      "flos": 19646297475840.0,
      "grad_norm": 2.1917376463492717,
      "language_loss": 0.84912968,
      "learning_rate": 3.998197932409363e-06,
      "loss": 0.87284011,
      "num_input_tokens_seen": 7582390,
      "step": 359,
      "time_per_iteration": 3.712926149368286
    },
    {
      "auxiliary_loss_clip": 0.01305837,
      "auxiliary_loss_mlp": 0.01054971,
      "balance_loss_clip": 1.08288252,
      "balance_loss_mlp": 1.04061794,
      "epoch": 0.04328744063007275,
      "flos": 22452320966400.0,
      "grad_norm": 2.0685205255133465,
      "language_loss": 0.86270547,
      "learning_rate": 3.9981647203537125e-06,
      "loss": 0.8863135,
      "num_input_tokens_seen": 7599890,
      "step": 360,
      "time_per_iteration": 2.6692707538604736
    },
    {
      "auxiliary_loss_clip": 0.01310273,
      "auxiliary_loss_mlp": 0.01049067,
      "balance_loss_clip": 1.08394361,
      "balance_loss_mlp": 1.03448772,
      "epoch": 0.04340768352071184,
      "flos": 21283033530240.0,
      "grad_norm": 5.586390571862419,
      "language_loss": 0.9600355,
      "learning_rate": 3.998131205178063e-06,
      "loss": 0.98362899,
      "num_input_tokens_seen": 7618360,
      "step": 361,
      "time_per_iteration": 3.7244393825531006
    },
    {
      "auxiliary_loss_clip": 0.01310165,
      "auxiliary_loss_mlp": 0.01045555,
      "balance_loss_clip": 1.08228409,
      "balance_loss_mlp": 1.03103518,
      "epoch": 0.04352792641135093,
      "flos": 11583705951360.0,
      "grad_norm": 5.047068737416688,
      "language_loss": 0.76703423,
      "learning_rate": 3.998097386887498e-06,
      "loss": 0.79059148,
      "num_input_tokens_seen": 7635435,
      "step": 362,
      "time_per_iteration": 2.6655263900756836
    },
    {
      "auxiliary_loss_clip": 0.01308384,
      "auxiliary_loss_mlp": 0.01043398,
      "balance_loss_clip": 1.08522797,
      "balance_loss_mlp": 1.02992773,
      "epoch": 0.04364816930199002,
      "flos": 23623547736960.0,
      "grad_norm": 1.6831606492282973,
      "language_loss": 0.84875262,
      "learning_rate": 3.998063265487148e-06,
      "loss": 0.87227046,
      "num_input_tokens_seen": 7656485,
      "step": 363,
      "time_per_iteration": 2.62864089012146
    },
    {
      "auxiliary_loss_clip": 0.01308401,
      "auxiliary_loss_mlp": 0.0105456,
      "balance_loss_clip": 1.08393574,
      "balance_loss_mlp": 1.04129767,
      "epoch": 0.043768412192629114,
      "flos": 14429734214400.0,
      "grad_norm": 1.779178090461871,
      "language_loss": 0.80952382,
      "learning_rate": 3.99802884098219e-06,
      "loss": 0.83315349,
      "num_input_tokens_seen": 7674595,
      "step": 364,
      "time_per_iteration": 2.689617156982422
    },
    {
      "auxiliary_loss_clip": 0.01310793,
      "auxiliary_loss_mlp": 0.01048566,
      "balance_loss_clip": 1.08249664,
      "balance_loss_mlp": 1.03397512,
      "epoch": 0.043888655083268203,
      "flos": 26468893641600.0,
      "grad_norm": 2.1870654740114257,
      "language_loss": 0.82337499,
      "learning_rate": 3.997994113377845e-06,
      "loss": 0.84696865,
      "num_input_tokens_seen": 7693495,
      "step": 365,
      "time_per_iteration": 2.6761844158172607
    },
    {
      "auxiliary_loss_clip": 0.01311434,
      "auxiliary_loss_mlp": 0.0104819,
      "balance_loss_clip": 1.08407724,
      "balance_loss_mlp": 1.03389657,
      "epoch": 0.04400889797390729,
      "flos": 27235263242880.0,
      "grad_norm": 1.907203169489908,
      "language_loss": 0.83125806,
      "learning_rate": 3.9979590826793815e-06,
      "loss": 0.85485429,
      "num_input_tokens_seen": 7714685,
      "step": 366,
      "time_per_iteration": 2.7311248779296875
    },
    {
      "auxiliary_loss_clip": 0.01313516,
      "auxiliary_loss_mlp": 0.0106135,
      "balance_loss_clip": 1.08654165,
      "balance_loss_mlp": 1.04514313,
      "epoch": 0.04412914086454638,
      "flos": 20119528183680.0,
      "grad_norm": 2.5625884563586574,
      "language_loss": 0.80940086,
      "learning_rate": 3.997923748892113e-06,
      "loss": 0.83314949,
      "num_input_tokens_seen": 7734005,
      "step": 367,
      "time_per_iteration": 2.691741704940796
    },
    {
      "auxiliary_loss_clip": 0.01304008,
      "auxiliary_loss_mlp": 0.01050519,
      "balance_loss_clip": 1.08326125,
      "balance_loss_mlp": 1.0370605,
      "epoch": 0.04424938375518547,
      "flos": 22604618632320.0,
      "grad_norm": 1.6151385664319264,
      "language_loss": 0.88594168,
      "learning_rate": 3.9978881120214015e-06,
      "loss": 0.90948701,
      "num_input_tokens_seen": 7755525,
      "step": 368,
      "time_per_iteration": 2.708890676498413
    },
    {
      "auxiliary_loss_clip": 0.01307372,
      "auxiliary_loss_mlp": 0.01055921,
      "balance_loss_clip": 1.08100963,
      "balance_loss_mlp": 1.04061413,
      "epoch": 0.04436962664582456,
      "flos": 24132365844480.0,
      "grad_norm": 2.2206795692509806,
      "language_loss": 0.79135525,
      "learning_rate": 3.997852172072652e-06,
      "loss": 0.8149882,
      "num_input_tokens_seen": 7776740,
      "step": 369,
      "time_per_iteration": 2.6990280151367188
    },
    {
      "auxiliary_loss_clip": 0.01309194,
      "auxiliary_loss_mlp": 0.0105118,
      "balance_loss_clip": 1.08349037,
      "balance_loss_mlp": 1.03601599,
      "epoch": 0.04448986953646366,
      "flos": 18222906251520.0,
      "grad_norm": 3.0950739396998848,
      "language_loss": 0.89158559,
      "learning_rate": 3.9978159290513155e-06,
      "loss": 0.91518933,
      "num_input_tokens_seen": 7794820,
      "step": 370,
      "time_per_iteration": 2.578137159347534
    },
    {
      "auxiliary_loss_clip": 0.01312843,
      "auxiliary_loss_mlp": 0.01046746,
      "balance_loss_clip": 1.08519232,
      "balance_loss_mlp": 1.03064716,
      "epoch": 0.04461011242710275,
      "flos": 30117920400000.0,
      "grad_norm": 1.8988711238286933,
      "language_loss": 0.80127704,
      "learning_rate": 3.997779382962892e-06,
      "loss": 0.82487291,
      "num_input_tokens_seen": 7817705,
      "step": 371,
      "time_per_iteration": 2.7590582370758057
    },
    {
      "auxiliary_loss_clip": 0.01302878,
      "auxiliary_loss_mlp": 0.01044011,
      "balance_loss_clip": 1.07976639,
      "balance_loss_mlp": 1.03043294,
      "epoch": 0.04473035531774184,
      "flos": 29752529299200.0,
      "grad_norm": 2.0023167868514022,
      "language_loss": 0.73859298,
      "learning_rate": 3.997742533812924e-06,
      "loss": 0.76206183,
      "num_input_tokens_seen": 7840970,
      "step": 372,
      "time_per_iteration": 2.678740978240967
    },
    {
      "auxiliary_loss_clip": 0.01305957,
      "auxiliary_loss_mlp": 0.01066339,
      "balance_loss_clip": 1.08253479,
      "balance_loss_mlp": 1.05224848,
      "epoch": 0.04485059820838093,
      "flos": 13151565676800.0,
      "grad_norm": 2.550772880124689,
      "language_loss": 0.92303097,
      "learning_rate": 3.997705381607001e-06,
      "loss": 0.94675392,
      "num_input_tokens_seen": 7857785,
      "step": 373,
      "time_per_iteration": 2.632939577102661
    },
    {
      "auxiliary_loss_clip": 0.01244759,
      "auxiliary_loss_mlp": 0.01015428,
      "balance_loss_clip": 1.09089804,
      "balance_loss_mlp": 1.01194692,
      "epoch": 0.04497084109902002,
      "flos": 68094209548800.0,
      "grad_norm": 0.9788669158636881,
      "language_loss": 0.60301638,
      "learning_rate": 3.997667926350761e-06,
      "loss": 0.62561822,
      "num_input_tokens_seen": 7916115,
      "step": 374,
      "time_per_iteration": 3.1555440425872803
    },
    {
      "auxiliary_loss_clip": 0.01243791,
      "auxiliary_loss_mlp": 0.01011847,
      "balance_loss_clip": 1.09032488,
      "balance_loss_mlp": 1.00831878,
      "epoch": 0.04509108398965911,
      "flos": 64342263346560.0,
      "grad_norm": 0.9009877031467165,
      "language_loss": 0.57825178,
      "learning_rate": 3.997630168049886e-06,
      "loss": 0.60080814,
      "num_input_tokens_seen": 7974480,
      "step": 375,
      "time_per_iteration": 3.210738182067871
    },
    {
      "auxiliary_loss_clip": 0.01311712,
      "auxiliary_loss_mlp": 0.01044607,
      "balance_loss_clip": 1.08556843,
      "balance_loss_mlp": 1.02913332,
      "epoch": 0.045211326880298205,
      "flos": 22271115830400.0,
      "grad_norm": 2.6339146033476344,
      "language_loss": 0.77113551,
      "learning_rate": 3.997592106710101e-06,
      "loss": 0.79469872,
      "num_input_tokens_seen": 7993940,
      "step": 376,
      "time_per_iteration": 2.7270381450653076
    },
    {
      "auxiliary_loss_clip": 0.01303294,
      "auxiliary_loss_mlp": 0.01057582,
      "balance_loss_clip": 1.08017671,
      "balance_loss_mlp": 1.04390228,
      "epoch": 0.045331569770937295,
      "flos": 32159441796480.0,
      "grad_norm": 2.9503259823677057,
      "language_loss": 0.65698373,
      "learning_rate": 3.997553742337182e-06,
      "loss": 0.68059248,
      "num_input_tokens_seen": 8013365,
      "step": 377,
      "time_per_iteration": 2.7393596172332764
    },
    {
      "auxiliary_loss_clip": 0.01303226,
      "auxiliary_loss_mlp": 0.01056609,
      "balance_loss_clip": 1.08079541,
      "balance_loss_mlp": 1.04226804,
      "epoch": 0.045451812661576385,
      "flos": 22163455791360.0,
      "grad_norm": 1.700252713545257,
      "language_loss": 0.9139753,
      "learning_rate": 3.997515074936949e-06,
      "loss": 0.93757361,
      "num_input_tokens_seen": 8034240,
      "step": 378,
      "time_per_iteration": 2.7585113048553467
    },
    {
      "auxiliary_loss_clip": 0.01301904,
      "auxiliary_loss_mlp": 0.01057262,
      "balance_loss_clip": 1.08181047,
      "balance_loss_mlp": 1.0431,
      "epoch": 0.045572055552215475,
      "flos": 16581968305920.0,
      "grad_norm": 3.5312804456682088,
      "language_loss": 0.87048078,
      "learning_rate": 3.997476104515268e-06,
      "loss": 0.89407253,
      "num_input_tokens_seen": 8052430,
      "step": 379,
      "time_per_iteration": 2.6487438678741455
    },
    {
      "auxiliary_loss_clip": 0.01299732,
      "auxiliary_loss_mlp": 0.01051479,
      "balance_loss_clip": 1.08306575,
      "balance_loss_mlp": 1.03822279,
      "epoch": 0.045692298442854565,
      "flos": 17603375448960.0,
      "grad_norm": 3.0388815411134473,
      "language_loss": 0.77610886,
      "learning_rate": 3.9974368310780485e-06,
      "loss": 0.79962099,
      "num_input_tokens_seen": 8069605,
      "step": 380,
      "time_per_iteration": 2.6738009452819824
    },
    {
      "auxiliary_loss_clip": 0.01315838,
      "auxiliary_loss_mlp": 0.01064405,
      "balance_loss_clip": 1.08576405,
      "balance_loss_mlp": 1.04735851,
      "epoch": 0.045812541333493655,
      "flos": 26761098781440.0,
      "grad_norm": 2.6643507342995036,
      "language_loss": 0.7409662,
      "learning_rate": 3.997397254631251e-06,
      "loss": 0.7647686,
      "num_input_tokens_seen": 8090225,
      "step": 381,
      "time_per_iteration": 2.6870789527893066
    },
    {
      "auxiliary_loss_clip": 0.01237332,
      "auxiliary_loss_mlp": 0.01005762,
      "balance_loss_clip": 1.08535814,
      "balance_loss_mlp": 1.00197077,
      "epoch": 0.04593278422413275,
      "flos": 60250349894400.0,
      "grad_norm": 0.8272019162785661,
      "language_loss": 0.60112029,
      "learning_rate": 3.997357375180878e-06,
      "loss": 0.62355125,
      "num_input_tokens_seen": 8154505,
      "step": 382,
      "time_per_iteration": 4.2475905418396
    },
    {
      "auxiliary_loss_clip": 0.01305558,
      "auxiliary_loss_mlp": 0.01066474,
      "balance_loss_clip": 1.08298588,
      "balance_loss_mlp": 1.0523119,
      "epoch": 0.04605302711477184,
      "flos": 21799249839360.0,
      "grad_norm": 2.510630372857572,
      "language_loss": 0.75018913,
      "learning_rate": 3.997317192732979e-06,
      "loss": 0.77390945,
      "num_input_tokens_seen": 8173285,
      "step": 383,
      "time_per_iteration": 3.5933191776275635
    },
    {
      "auxiliary_loss_clip": 0.01306454,
      "auxiliary_loss_mlp": 0.01058951,
      "balance_loss_clip": 1.08141565,
      "balance_loss_mlp": 1.04366875,
      "epoch": 0.04617327000541093,
      "flos": 19459705299840.0,
      "grad_norm": 1.9866096528120654,
      "language_loss": 0.82302672,
      "learning_rate": 3.99727670729365e-06,
      "loss": 0.84668076,
      "num_input_tokens_seen": 8191845,
      "step": 384,
      "time_per_iteration": 2.6424667835235596
    },
    {
      "auxiliary_loss_clip": 0.0130313,
      "auxiliary_loss_mlp": 0.01063037,
      "balance_loss_clip": 1.08451939,
      "balance_loss_mlp": 1.04822564,
      "epoch": 0.04629351289605002,
      "flos": 25411468135680.0,
      "grad_norm": 2.793857661552962,
      "language_loss": 0.78014314,
      "learning_rate": 3.997235918869033e-06,
      "loss": 0.80380481,
      "num_input_tokens_seen": 8212880,
      "step": 385,
      "time_per_iteration": 2.7229650020599365
    },
    {
      "auxiliary_loss_clip": 0.01299555,
      "auxiliary_loss_mlp": 0.01049934,
      "balance_loss_clip": 1.08111668,
      "balance_loss_mlp": 1.03593838,
      "epoch": 0.04641375578668911,
      "flos": 20558284813440.0,
      "grad_norm": 2.064704151651639,
      "language_loss": 0.82733679,
      "learning_rate": 3.997194827465315e-06,
      "loss": 0.85083175,
      "num_input_tokens_seen": 8231475,
      "step": 386,
      "time_per_iteration": 3.6414692401885986
    },
    {
      "auxiliary_loss_clip": 0.01303267,
      "auxiliary_loss_mlp": 0.01055681,
      "balance_loss_clip": 1.08253372,
      "balance_loss_mlp": 1.0424366,
      "epoch": 0.0465339986773282,
      "flos": 13188661447680.0,
      "grad_norm": 3.2801590645859444,
      "language_loss": 0.9128201,
      "learning_rate": 3.997153433088728e-06,
      "loss": 0.93640959,
      "num_input_tokens_seen": 8248600,
      "step": 387,
      "time_per_iteration": 2.63128399848938
    },
    {
      "auxiliary_loss_clip": 0.01305161,
      "auxiliary_loss_mlp": 0.01052536,
      "balance_loss_clip": 1.08091116,
      "balance_loss_mlp": 1.03869605,
      "epoch": 0.0466542415679673,
      "flos": 25556547168000.0,
      "grad_norm": 2.2350831995675082,
      "language_loss": 0.81206566,
      "learning_rate": 3.997111735745554e-06,
      "loss": 0.83564264,
      "num_input_tokens_seen": 8271570,
      "step": 388,
      "time_per_iteration": 3.922584056854248
    },
    {
      "auxiliary_loss_clip": 0.01306798,
      "auxiliary_loss_mlp": 0.01048335,
      "balance_loss_clip": 1.08403635,
      "balance_loss_mlp": 1.03411281,
      "epoch": 0.04677448445860639,
      "flos": 22236749493120.0,
      "grad_norm": 1.7982286653002377,
      "language_loss": 0.82561862,
      "learning_rate": 3.997069735442118e-06,
      "loss": 0.84916997,
      "num_input_tokens_seen": 8291265,
      "step": 389,
      "time_per_iteration": 2.6688811779022217
    },
    {
      "auxiliary_loss_clip": 0.01297453,
      "auxiliary_loss_mlp": 0.0105441,
      "balance_loss_clip": 1.08062851,
      "balance_loss_mlp": 1.04036689,
      "epoch": 0.04689472734924548,
      "flos": 28147825198080.0,
      "grad_norm": 1.4761750189362786,
      "language_loss": 0.80297285,
      "learning_rate": 3.997027432184792e-06,
      "loss": 0.82649148,
      "num_input_tokens_seen": 8315925,
      "step": 390,
      "time_per_iteration": 2.7295639514923096
    },
    {
      "auxiliary_loss_clip": 0.01302496,
      "auxiliary_loss_mlp": 0.01047206,
      "balance_loss_clip": 1.08204091,
      "balance_loss_mlp": 1.03318679,
      "epoch": 0.04701497023988457,
      "flos": 23148952312320.0,
      "grad_norm": 2.1634375387020257,
      "language_loss": 0.893435,
      "learning_rate": 3.99698482597999e-06,
      "loss": 0.91693205,
      "num_input_tokens_seen": 8333605,
      "step": 391,
      "time_per_iteration": 2.7145674228668213
    },
    {
      "auxiliary_loss_clip": 0.01226638,
      "auxiliary_loss_mlp": 0.01002428,
      "balance_loss_clip": 1.07853508,
      "balance_loss_mlp": 0.9986375,
      "epoch": 0.04713521313052366,
      "flos": 64827668764800.0,
      "grad_norm": 0.8688145392703843,
      "language_loss": 0.63935566,
      "learning_rate": 3.99694191683418e-06,
      "loss": 0.66164637,
      "num_input_tokens_seen": 8394405,
      "step": 392,
      "time_per_iteration": 3.3468165397644043
    },
    {
      "auxiliary_loss_clip": 0.01308938,
      "auxiliary_loss_mlp": 0.01057363,
      "balance_loss_clip": 1.08629978,
      "balance_loss_mlp": 1.04184222,
      "epoch": 0.047255456021162746,
      "flos": 18771585477120.0,
      "grad_norm": 4.656468174349219,
      "language_loss": 0.81636757,
      "learning_rate": 3.996898704753867e-06,
      "loss": 0.84003055,
      "num_input_tokens_seen": 8412355,
      "step": 393,
      "time_per_iteration": 2.7445671558380127
    },
    {
      "auxiliary_loss_clip": 0.01303846,
      "auxiliary_loss_mlp": 0.01049867,
      "balance_loss_clip": 1.07951319,
      "balance_loss_mlp": 1.03603911,
      "epoch": 0.04737569891180184,
      "flos": 22053820504320.0,
      "grad_norm": 2.2455089186162818,
      "language_loss": 0.87454879,
      "learning_rate": 3.996855189745609e-06,
      "loss": 0.89808583,
      "num_input_tokens_seen": 8431620,
      "step": 394,
      "time_per_iteration": 2.863042116165161
    },
    {
      "auxiliary_loss_clip": 0.01299552,
      "auxiliary_loss_mlp": 0.01060217,
      "balance_loss_clip": 1.07922339,
      "balance_loss_mlp": 1.04650211,
      "epoch": 0.04749594180244093,
      "flos": 29057370410880.0,
      "grad_norm": 1.9486256311046461,
      "language_loss": 0.9241792,
      "learning_rate": 3.996811371816007e-06,
      "loss": 0.94777679,
      "num_input_tokens_seen": 8454045,
      "step": 395,
      "time_per_iteration": 2.8059115409851074
    },
    {
      "auxiliary_loss_clip": 0.01300213,
      "auxiliary_loss_mlp": 0.0105218,
      "balance_loss_clip": 1.08230138,
      "balance_loss_mlp": 1.03908515,
      "epoch": 0.04761618469308002,
      "flos": 35112268172160.0,
      "grad_norm": 1.8461607443413453,
      "language_loss": 0.77558023,
      "learning_rate": 3.996767250971707e-06,
      "loss": 0.79910421,
      "num_input_tokens_seen": 8476785,
      "step": 396,
      "time_per_iteration": 2.7634549140930176
    },
    {
      "auxiliary_loss_clip": 0.01303893,
      "auxiliary_loss_mlp": 0.01047128,
      "balance_loss_clip": 1.08297515,
      "balance_loss_mlp": 1.03335333,
      "epoch": 0.04773642758371911,
      "flos": 25630702796160.0,
      "grad_norm": 2.360876605623183,
      "language_loss": 0.86833811,
      "learning_rate": 3.996722827219403e-06,
      "loss": 0.89184833,
      "num_input_tokens_seen": 8498400,
      "step": 397,
      "time_per_iteration": 2.737384080886841
    },
    {
      "auxiliary_loss_clip": 0.01309283,
      "auxiliary_loss_mlp": 0.01062135,
      "balance_loss_clip": 1.08520222,
      "balance_loss_mlp": 1.04843247,
      "epoch": 0.0478566704743582,
      "flos": 20631506688000.0,
      "grad_norm": 2.298009456175315,
      "language_loss": 0.82512242,
      "learning_rate": 3.996678100565833e-06,
      "loss": 0.8488366,
      "num_input_tokens_seen": 8517455,
      "step": 398,
      "time_per_iteration": 2.720529794692993
    },
    {
      "auxiliary_loss_clip": 0.01302348,
      "auxiliary_loss_mlp": 0.01057985,
      "balance_loss_clip": 1.08234978,
      "balance_loss_mlp": 1.04391241,
      "epoch": 0.04797691336499729,
      "flos": 18835721210880.0,
      "grad_norm": 2.8461629438695395,
      "language_loss": 0.88654321,
      "learning_rate": 3.996633071017783e-06,
      "loss": 0.91014659,
      "num_input_tokens_seen": 8534085,
      "step": 399,
      "time_per_iteration": 2.7168006896972656
    },
    {
      "auxiliary_loss_clip": 0.01302522,
      "auxiliary_loss_mlp": 0.01051207,
      "balance_loss_clip": 1.0822984,
      "balance_loss_mlp": 1.03783154,
      "epoch": 0.04809715625563638,
      "flos": 21099673578240.0,
      "grad_norm": 3.0148563663759838,
      "language_loss": 0.8162477,
      "learning_rate": 3.996587738582084e-06,
      "loss": 0.83978498,
      "num_input_tokens_seen": 8550885,
      "step": 400,
      "time_per_iteration": 2.804020404815674
    },
    {
      "auxiliary_loss_clip": 0.01299091,
      "auxiliary_loss_mlp": 0.01047174,
      "balance_loss_clip": 1.08003986,
      "balance_loss_mlp": 1.0333811,
      "epoch": 0.04821739914627548,
      "flos": 23805650712960.0,
      "grad_norm": 2.3023486552959738,
      "language_loss": 0.86124027,
      "learning_rate": 3.9965421032656115e-06,
      "loss": 0.88470292,
      "num_input_tokens_seen": 8570815,
      "step": 401,
      "time_per_iteration": 2.7791588306427
    },
    {
      "auxiliary_loss_clip": 0.01303628,
      "auxiliary_loss_mlp": 0.0104477,
      "balance_loss_clip": 1.08378816,
      "balance_loss_mlp": 1.03107905,
      "epoch": 0.04833764203691457,
      "flos": 22200587475840.0,
      "grad_norm": 2.806805159694416,
      "language_loss": 0.93771791,
      "learning_rate": 3.99649616507529e-06,
      "loss": 0.96120191,
      "num_input_tokens_seen": 8589910,
      "step": 402,
      "time_per_iteration": 2.8102262020111084
    },
    {
      "auxiliary_loss_clip": 0.01225559,
      "auxiliary_loss_mlp": 0.01006421,
      "balance_loss_clip": 1.07556558,
      "balance_loss_mlp": 1.00191462,
      "epoch": 0.04845788492755366,
      "flos": 65904376896000.0,
      "grad_norm": 0.8940606838158041,
      "language_loss": 0.63193423,
      "learning_rate": 3.996449924018088e-06,
      "loss": 0.65425402,
      "num_input_tokens_seen": 8650370,
      "step": 403,
      "time_per_iteration": 3.2153587341308594
    },
    {
      "auxiliary_loss_clip": 0.01292717,
      "auxiliary_loss_mlp": 0.01051253,
      "balance_loss_clip": 1.07979178,
      "balance_loss_mlp": 1.03717446,
      "epoch": 0.04857812781819275,
      "flos": 19281301424640.0,
      "grad_norm": 1.9178217905059813,
      "language_loss": 0.7962895,
      "learning_rate": 3.99640338010102e-06,
      "loss": 0.81972921,
      "num_input_tokens_seen": 8669475,
      "step": 404,
      "time_per_iteration": 2.8024115562438965
    },
    {
      "auxiliary_loss_clip": 0.01294774,
      "auxiliary_loss_mlp": 0.0105714,
      "balance_loss_clip": 1.07887554,
      "balance_loss_mlp": 1.04385984,
      "epoch": 0.04869837070883184,
      "flos": 24062376193920.0,
      "grad_norm": 1.7654195265383612,
      "language_loss": 0.78397721,
      "learning_rate": 3.996356533331146e-06,
      "loss": 0.80749631,
      "num_input_tokens_seen": 8691345,
      "step": 405,
      "time_per_iteration": 2.7982988357543945
    },
    {
      "auxiliary_loss_clip": 0.01307778,
      "auxiliary_loss_mlp": 0.01047884,
      "balance_loss_clip": 1.07953382,
      "balance_loss_mlp": 1.03390682,
      "epoch": 0.04881861359947093,
      "flos": 25187169657600.0,
      "grad_norm": 3.1696760448133263,
      "language_loss": 0.61860347,
      "learning_rate": 3.996309383715573e-06,
      "loss": 0.64216006,
      "num_input_tokens_seen": 8710125,
      "step": 406,
      "time_per_iteration": 2.852543592453003
    },
    {
      "auxiliary_loss_clip": 0.01298436,
      "auxiliary_loss_mlp": 0.01052163,
      "balance_loss_clip": 1.08044326,
      "balance_loss_mlp": 1.03843009,
      "epoch": 0.048938856490110025,
      "flos": 16362913213440.0,
      "grad_norm": 2.03284004679079,
      "language_loss": 0.73821557,
      "learning_rate": 3.996261931261454e-06,
      "loss": 0.76172155,
      "num_input_tokens_seen": 8728705,
      "step": 407,
      "time_per_iteration": 2.757646083831787
    },
    {
      "auxiliary_loss_clip": 0.0130223,
      "auxiliary_loss_mlp": 0.01055854,
      "balance_loss_clip": 1.08393884,
      "balance_loss_mlp": 1.04143596,
      "epoch": 0.049059099380749115,
      "flos": 29895094379520.0,
      "grad_norm": 1.9708803354298638,
      "language_loss": 0.86447179,
      "learning_rate": 3.996214175975987e-06,
      "loss": 0.88805258,
      "num_input_tokens_seen": 8749225,
      "step": 408,
      "time_per_iteration": 3.8407833576202393
    },
    {
      "auxiliary_loss_clip": 0.01306893,
      "auxiliary_loss_mlp": 0.01044942,
      "balance_loss_clip": 1.0853188,
      "balance_loss_mlp": 1.03178167,
      "epoch": 0.049179342271388204,
      "flos": 35918858027520.0,
      "grad_norm": 2.539963397182782,
      "language_loss": 0.78981268,
      "learning_rate": 3.996166117866417e-06,
      "loss": 0.81333101,
      "num_input_tokens_seen": 8771160,
      "step": 409,
      "time_per_iteration": 2.957381248474121
    },
    {
      "auxiliary_loss_clip": 0.01291089,
      "auxiliary_loss_mlp": 0.01060222,
      "balance_loss_clip": 1.07647276,
      "balance_loss_mlp": 1.04672134,
      "epoch": 0.049299585162027294,
      "flos": 14611226659200.0,
      "grad_norm": 2.0739347436710034,
      "language_loss": 0.86854577,
      "learning_rate": 3.996117756940035e-06,
      "loss": 0.89205885,
      "num_input_tokens_seen": 8787845,
      "step": 410,
      "time_per_iteration": 3.9792325496673584
    },
    {
      "auxiliary_loss_clip": 0.01302285,
      "auxiliary_loss_mlp": 0.01046476,
      "balance_loss_clip": 1.08278239,
      "balance_loss_mlp": 1.03255868,
      "epoch": 0.049419828052666384,
      "flos": 19567939956480.0,
      "grad_norm": 3.5734240868914893,
      "language_loss": 0.97706866,
      "learning_rate": 3.996069093204175e-06,
      "loss": 1.00055623,
      "num_input_tokens_seen": 8803805,
      "step": 411,
      "time_per_iteration": 2.780336618423462
    },
    {
      "auxiliary_loss_clip": 0.01308218,
      "auxiliary_loss_mlp": 0.01055178,
      "balance_loss_clip": 1.08717453,
      "balance_loss_mlp": 1.04155207,
      "epoch": 0.049540070943305474,
      "flos": 13659916907520.0,
      "grad_norm": 2.541651457573741,
      "language_loss": 0.87735403,
      "learning_rate": 3.996020126666221e-06,
      "loss": 0.90098804,
      "num_input_tokens_seen": 8820785,
      "step": 412,
      "time_per_iteration": 2.8813061714172363
    },
    {
      "auxiliary_loss_clip": 0.0129789,
      "auxiliary_loss_mlp": 0.01046921,
      "balance_loss_clip": 1.08190477,
      "balance_loss_mlp": 1.0333668,
      "epoch": 0.04966031383394457,
      "flos": 21832035978240.0,
      "grad_norm": 1.9566419092181626,
      "language_loss": 0.82001114,
      "learning_rate": 3.995970857333601e-06,
      "loss": 0.84345925,
      "num_input_tokens_seen": 8841195,
      "step": 413,
      "time_per_iteration": 3.6949756145477295
    },
    {
      "auxiliary_loss_clip": 0.01296424,
      "auxiliary_loss_mlp": 0.01055976,
      "balance_loss_clip": 1.07791817,
      "balance_loss_mlp": 1.04226708,
      "epoch": 0.04978055672458366,
      "flos": 28618793349120.0,
      "grad_norm": 2.3952816675895487,
      "language_loss": 0.79740793,
      "learning_rate": 3.995921285213789e-06,
      "loss": 0.82093191,
      "num_input_tokens_seen": 8861455,
      "step": 414,
      "time_per_iteration": 3.7385382652282715
    },
    {
      "auxiliary_loss_clip": 0.01301266,
      "auxiliary_loss_mlp": 0.010514,
      "balance_loss_clip": 1.08418226,
      "balance_loss_mlp": 1.03772092,
      "epoch": 0.04990079961522275,
      "flos": 19828220883840.0,
      "grad_norm": 2.32281838768489,
      "language_loss": 0.80557746,
      "learning_rate": 3.995871410314305e-06,
      "loss": 0.82910413,
      "num_input_tokens_seen": 8880015,
      "step": 415,
      "time_per_iteration": 2.788475513458252
    },
    {
      "auxiliary_loss_clip": 0.01191978,
      "auxiliary_loss_mlp": 0.01007733,
      "balance_loss_clip": 1.06921959,
      "balance_loss_mlp": 1.0034411,
      "epoch": 0.05002104250586184,
      "flos": 62735045293440.0,
      "grad_norm": 0.9053118944899345,
      "language_loss": 0.59651512,
      "learning_rate": 3.995821232642714e-06,
      "loss": 0.61851221,
      "num_input_tokens_seen": 8938420,
      "step": 416,
      "time_per_iteration": 3.3809609413146973
    },
    {
      "auxiliary_loss_clip": 0.01264702,
      "auxiliary_loss_mlp": 0.01047453,
      "balance_loss_clip": 1.07631612,
      "balance_loss_mlp": 1.03480506,
      "epoch": 0.05014128539650093,
      "flos": 27928518710400.0,
      "grad_norm": 2.192751814065702,
      "language_loss": 0.82565057,
      "learning_rate": 3.995770752206629e-06,
      "loss": 0.84877217,
      "num_input_tokens_seen": 8959495,
      "step": 417,
      "time_per_iteration": 2.737752676010132
    },
    {
      "auxiliary_loss_clip": 0.01249335,
      "auxiliary_loss_mlp": 0.00767153,
      "balance_loss_clip": 1.07315457,
      "balance_loss_mlp": 1.00081491,
      "epoch": 0.05026152828714002,
      "flos": 17705576620800.0,
      "grad_norm": 1.9737500114258117,
      "language_loss": 0.96979439,
      "learning_rate": 3.995719969013709e-06,
      "loss": 0.98995936,
      "num_input_tokens_seen": 8976675,
      "step": 418,
      "time_per_iteration": 2.782500982284546
    },
    {
      "auxiliary_loss_clip": 0.01216157,
      "auxiliary_loss_mlp": 0.01043101,
      "balance_loss_clip": 1.06735301,
      "balance_loss_mlp": 1.02974367,
      "epoch": 0.05038177117777912,
      "flos": 19133277477120.0,
      "grad_norm": 3.1826477938136524,
      "language_loss": 0.86121124,
      "learning_rate": 3.995668883071655e-06,
      "loss": 0.88380373,
      "num_input_tokens_seen": 8992900,
      "step": 419,
      "time_per_iteration": 2.7924909591674805
    },
    {
      "auxiliary_loss_clip": 0.01216747,
      "auxiliary_loss_mlp": 0.01046069,
      "balance_loss_clip": 1.07008219,
      "balance_loss_mlp": 1.03181171,
      "epoch": 0.050502014068418206,
      "flos": 20667704618880.0,
      "grad_norm": 2.3684104482641306,
      "language_loss": 0.90787733,
      "learning_rate": 3.995617494388219e-06,
      "loss": 0.93050551,
      "num_input_tokens_seen": 9011020,
      "step": 420,
      "time_per_iteration": 2.8883655071258545
    },
    {
      "auxiliary_loss_clip": 0.01251536,
      "auxiliary_loss_mlp": 0.01048465,
      "balance_loss_clip": 1.07424045,
      "balance_loss_mlp": 1.03385043,
      "epoch": 0.050622256959057296,
      "flos": 21361103740800.0,
      "grad_norm": 1.934695164873366,
      "language_loss": 0.80531216,
      "learning_rate": 3.995565802971196e-06,
      "loss": 0.82831216,
      "num_input_tokens_seen": 9030995,
      "step": 421,
      "time_per_iteration": 2.770827054977417
    },
    {
      "auxiliary_loss_clip": 0.01229626,
      "auxiliary_loss_mlp": 0.01059305,
      "balance_loss_clip": 1.07289386,
      "balance_loss_mlp": 1.04555392,
      "epoch": 0.050742499849696386,
      "flos": 27673588909440.0,
      "grad_norm": 1.8922434497552603,
      "language_loss": 0.67338943,
      "learning_rate": 3.995513808828427e-06,
      "loss": 0.69627875,
      "num_input_tokens_seen": 9053790,
      "step": 422,
      "time_per_iteration": 2.934518575668335
    },
    {
      "auxiliary_loss_clip": 0.01254995,
      "auxiliary_loss_mlp": 0.01055182,
      "balance_loss_clip": 1.07427216,
      "balance_loss_mlp": 1.04211044,
      "epoch": 0.050862742740335476,
      "flos": 19865999013120.0,
      "grad_norm": 1.817579069739612,
      "language_loss": 0.76531684,
      "learning_rate": 3.9954615119678e-06,
      "loss": 0.78841865,
      "num_input_tokens_seen": 9072345,
      "step": 423,
      "time_per_iteration": 2.8514490127563477
    },
    {
      "auxiliary_loss_clip": 0.0122921,
      "auxiliary_loss_mlp": 0.01042334,
      "balance_loss_clip": 1.07535827,
      "balance_loss_mlp": 1.02907801,
      "epoch": 0.050982985630974566,
      "flos": 22085098272000.0,
      "grad_norm": 2.0598129223556394,
      "language_loss": 0.80400193,
      "learning_rate": 3.995408912397248e-06,
      "loss": 0.82671738,
      "num_input_tokens_seen": 9090240,
      "step": 424,
      "time_per_iteration": 2.8033599853515625
    },
    {
      "auxiliary_loss_clip": 0.01244067,
      "auxiliary_loss_mlp": 0.01054785,
      "balance_loss_clip": 1.07289267,
      "balance_loss_mlp": 1.04110587,
      "epoch": 0.05110322852161366,
      "flos": 20740962407040.0,
      "grad_norm": 2.2755180055278634,
      "language_loss": 0.93262291,
      "learning_rate": 3.99535601012475e-06,
      "loss": 0.95561147,
      "num_input_tokens_seen": 9105570,
      "step": 425,
      "time_per_iteration": 2.759328603744507
    },
    {
      "auxiliary_loss_clip": 0.01260822,
      "auxiliary_loss_mlp": 0.01059184,
      "balance_loss_clip": 1.07662225,
      "balance_loss_mlp": 1.04692936,
      "epoch": 0.05122347141225275,
      "flos": 28547295327360.0,
      "grad_norm": 1.6720334014492262,
      "language_loss": 0.75446081,
      "learning_rate": 3.995302805158333e-06,
      "loss": 0.77766085,
      "num_input_tokens_seen": 9128225,
      "step": 426,
      "time_per_iteration": 2.797907829284668
    },
    {
      "auxiliary_loss_clip": 0.01301512,
      "auxiliary_loss_mlp": 0.0104579,
      "balance_loss_clip": 1.08100164,
      "balance_loss_mlp": 1.03115129,
      "epoch": 0.05134371430289184,
      "flos": 19722679747200.0,
      "grad_norm": 1.8591051278952873,
      "language_loss": 0.83571732,
      "learning_rate": 3.9952492975060665e-06,
      "loss": 0.8591904,
      "num_input_tokens_seen": 9148295,
      "step": 427,
      "time_per_iteration": 2.7395427227020264
    },
    {
      "auxiliary_loss_clip": 0.01261298,
      "auxiliary_loss_mlp": 0.01045446,
      "balance_loss_clip": 1.07688451,
      "balance_loss_mlp": 1.03073561,
      "epoch": 0.05146395719353093,
      "flos": 34458945649920.0,
      "grad_norm": 2.174255889064568,
      "language_loss": 0.84829247,
      "learning_rate": 3.995195487176067e-06,
      "loss": 0.87135994,
      "num_input_tokens_seen": 9168525,
      "step": 428,
      "time_per_iteration": 2.7669336795806885
    },
    {
      "auxiliary_loss_clip": 0.01298145,
      "auxiliary_loss_mlp": 0.01043878,
      "balance_loss_clip": 1.08056092,
      "balance_loss_mlp": 1.03125381,
      "epoch": 0.05158420008417002,
      "flos": 21760286561280.0,
      "grad_norm": 2.071236349687691,
      "language_loss": 0.85507381,
      "learning_rate": 3.995141374176499e-06,
      "loss": 0.87849402,
      "num_input_tokens_seen": 9186920,
      "step": 429,
      "time_per_iteration": 2.6871206760406494
    },
    {
      "auxiliary_loss_clip": 0.01149028,
      "auxiliary_loss_mlp": 0.01004841,
      "balance_loss_clip": 1.05229282,
      "balance_loss_mlp": 1.00124061,
      "epoch": 0.05170444297480911,
      "flos": 72553956226560.0,
      "grad_norm": 0.8763389387103186,
      "language_loss": 0.63106126,
      "learning_rate": 3.995086958515572e-06,
      "loss": 0.65259993,
      "num_input_tokens_seen": 9244940,
      "step": 430,
      "time_per_iteration": 3.337149143218994
    },
    {
      "auxiliary_loss_clip": 0.0117305,
      "auxiliary_loss_mlp": 0.01003966,
      "balance_loss_clip": 1.06000757,
      "balance_loss_mlp": 1.00043714,
      "epoch": 0.05182468586544821,
      "flos": 62416159326720.0,
      "grad_norm": 0.8618634829108771,
      "language_loss": 0.59957814,
      "learning_rate": 3.995032240201538e-06,
      "loss": 0.62134826,
      "num_input_tokens_seen": 9307335,
      "step": 431,
      "time_per_iteration": 3.2603416442871094
    },
    {
      "auxiliary_loss_clip": 0.01132838,
      "auxiliary_loss_mlp": 0.01002949,
      "balance_loss_clip": 1.05508614,
      "balance_loss_mlp": 0.99932492,
      "epoch": 0.0519449287560873,
      "flos": 41225989432320.0,
      "grad_norm": 0.9344344721486582,
      "language_loss": 0.631271,
      "learning_rate": 3.9949772192427e-06,
      "loss": 0.6526289,
      "num_input_tokens_seen": 9353960,
      "step": 432,
      "time_per_iteration": 3.605565071105957
    },
    {
      "auxiliary_loss_clip": 0.01198378,
      "auxiliary_loss_mlp": 0.01048763,
      "balance_loss_clip": 1.06697989,
      "balance_loss_mlp": 1.0351913,
      "epoch": 0.05206517164672639,
      "flos": 17494530261120.0,
      "grad_norm": 1.8292718256128941,
      "language_loss": 0.79646552,
      "learning_rate": 3.994921895647405e-06,
      "loss": 0.81893694,
      "num_input_tokens_seen": 9372130,
      "step": 433,
      "time_per_iteration": 3.3710289001464844
    },
    {
      "auxiliary_loss_clip": 0.01206258,
      "auxiliary_loss_mlp": 0.01003181,
      "balance_loss_clip": 1.0632658,
      "balance_loss_mlp": 0.99958116,
      "epoch": 0.05218541453736548,
      "flos": 64002762973440.0,
      "grad_norm": 0.8490532510759982,
      "language_loss": 0.55342758,
      "learning_rate": 3.994866269424043e-06,
      "loss": 0.57552195,
      "num_input_tokens_seen": 9428500,
      "step": 434,
      "time_per_iteration": 3.1670472621917725
    },
    {
      "auxiliary_loss_clip": 0.01229722,
      "auxiliary_loss_mlp": 0.01054038,
      "balance_loss_clip": 1.06572723,
      "balance_loss_mlp": 1.04015064,
      "epoch": 0.05230565742800457,
      "flos": 19317319787520.0,
      "grad_norm": 2.294857939719833,
      "language_loss": 0.78339857,
      "learning_rate": 3.9948103405810545e-06,
      "loss": 0.80623615,
      "num_input_tokens_seen": 9447450,
      "step": 435,
      "time_per_iteration": 3.7108261585235596
    },
    {
      "auxiliary_loss_clip": 0.01235813,
      "auxiliary_loss_mlp": 0.01061251,
      "balance_loss_clip": 1.06915009,
      "balance_loss_mlp": 1.04821587,
      "epoch": 0.05242590031864366,
      "flos": 25298636538240.0,
      "grad_norm": 1.8461769133150499,
      "language_loss": 0.85539401,
      "learning_rate": 3.994754109126923e-06,
      "loss": 0.87836468,
      "num_input_tokens_seen": 9468945,
      "step": 436,
      "time_per_iteration": 3.769427537918091
    },
    {
      "auxiliary_loss_clip": 0.01207031,
      "auxiliary_loss_mlp": 0.01056033,
      "balance_loss_clip": 1.06850767,
      "balance_loss_mlp": 1.04351008,
      "epoch": 0.052546143209282754,
      "flos": 26211629456640.0,
      "grad_norm": 10.028668598564568,
      "language_loss": 0.93421221,
      "learning_rate": 3.994697575070181e-06,
      "loss": 0.95684278,
      "num_input_tokens_seen": 9488405,
      "step": 437,
      "time_per_iteration": 2.8796656131744385
    },
    {
      "auxiliary_loss_clip": 0.01249864,
      "auxiliary_loss_mlp": 0.01051432,
      "balance_loss_clip": 1.07710934,
      "balance_loss_mlp": 1.03822327,
      "epoch": 0.052666386099921844,
      "flos": 22158140578560.0,
      "grad_norm": 1.8261651793402376,
      "language_loss": 0.91616976,
      "learning_rate": 3.994640738419402e-06,
      "loss": 0.93918264,
      "num_input_tokens_seen": 9507780,
      "step": 438,
      "time_per_iteration": 3.6649110317230225
    },
    {
      "auxiliary_loss_clip": 0.01246844,
      "auxiliary_loss_mlp": 0.00766317,
      "balance_loss_clip": 1.07273006,
      "balance_loss_mlp": 1.00064349,
      "epoch": 0.052786628990560934,
      "flos": 23881817502720.0,
      "grad_norm": 1.923740111600316,
      "language_loss": 0.80925816,
      "learning_rate": 3.9945835991832075e-06,
      "loss": 0.82938975,
      "num_input_tokens_seen": 9529665,
      "step": 439,
      "time_per_iteration": 2.8421056270599365
    },
    {
      "auxiliary_loss_clip": 0.01282018,
      "auxiliary_loss_mlp": 0.00765776,
      "balance_loss_clip": 1.08139694,
      "balance_loss_mlp": 1.0005939,
      "epoch": 0.052906871881200024,
      "flos": 24605021934720.0,
      "grad_norm": 2.348333718543125,
      "language_loss": 0.92772901,
      "learning_rate": 3.994526157370268e-06,
      "loss": 0.94820702,
      "num_input_tokens_seen": 9548280,
      "step": 440,
      "time_per_iteration": 3.7695653438568115
    },
    {
      "auxiliary_loss_clip": 0.01105102,
      "auxiliary_loss_mlp": 0.01008993,
      "balance_loss_clip": 1.04353333,
      "balance_loss_mlp": 1.00539255,
      "epoch": 0.053027114771839114,
      "flos": 56461631143680.0,
      "grad_norm": 0.98809924932657,
      "language_loss": 0.5928055,
      "learning_rate": 3.994468412989296e-06,
      "loss": 0.61394644,
      "num_input_tokens_seen": 9609690,
      "step": 441,
      "time_per_iteration": 4.041775465011597
    },
    {
      "auxiliary_loss_clip": 0.01226934,
      "auxiliary_loss_mlp": 0.01055952,
      "balance_loss_clip": 1.06848693,
      "balance_loss_mlp": 1.04277372,
      "epoch": 0.053147357662478203,
      "flos": 17311098481920.0,
      "grad_norm": 3.2032865542138995,
      "language_loss": 0.92679691,
      "learning_rate": 3.994410366049052e-06,
      "loss": 0.94962585,
      "num_input_tokens_seen": 9627550,
      "step": 442,
      "time_per_iteration": 3.091188669204712
    },
    {
      "auxiliary_loss_clip": 0.01279137,
      "auxiliary_loss_mlp": 0.01049624,
      "balance_loss_clip": 1.07708251,
      "balance_loss_mlp": 1.03550982,
      "epoch": 0.0532676005531173,
      "flos": 17164977955200.0,
      "grad_norm": 2.5026731866085936,
      "language_loss": 0.83197701,
      "learning_rate": 3.994352016558341e-06,
      "loss": 0.85526454,
      "num_input_tokens_seen": 9644855,
      "step": 443,
      "time_per_iteration": 2.7488584518432617
    },
    {
      "auxiliary_loss_clip": 0.01253325,
      "auxiliary_loss_mlp": 0.01051676,
      "balance_loss_clip": 1.07544339,
      "balance_loss_mlp": 1.03889632,
      "epoch": 0.05338784344375639,
      "flos": 27819960831360.0,
      "grad_norm": 2.8616851447591323,
      "language_loss": 0.73892146,
      "learning_rate": 3.994293364526014e-06,
      "loss": 0.76197147,
      "num_input_tokens_seen": 9665740,
      "step": 444,
      "time_per_iteration": 2.8398687839508057
    },
    {
      "auxiliary_loss_clip": 0.01252185,
      "auxiliary_loss_mlp": 0.01054405,
      "balance_loss_clip": 1.0787468,
      "balance_loss_mlp": 1.04109478,
      "epoch": 0.05350808633439548,
      "flos": 21507691144320.0,
      "grad_norm": 1.935335225758326,
      "language_loss": 0.84737128,
      "learning_rate": 3.99423440996097e-06,
      "loss": 0.87043715,
      "num_input_tokens_seen": 9685280,
      "step": 445,
      "time_per_iteration": 2.754835844039917
    },
    {
      "auxiliary_loss_clip": 0.01250694,
      "auxiliary_loss_mlp": 0.01055103,
      "balance_loss_clip": 1.07215524,
      "balance_loss_mlp": 1.04225791,
      "epoch": 0.05362832922503457,
      "flos": 20084299920000.0,
      "grad_norm": 3.3514393988049953,
      "language_loss": 0.81659889,
      "learning_rate": 3.994175152872152e-06,
      "loss": 0.83965683,
      "num_input_tokens_seen": 9704365,
      "step": 446,
      "time_per_iteration": 2.9133408069610596
    },
    {
      "auxiliary_loss_clip": 0.01161675,
      "auxiliary_loss_mlp": 0.01053225,
      "balance_loss_clip": 1.06331253,
      "balance_loss_mlp": 1.04050541,
      "epoch": 0.05374857211567366,
      "flos": 26137222433280.0,
      "grad_norm": 1.9894290763258446,
      "language_loss": 0.78565848,
      "learning_rate": 3.994115593268548e-06,
      "loss": 0.80780751,
      "num_input_tokens_seen": 9724145,
      "step": 447,
      "time_per_iteration": 2.984713554382324
    },
    {
      "auxiliary_loss_clip": 0.01274173,
      "auxiliary_loss_mlp": 0.01049952,
      "balance_loss_clip": 1.076805,
      "balance_loss_mlp": 1.03679097,
      "epoch": 0.05386881500631275,
      "flos": 27486817165440.0,
      "grad_norm": 2.0807198734422667,
      "language_loss": 0.8194797,
      "learning_rate": 3.994055731159195e-06,
      "loss": 0.84272099,
      "num_input_tokens_seen": 9741615,
      "step": 448,
      "time_per_iteration": 2.754833936691284
    },
    {
      "auxiliary_loss_clip": 0.01266695,
      "auxiliary_loss_mlp": 0.01052837,
      "balance_loss_clip": 1.07710242,
      "balance_loss_mlp": 1.04039764,
      "epoch": 0.053989057896951846,
      "flos": 23585087249280.0,
      "grad_norm": 2.067813553540972,
      "language_loss": 0.86992043,
      "learning_rate": 3.993995566553172e-06,
      "loss": 0.89311576,
      "num_input_tokens_seen": 9760580,
      "step": 449,
      "time_per_iteration": 2.7904720306396484
    },
    {
      "auxiliary_loss_clip": 0.01274481,
      "auxiliary_loss_mlp": 0.01055757,
      "balance_loss_clip": 1.07662392,
      "balance_loss_mlp": 1.0416311,
      "epoch": 0.054109300787590936,
      "flos": 25228862369280.0,
      "grad_norm": 1.6188396276574197,
      "language_loss": 0.77261066,
      "learning_rate": 3.993935099459607e-06,
      "loss": 0.7959131,
      "num_input_tokens_seen": 9782195,
      "step": 450,
      "time_per_iteration": 2.755995750427246
    },
    {
      "auxiliary_loss_clip": 0.01256033,
      "auxiliary_loss_mlp": 0.01046298,
      "balance_loss_clip": 1.07364225,
      "balance_loss_mlp": 1.033638,
      "epoch": 0.054229543678230026,
      "flos": 23841525421440.0,
      "grad_norm": 1.9219555587340924,
      "language_loss": 0.73675978,
      "learning_rate": 3.993874329887673e-06,
      "loss": 0.75978315,
      "num_input_tokens_seen": 9800850,
      "step": 451,
      "time_per_iteration": 2.8470308780670166
    },
    {
      "auxiliary_loss_clip": 0.01293111,
      "auxiliary_loss_mlp": 0.01049838,
      "balance_loss_clip": 1.07872689,
      "balance_loss_mlp": 1.03663564,
      "epoch": 0.054349786568869116,
      "flos": 16320933192960.0,
      "grad_norm": 5.319862933688955,
      "language_loss": 0.86456263,
      "learning_rate": 3.993813257846589e-06,
      "loss": 0.88799214,
      "num_input_tokens_seen": 9817605,
      "step": 452,
      "time_per_iteration": 2.668957471847534
    },
    {
      "auxiliary_loss_clip": 0.01301953,
      "auxiliary_loss_mlp": 0.01063784,
      "balance_loss_clip": 1.08361101,
      "balance_loss_mlp": 1.05050397,
      "epoch": 0.054470029459508205,
      "flos": 18660729127680.0,
      "grad_norm": 2.3972397695708385,
      "language_loss": 0.9286055,
      "learning_rate": 3.993751883345619e-06,
      "loss": 0.95226288,
      "num_input_tokens_seen": 9835965,
      "step": 453,
      "time_per_iteration": 2.675686836242676
    },
    {
      "auxiliary_loss_clip": 0.01266439,
      "auxiliary_loss_mlp": 0.01060299,
      "balance_loss_clip": 1.0825243,
      "balance_loss_mlp": 1.04709053,
      "epoch": 0.054590272350147295,
      "flos": 17785298856960.0,
      "grad_norm": 2.455508325863405,
      "language_loss": 0.87646979,
      "learning_rate": 3.993690206394073e-06,
      "loss": 0.89973712,
      "num_input_tokens_seen": 9852265,
      "step": 454,
      "time_per_iteration": 2.870145559310913
    },
    {
      "auxiliary_loss_clip": 0.01276906,
      "auxiliary_loss_mlp": 0.01046706,
      "balance_loss_clip": 1.07962263,
      "balance_loss_mlp": 1.03425431,
      "epoch": 0.054710515240786385,
      "flos": 17785945301760.0,
      "grad_norm": 2.6629059372790826,
      "language_loss": 0.87560081,
      "learning_rate": 3.993628227001307e-06,
      "loss": 0.89883697,
      "num_input_tokens_seen": 9870465,
      "step": 455,
      "time_per_iteration": 2.703948497772217
    },
    {
      "auxiliary_loss_clip": 0.01194561,
      "auxiliary_loss_mlp": 0.01054915,
      "balance_loss_clip": 1.06485963,
      "balance_loss_mlp": 1.04241037,
      "epoch": 0.05483075813142548,
      "flos": 48210900180480.0,
      "grad_norm": 2.1080307312367355,
      "language_loss": 0.71662396,
      "learning_rate": 3.993565945176726e-06,
      "loss": 0.73911875,
      "num_input_tokens_seen": 9891490,
      "step": 456,
      "time_per_iteration": 3.102400779724121
    },
    {
      "auxiliary_loss_clip": 0.0122082,
      "auxiliary_loss_mlp": 0.01059041,
      "balance_loss_clip": 1.0675385,
      "balance_loss_mlp": 1.04726362,
      "epoch": 0.05495100102206457,
      "flos": 19682244011520.0,
      "grad_norm": 1.922040048275448,
      "language_loss": 0.83961725,
      "learning_rate": 3.993503360929776e-06,
      "loss": 0.86241591,
      "num_input_tokens_seen": 9910375,
      "step": 457,
      "time_per_iteration": 2.762054443359375
    },
    {
      "auxiliary_loss_clip": 0.01239713,
      "auxiliary_loss_mlp": 0.01050854,
      "balance_loss_clip": 1.06670547,
      "balance_loss_mlp": 1.0376277,
      "epoch": 0.05507124391270366,
      "flos": 26360048453760.0,
      "grad_norm": 1.5771809640261256,
      "language_loss": 0.81086016,
      "learning_rate": 3.99344047426995e-06,
      "loss": 0.8337658,
      "num_input_tokens_seen": 9931635,
      "step": 458,
      "time_per_iteration": 2.7595772743225098
    },
    {
      "auxiliary_loss_clip": 0.01248632,
      "auxiliary_loss_mlp": 0.01046912,
      "balance_loss_clip": 1.07288718,
      "balance_loss_mlp": 1.03264236,
      "epoch": 0.05519148680334275,
      "flos": 22601314581120.0,
      "grad_norm": 2.332213877271301,
      "language_loss": 0.9339869,
      "learning_rate": 3.993377285206789e-06,
      "loss": 0.95694232,
      "num_input_tokens_seen": 9951420,
      "step": 459,
      "time_per_iteration": 2.7911124229431152
    },
    {
      "auxiliary_loss_clip": 0.0127401,
      "auxiliary_loss_mlp": 0.01055453,
      "balance_loss_clip": 1.0758419,
      "balance_loss_mlp": 1.04233992,
      "epoch": 0.05531172969398184,
      "flos": 40552519380480.0,
      "grad_norm": 1.8560809312962359,
      "language_loss": 0.86464471,
      "learning_rate": 3.99331379374988e-06,
      "loss": 0.88793933,
      "num_input_tokens_seen": 9975025,
      "step": 460,
      "time_per_iteration": 4.26049280166626
    },
    {
      "auxiliary_loss_clip": 0.01273643,
      "auxiliary_loss_mlp": 0.01048337,
      "balance_loss_clip": 1.07645559,
      "balance_loss_mlp": 1.03524184,
      "epoch": 0.05543197258462093,
      "flos": 23477894087040.0,
      "grad_norm": 1.9264718174415607,
      "language_loss": 0.79945719,
      "learning_rate": 3.993249999908852e-06,
      "loss": 0.82267702,
      "num_input_tokens_seen": 9995175,
      "step": 461,
      "time_per_iteration": 3.9426348209381104
    },
    {
      "auxiliary_loss_clip": 0.01199539,
      "auxiliary_loss_mlp": 0.01051028,
      "balance_loss_clip": 1.06356406,
      "balance_loss_mlp": 1.03742075,
      "epoch": 0.05555221547526003,
      "flos": 18624603024000.0,
      "grad_norm": 1.8904501407612937,
      "language_loss": 0.87353909,
      "learning_rate": 3.993185903693384e-06,
      "loss": 0.89604479,
      "num_input_tokens_seen": 10011975,
      "step": 462,
      "time_per_iteration": 2.8673253059387207
    },
    {
      "auxiliary_loss_clip": 0.01255512,
      "auxiliary_loss_mlp": 0.00765883,
      "balance_loss_clip": 1.07079959,
      "balance_loss_mlp": 1.00043082,
      "epoch": 0.05567245836589912,
      "flos": 23587098410880.0,
      "grad_norm": 2.1896927155880346,
      "language_loss": 0.82376254,
      "learning_rate": 3.9931215051131995e-06,
      "loss": 0.8439765,
      "num_input_tokens_seen": 10032620,
      "step": 463,
      "time_per_iteration": 2.753380060195923
    },
    {
      "auxiliary_loss_clip": 0.0122524,
      "auxiliary_loss_mlp": 0.01052845,
      "balance_loss_clip": 1.06740689,
      "balance_loss_mlp": 1.03976727,
      "epoch": 0.05579270125653821,
      "flos": 27746667129600.0,
      "grad_norm": 1.9527377543725617,
      "language_loss": 0.8001951,
      "learning_rate": 3.993056804178068e-06,
      "loss": 0.82297593,
      "num_input_tokens_seen": 10054165,
      "step": 464,
      "time_per_iteration": 3.7003660202026367
    },
    {
      "auxiliary_loss_clip": 0.01244441,
      "auxiliary_loss_mlp": 0.01055504,
      "balance_loss_clip": 1.07377958,
      "balance_loss_mlp": 1.04236162,
      "epoch": 0.0559129441471773,
      "flos": 27014161075200.0,
      "grad_norm": 2.546297173863537,
      "language_loss": 0.84191281,
      "learning_rate": 3.992991800897803e-06,
      "loss": 0.86491227,
      "num_input_tokens_seen": 10073970,
      "step": 465,
      "time_per_iteration": 2.8199872970581055
    },
    {
      "auxiliary_loss_clip": 0.01265387,
      "auxiliary_loss_mlp": 0.01048507,
      "balance_loss_clip": 1.07101226,
      "balance_loss_mlp": 1.03564429,
      "epoch": 0.05603318703781639,
      "flos": 15229787794560.0,
      "grad_norm": 2.7953895560134248,
      "language_loss": 0.89702922,
      "learning_rate": 3.9929264952822665e-06,
      "loss": 0.92016816,
      "num_input_tokens_seen": 10091505,
      "step": 466,
      "time_per_iteration": 3.6730782985687256
    },
    {
      "auxiliary_loss_clip": 0.01276536,
      "auxiliary_loss_mlp": 0.01053713,
      "balance_loss_clip": 1.07572579,
      "balance_loss_mlp": 1.04023027,
      "epoch": 0.05615342992845548,
      "flos": 22266482976000.0,
      "grad_norm": 1.906356395775752,
      "language_loss": 0.88230783,
      "learning_rate": 3.992860887341366e-06,
      "loss": 0.90561026,
      "num_input_tokens_seen": 10109675,
      "step": 467,
      "time_per_iteration": 2.807168960571289
    },
    {
      "auxiliary_loss_clip": 0.01286184,
      "auxiliary_loss_mlp": 0.01049843,
      "balance_loss_clip": 1.07761502,
      "balance_loss_mlp": 1.03720689,
      "epoch": 0.056273672819094574,
      "flos": 23584979508480.0,
      "grad_norm": 1.965833293322413,
      "language_loss": 0.81353456,
      "learning_rate": 3.992794977085052e-06,
      "loss": 0.83689487,
      "num_input_tokens_seen": 10127675,
      "step": 468,
      "time_per_iteration": 2.70011043548584
    },
    {
      "auxiliary_loss_clip": 0.01253348,
      "auxiliary_loss_mlp": 0.01043555,
      "balance_loss_clip": 1.07017887,
      "balance_loss_mlp": 1.03076422,
      "epoch": 0.056393915709733664,
      "flos": 19858708552320.0,
      "grad_norm": 4.20299867450432,
      "language_loss": 0.84898829,
      "learning_rate": 3.992728764523326e-06,
      "loss": 0.8719573,
      "num_input_tokens_seen": 10146620,
      "step": 469,
      "time_per_iteration": 2.7194063663482666
    },
    {
      "auxiliary_loss_clip": 0.01259116,
      "auxiliary_loss_mlp": 0.01040682,
      "balance_loss_clip": 1.07614744,
      "balance_loss_mlp": 1.0276643,
      "epoch": 0.05651415860037275,
      "flos": 22163779013760.0,
      "grad_norm": 1.9936188188541,
      "language_loss": 0.80662072,
      "learning_rate": 3.99266224966623e-06,
      "loss": 0.82961869,
      "num_input_tokens_seen": 10167535,
      "step": 470,
      "time_per_iteration": 2.7826030254364014
    },
    {
      "auxiliary_loss_clip": 0.01288235,
      "auxiliary_loss_mlp": 0.01046536,
      "balance_loss_clip": 1.07948232,
      "balance_loss_mlp": 1.03417432,
      "epoch": 0.05663440149101184,
      "flos": 19463548055040.0,
      "grad_norm": 2.003351473337449,
      "language_loss": 0.87716889,
      "learning_rate": 3.992595432523855e-06,
      "loss": 0.90051657,
      "num_input_tokens_seen": 10184825,
      "step": 471,
      "time_per_iteration": 2.6752731800079346
    },
    {
      "auxiliary_loss_clip": 0.01209328,
      "auxiliary_loss_mlp": 0.01049901,
      "balance_loss_clip": 1.06650007,
      "balance_loss_mlp": 1.03759289,
      "epoch": 0.05675464438165093,
      "flos": 22670226823680.0,
      "grad_norm": 2.0440096937179817,
      "language_loss": 0.86106765,
      "learning_rate": 3.992528313106338e-06,
      "loss": 0.88365996,
      "num_input_tokens_seen": 10203025,
      "step": 472,
      "time_per_iteration": 2.8300766944885254
    },
    {
      "auxiliary_loss_clip": 0.01241309,
      "auxiliary_loss_mlp": 0.01052863,
      "balance_loss_clip": 1.07233238,
      "balance_loss_mlp": 1.03970826,
      "epoch": 0.05687488727229002,
      "flos": 16901177495040.0,
      "grad_norm": 5.49488811009901,
      "language_loss": 0.82274771,
      "learning_rate": 3.9924608914238595e-06,
      "loss": 0.84568942,
      "num_input_tokens_seen": 10218020,
      "step": 473,
      "time_per_iteration": 2.7529947757720947
    },
    {
      "auxiliary_loss_clip": 0.01292185,
      "auxiliary_loss_mlp": 0.01042105,
      "balance_loss_clip": 1.07761538,
      "balance_loss_mlp": 1.02855706,
      "epoch": 0.05699513016292912,
      "flos": 29168980945920.0,
      "grad_norm": 2.4790967772042207,
      "language_loss": 0.8394773,
      "learning_rate": 3.992393167486648e-06,
      "loss": 0.86282021,
      "num_input_tokens_seen": 10237170,
      "step": 474,
      "time_per_iteration": 2.773357629776001
    },
    {
      "auxiliary_loss_clip": 0.01239844,
      "auxiliary_loss_mlp": 0.0076588,
      "balance_loss_clip": 1.07526755,
      "balance_loss_mlp": 1.00039124,
      "epoch": 0.05711537305356821,
      "flos": 18916197632640.0,
      "grad_norm": 2.3024808812463675,
      "language_loss": 0.8063395,
      "learning_rate": 3.992325141304977e-06,
      "loss": 0.8263967,
      "num_input_tokens_seen": 10255125,
      "step": 475,
      "time_per_iteration": 2.7353572845458984
    },
    {
      "auxiliary_loss_clip": 0.01233282,
      "auxiliary_loss_mlp": 0.01062595,
      "balance_loss_clip": 1.06623697,
      "balance_loss_mlp": 1.04922009,
      "epoch": 0.0572356159442073,
      "flos": 26758979879040.0,
      "grad_norm": 2.1561760335112923,
      "language_loss": 0.86562079,
      "learning_rate": 3.992256812889166e-06,
      "loss": 0.88857955,
      "num_input_tokens_seen": 10271230,
      "step": 476,
      "time_per_iteration": 2.778930425643921
    },
    {
      "auxiliary_loss_clip": 0.01231145,
      "auxiliary_loss_mlp": 0.00765637,
      "balance_loss_clip": 1.070117,
      "balance_loss_mlp": 1.00046897,
      "epoch": 0.05735585883484639,
      "flos": 35116146840960.0,
      "grad_norm": 2.4616080173889467,
      "language_loss": 0.76602739,
      "learning_rate": 3.992188182249582e-06,
      "loss": 0.78599513,
      "num_input_tokens_seen": 10293125,
      "step": 477,
      "time_per_iteration": 2.864802360534668
    },
    {
      "auxiliary_loss_clip": 0.01262313,
      "auxiliary_loss_mlp": 0.01046813,
      "balance_loss_clip": 1.07697272,
      "balance_loss_mlp": 1.03431964,
      "epoch": 0.05747610172548548,
      "flos": 18734381965440.0,
      "grad_norm": 2.0864750184750234,
      "language_loss": 0.90598023,
      "learning_rate": 3.992119249396633e-06,
      "loss": 0.92907143,
      "num_input_tokens_seen": 10311810,
      "step": 478,
      "time_per_iteration": 2.727592706680298
    },
    {
      "auxiliary_loss_clip": 0.01247987,
      "auxiliary_loss_mlp": 0.01052865,
      "balance_loss_clip": 1.07289124,
      "balance_loss_mlp": 1.03970456,
      "epoch": 0.05759634461612457,
      "flos": 27964752554880.0,
      "grad_norm": 1.8421806008075916,
      "language_loss": 0.81856853,
      "learning_rate": 3.992050014340778e-06,
      "loss": 0.84157705,
      "num_input_tokens_seen": 10332165,
      "step": 479,
      "time_per_iteration": 2.771514892578125
    },
    {
      "auxiliary_loss_clip": 0.01114682,
      "auxiliary_loss_mlp": 0.01010552,
      "balance_loss_clip": 1.04025578,
      "balance_loss_mlp": 1.00506806,
      "epoch": 0.057716587506763666,
      "flos": 69292009405440.0,
      "grad_norm": 0.8520693504510319,
      "language_loss": 0.5501129,
      "learning_rate": 3.99198047709252e-06,
      "loss": 0.57136524,
      "num_input_tokens_seen": 10393685,
      "step": 480,
      "time_per_iteration": 3.8218774795532227
    },
    {
      "auxiliary_loss_clip": 0.01265919,
      "auxiliary_loss_mlp": 0.01034572,
      "balance_loss_clip": 1.07030773,
      "balance_loss_mlp": 1.02215075,
      "epoch": 0.057836830397402755,
      "flos": 25009196745600.0,
      "grad_norm": 2.2807298228179502,
      "language_loss": 0.7885704,
      "learning_rate": 3.991910637662408e-06,
      "loss": 0.81157523,
      "num_input_tokens_seen": 10413975,
      "step": 481,
      "time_per_iteration": 2.986503839492798
    },
    {
      "auxiliary_loss_clip": 0.0125541,
      "auxiliary_loss_mlp": 0.00766289,
      "balance_loss_clip": 1.0723002,
      "balance_loss_mlp": 1.0003593,
      "epoch": 0.057957073288041845,
      "flos": 25593894334080.0,
      "grad_norm": 2.131345365586665,
      "language_loss": 0.80712348,
      "learning_rate": 3.9918404960610355e-06,
      "loss": 0.82734048,
      "num_input_tokens_seen": 10433005,
      "step": 482,
      "time_per_iteration": 2.7998664379119873
    },
    {
      "auxiliary_loss_clip": 0.01237465,
      "auxiliary_loss_mlp": 0.01053403,
      "balance_loss_clip": 1.07450485,
      "balance_loss_mlp": 1.04040933,
      "epoch": 0.058077316178680935,
      "flos": 20777411733120.0,
      "grad_norm": 2.7600971738067375,
      "language_loss": 0.77857363,
      "learning_rate": 3.991770052299043e-06,
      "loss": 0.80148226,
      "num_input_tokens_seen": 10451235,
      "step": 483,
      "time_per_iteration": 2.828376293182373
    },
    {
      "auxiliary_loss_clip": 0.01218622,
      "auxiliary_loss_mlp": 0.01068347,
      "balance_loss_clip": 1.06775796,
      "balance_loss_mlp": 1.05607486,
      "epoch": 0.058197559069320025,
      "flos": 18916484941440.0,
      "grad_norm": 20.16862212697162,
      "language_loss": 0.87744957,
      "learning_rate": 3.991699306387118e-06,
      "loss": 0.90031922,
      "num_input_tokens_seen": 10469705,
      "step": 484,
      "time_per_iteration": 2.747410535812378
    },
    {
      "auxiliary_loss_clip": 0.01240627,
      "auxiliary_loss_mlp": 0.01069422,
      "balance_loss_clip": 1.06835198,
      "balance_loss_mlp": 1.05582023,
      "epoch": 0.058317801959959115,
      "flos": 24863327614080.0,
      "grad_norm": 2.244760039514846,
      "language_loss": 0.78105193,
      "learning_rate": 3.991628258335991e-06,
      "loss": 0.80415237,
      "num_input_tokens_seen": 10491910,
      "step": 485,
      "time_per_iteration": 2.7792489528656006
    },
    {
      "auxiliary_loss_clip": 0.01220355,
      "auxiliary_loss_mlp": 0.01045629,
      "balance_loss_clip": 1.06339955,
      "balance_loss_mlp": 1.03322482,
      "epoch": 0.05843804485059821,
      "flos": 23257977068160.0,
      "grad_norm": 3.425247236969045,
      "language_loss": 0.87794924,
      "learning_rate": 3.991556908156442e-06,
      "loss": 0.90060914,
      "num_input_tokens_seen": 10508435,
      "step": 486,
      "time_per_iteration": 3.8099453449249268
    },
    {
      "auxiliary_loss_clip": 0.01242706,
      "auxiliary_loss_mlp": 0.01050873,
      "balance_loss_clip": 1.06779325,
      "balance_loss_mlp": 1.03713417,
      "epoch": 0.0585582877412373,
      "flos": 23150532510720.0,
      "grad_norm": 1.981216881893937,
      "language_loss": 0.87673521,
      "learning_rate": 3.9914852558592914e-06,
      "loss": 0.89967102,
      "num_input_tokens_seen": 10529485,
      "step": 487,
      "time_per_iteration": 3.7638819217681885
    },
    {
      "auxiliary_loss_clip": 0.01188856,
      "auxiliary_loss_mlp": 0.00766155,
      "balance_loss_clip": 1.06191099,
      "balance_loss_mlp": 1.00041234,
      "epoch": 0.05867853063187639,
      "flos": 23506406507520.0,
      "grad_norm": 3.8824321339954033,
      "language_loss": 0.80706155,
      "learning_rate": 3.991413301455413e-06,
      "loss": 0.82661164,
      "num_input_tokens_seen": 10545935,
      "step": 488,
      "time_per_iteration": 3.1141598224639893
    },
    {
      "auxiliary_loss_clip": 0.01229409,
      "auxiliary_loss_mlp": 0.00765929,
      "balance_loss_clip": 1.06821108,
      "balance_loss_mlp": 1.00039113,
      "epoch": 0.05879877352251548,
      "flos": 29495803818240.0,
      "grad_norm": 2.38306458984647,
      "language_loss": 0.77591097,
      "learning_rate": 3.991341044955719e-06,
      "loss": 0.79586434,
      "num_input_tokens_seen": 10565690,
      "step": 489,
      "time_per_iteration": 2.894582748413086
    },
    {
      "auxiliary_loss_clip": 0.01198993,
      "auxiliary_loss_mlp": 0.01046152,
      "balance_loss_clip": 1.05997705,
      "balance_loss_mlp": 1.03264582,
      "epoch": 0.05891901641315457,
      "flos": 20157485880960.0,
      "grad_norm": 2.177057182866959,
      "language_loss": 0.81580877,
      "learning_rate": 3.991268486371172e-06,
      "loss": 0.83826029,
      "num_input_tokens_seen": 10584245,
      "step": 490,
      "time_per_iteration": 3.7419581413269043
    },
    {
      "auxiliary_loss_clip": 0.01237803,
      "auxiliary_loss_mlp": 0.01049657,
      "balance_loss_clip": 1.06567061,
      "balance_loss_mlp": 1.03676438,
      "epoch": 0.05903925930379366,
      "flos": 24644200694400.0,
      "grad_norm": 5.683784846365164,
      "language_loss": 0.88299096,
      "learning_rate": 3.991195625712779e-06,
      "loss": 0.90586549,
      "num_input_tokens_seen": 10601210,
      "step": 491,
      "time_per_iteration": 3.7297658920288086
    },
    {
      "auxiliary_loss_clip": 0.01250558,
      "auxiliary_loss_mlp": 0.01047942,
      "balance_loss_clip": 1.07450366,
      "balance_loss_mlp": 1.03587842,
      "epoch": 0.05915950219443276,
      "flos": 21250391045760.0,
      "grad_norm": 2.3872375469459715,
      "language_loss": 0.81239927,
      "learning_rate": 3.991122462991592e-06,
      "loss": 0.83538425,
      "num_input_tokens_seen": 10620730,
      "step": 492,
      "time_per_iteration": 2.771327495574951
    },
    {
      "auxiliary_loss_clip": 0.0125353,
      "auxiliary_loss_mlp": 0.01049599,
      "balance_loss_clip": 1.07335567,
      "balance_loss_mlp": 1.03660476,
      "epoch": 0.05927974508507185,
      "flos": 9902727319680.0,
      "grad_norm": 3.6879528377756357,
      "language_loss": 0.80880737,
      "learning_rate": 3.991048998218712e-06,
      "loss": 0.83183867,
      "num_input_tokens_seen": 10634035,
      "step": 493,
      "time_per_iteration": 2.7043185234069824
    },
    {
      "auxiliary_loss_clip": 0.01230951,
      "auxiliary_loss_mlp": 0.01043089,
      "balance_loss_clip": 1.07002711,
      "balance_loss_mlp": 1.03047681,
      "epoch": 0.05939998797571094,
      "flos": 18259499232000.0,
      "grad_norm": 2.2786313647243808,
      "language_loss": 0.76079404,
      "learning_rate": 3.990975231405281e-06,
      "loss": 0.78353441,
      "num_input_tokens_seen": 10652485,
      "step": 494,
      "time_per_iteration": 2.788450241088867
    },
    {
      "auxiliary_loss_clip": 0.01271389,
      "auxiliary_loss_mlp": 0.01057262,
      "balance_loss_clip": 1.07815623,
      "balance_loss_mlp": 1.04427457,
      "epoch": 0.05952023086635003,
      "flos": 28256598558720.0,
      "grad_norm": 2.0446935666309587,
      "language_loss": 0.78768957,
      "learning_rate": 3.990901162562491e-06,
      "loss": 0.81097603,
      "num_input_tokens_seen": 10673175,
      "step": 495,
      "time_per_iteration": 2.742783784866333
    },
    {
      "auxiliary_loss_clip": 0.01231927,
      "auxiliary_loss_mlp": 0.01047283,
      "balance_loss_clip": 1.06948161,
      "balance_loss_mlp": 1.03505838,
      "epoch": 0.05964047375698912,
      "flos": 14902498045440.0,
      "grad_norm": 2.5522499746358553,
      "language_loss": 0.90563208,
      "learning_rate": 3.9908267917015765e-06,
      "loss": 0.92842424,
      "num_input_tokens_seen": 10691235,
      "step": 496,
      "time_per_iteration": 2.768127918243408
    },
    {
      "auxiliary_loss_clip": 0.01174336,
      "auxiliary_loss_mlp": 0.01049085,
      "balance_loss_clip": 1.05684531,
      "balance_loss_mlp": 1.03601956,
      "epoch": 0.059760716647628206,
      "flos": 23185581206400.0,
      "grad_norm": 2.282581371264378,
      "language_loss": 0.93118787,
      "learning_rate": 3.990752118833821e-06,
      "loss": 0.95342213,
      "num_input_tokens_seen": 10708675,
      "step": 497,
      "time_per_iteration": 3.1872541904449463
    },
    {
      "auxiliary_loss_clip": 0.01241101,
      "auxiliary_loss_mlp": 0.0104342,
      "balance_loss_clip": 1.07420492,
      "balance_loss_mlp": 1.03030729,
      "epoch": 0.0598809595382673,
      "flos": 22746968231040.0,
      "grad_norm": 2.002180441625928,
      "language_loss": 0.77821642,
      "learning_rate": 3.990677143970553e-06,
      "loss": 0.80106163,
      "num_input_tokens_seen": 10729485,
      "step": 498,
      "time_per_iteration": 3.070021867752075
    },
    {
      "auxiliary_loss_clip": 0.01211677,
      "auxiliary_loss_mlp": 0.01046724,
      "balance_loss_clip": 1.06554842,
      "balance_loss_mlp": 1.03448701,
      "epoch": 0.06000120242890639,
      "flos": 22127221946880.0,
      "grad_norm": 5.229793186786031,
      "language_loss": 0.81122047,
      "learning_rate": 3.990601867123144e-06,
      "loss": 0.83380449,
      "num_input_tokens_seen": 10749210,
      "step": 499,
      "time_per_iteration": 2.7936851978302
    },
    {
      "auxiliary_loss_clip": 0.01221846,
      "auxiliary_loss_mlp": 0.01047316,
      "balance_loss_clip": 1.0700649,
      "balance_loss_mlp": 1.03473353,
      "epoch": 0.06012144531954548,
      "flos": 19171773878400.0,
      "grad_norm": 2.196317996621545,
      "language_loss": 0.85010123,
      "learning_rate": 3.990526288303014e-06,
      "loss": 0.87279284,
      "num_input_tokens_seen": 10768000,
      "step": 500,
      "time_per_iteration": 2.756338596343994
    },
    {
      "auxiliary_loss_clip": 0.01225803,
      "auxiliary_loss_mlp": 0.01044606,
      "balance_loss_clip": 1.06672168,
      "balance_loss_mlp": 1.03214836,
      "epoch": 0.06024168821018457,
      "flos": 22783345729920.0,
      "grad_norm": 1.8773505849354015,
      "language_loss": 0.90960813,
      "learning_rate": 3.9904504075216295e-06,
      "loss": 0.93231219,
      "num_input_tokens_seen": 10788760,
      "step": 501,
      "time_per_iteration": 2.8044111728668213
    },
    {
      "auxiliary_loss_clip": 0.01236203,
      "auxiliary_loss_mlp": 0.01048959,
      "balance_loss_clip": 1.06733561,
      "balance_loss_mlp": 1.03539276,
      "epoch": 0.06036193110082366,
      "flos": 18770687637120.0,
      "grad_norm": 2.6037399604540763,
      "language_loss": 0.94044423,
      "learning_rate": 3.990374224790501e-06,
      "loss": 0.96329588,
      "num_input_tokens_seen": 10806965,
      "step": 502,
      "time_per_iteration": 2.7652711868286133
    },
    {
      "auxiliary_loss_clip": 0.01247405,
      "auxiliary_loss_mlp": 0.01042228,
      "balance_loss_clip": 1.0677278,
      "balance_loss_mlp": 1.03011084,
      "epoch": 0.06048217399146275,
      "flos": 17201570935680.0,
      "grad_norm": 2.1862289534805472,
      "language_loss": 0.70986432,
      "learning_rate": 3.990297740121185e-06,
      "loss": 0.73276067,
      "num_input_tokens_seen": 10824900,
      "step": 503,
      "time_per_iteration": 2.747790575027466
    },
    {
      "auxiliary_loss_clip": 0.01226522,
      "auxiliary_loss_mlp": 0.01046008,
      "balance_loss_clip": 1.06485891,
      "balance_loss_mlp": 1.03367019,
      "epoch": 0.06060241688210185,
      "flos": 24024131187840.0,
      "grad_norm": 2.7241058514373826,
      "language_loss": 0.7811026,
      "learning_rate": 3.990220953525284e-06,
      "loss": 0.80382788,
      "num_input_tokens_seen": 10842010,
      "step": 504,
      "time_per_iteration": 2.8698737621307373
    },
    {
      "auxiliary_loss_clip": 0.01233294,
      "auxiliary_loss_mlp": 0.0076589,
      "balance_loss_clip": 1.0704608,
      "balance_loss_mlp": 1.00030541,
      "epoch": 0.06072265977274094,
      "flos": 14611190745600.0,
      "grad_norm": 3.706999360790643,
      "language_loss": 0.73739618,
      "learning_rate": 3.9901438650144465e-06,
      "loss": 0.757388,
      "num_input_tokens_seen": 10858260,
      "step": 505,
      "time_per_iteration": 2.7294528484344482
    },
    {
      "auxiliary_loss_clip": 0.0126077,
      "auxiliary_loss_mlp": 0.01048228,
      "balance_loss_clip": 1.07033896,
      "balance_loss_mlp": 1.0348767,
      "epoch": 0.06084290266338003,
      "flos": 20558284813440.0,
      "grad_norm": 4.282434673307724,
      "language_loss": 0.92314982,
      "learning_rate": 3.990066474600367e-06,
      "loss": 0.94623977,
      "num_input_tokens_seen": 10876230,
      "step": 506,
      "time_per_iteration": 2.7259409427642822
    },
    {
      "auxiliary_loss_clip": 0.01220259,
      "auxiliary_loss_mlp": 0.01042731,
      "balance_loss_clip": 1.06325698,
      "balance_loss_mlp": 1.03033948,
      "epoch": 0.06096314555401912,
      "flos": 22309217182080.0,
      "grad_norm": 1.904567725008083,
      "language_loss": 0.67984492,
      "learning_rate": 3.989988782294786e-06,
      "loss": 0.70247483,
      "num_input_tokens_seen": 10896320,
      "step": 507,
      "time_per_iteration": 2.7579636573791504
    },
    {
      "auxiliary_loss_clip": 0.01283776,
      "auxiliary_loss_mlp": 0.01049641,
      "balance_loss_clip": 1.07794857,
      "balance_loss_mlp": 1.03766012,
      "epoch": 0.06108338844465821,
      "flos": 19131374056320.0,
      "grad_norm": 1.965875357055065,
      "language_loss": 0.94597358,
      "learning_rate": 3.989910788109489e-06,
      "loss": 0.96930778,
      "num_input_tokens_seen": 10912970,
      "step": 508,
      "time_per_iteration": 2.746616840362549
    },
    {
      "auxiliary_loss_clip": 0.01210622,
      "auxiliary_loss_mlp": 0.01056351,
      "balance_loss_clip": 1.05931067,
      "balance_loss_mlp": 1.0431304,
      "epoch": 0.0612036313352973,
      "flos": 33584018169600.0,
      "grad_norm": 2.20475309017018,
      "language_loss": 0.74871826,
      "learning_rate": 3.989832492056307e-06,
      "loss": 0.77138799,
      "num_input_tokens_seen": 10933995,
      "step": 509,
      "time_per_iteration": 2.9636499881744385
    },
    {
      "auxiliary_loss_clip": 0.01253036,
      "auxiliary_loss_mlp": 0.01050986,
      "balance_loss_clip": 1.07452464,
      "balance_loss_mlp": 1.03839743,
      "epoch": 0.06132387422593639,
      "flos": 27490552179840.0,
      "grad_norm": 3.5310507366421526,
      "language_loss": 0.81026208,
      "learning_rate": 3.989753894147119e-06,
      "loss": 0.83330232,
      "num_input_tokens_seen": 10954120,
      "step": 510,
      "time_per_iteration": 2.8901736736297607
    },
    {
      "auxiliary_loss_clip": 0.01199705,
      "auxiliary_loss_mlp": 0.00765262,
      "balance_loss_clip": 1.06516218,
      "balance_loss_mlp": 1.00045896,
      "epoch": 0.061444117116575485,
      "flos": 25885057979520.0,
      "grad_norm": 1.7491209954211846,
      "language_loss": 0.80091047,
      "learning_rate": 3.989674994393846e-06,
      "loss": 0.82056016,
      "num_input_tokens_seen": 10973595,
      "step": 511,
      "time_per_iteration": 2.9350991249084473
    },
    {
      "auxiliary_loss_clip": 0.01239406,
      "auxiliary_loss_mlp": 0.01043901,
      "balance_loss_clip": 1.06828356,
      "balance_loss_mlp": 1.03149724,
      "epoch": 0.061564360007214575,
      "flos": 28512031150080.0,
      "grad_norm": 2.3872042995594995,
      "language_loss": 0.93917668,
      "learning_rate": 3.98959579280846e-06,
      "loss": 0.96200967,
      "num_input_tokens_seen": 10991995,
      "step": 512,
      "time_per_iteration": 3.733184576034546
    },
    {
      "auxiliary_loss_clip": 0.0124867,
      "auxiliary_loss_mlp": 0.01053056,
      "balance_loss_clip": 1.0757637,
      "balance_loss_mlp": 1.04058647,
      "epoch": 0.061684602897853665,
      "flos": 12094355652480.0,
      "grad_norm": 8.126737592209796,
      "language_loss": 0.82727695,
      "learning_rate": 3.989516289402973e-06,
      "loss": 0.85029423,
      "num_input_tokens_seen": 11007625,
      "step": 513,
      "time_per_iteration": 3.64353084564209
    },
    {
      "auxiliary_loss_clip": 0.01210655,
      "auxiliary_loss_mlp": 0.01046173,
      "balance_loss_clip": 1.06912756,
      "balance_loss_mlp": 1.03397799,
      "epoch": 0.061804845788492754,
      "flos": 19532639865600.0,
      "grad_norm": 2.657356632565099,
      "language_loss": 0.80197191,
      "learning_rate": 3.989436484189447e-06,
      "loss": 0.82454026,
      "num_input_tokens_seen": 11025570,
      "step": 514,
      "time_per_iteration": 2.788321018218994
    },
    {
      "auxiliary_loss_clip": 0.01196173,
      "auxiliary_loss_mlp": 0.01048188,
      "balance_loss_clip": 1.05777311,
      "balance_loss_mlp": 1.03600526,
      "epoch": 0.061925088679131844,
      "flos": 15341111020800.0,
      "grad_norm": 2.5830039854448428,
      "language_loss": 0.8104459,
      "learning_rate": 3.9893563771799885e-06,
      "loss": 0.83288956,
      "num_input_tokens_seen": 11042045,
      "step": 515,
      "time_per_iteration": 2.7761101722717285
    },
    {
      "auxiliary_loss_clip": 0.01241107,
      "auxiliary_loss_mlp": 0.01046074,
      "balance_loss_clip": 1.0691185,
      "balance_loss_mlp": 1.03366423,
      "epoch": 0.062045331569770934,
      "flos": 25919927107200.0,
      "grad_norm": 3.453291886529484,
      "language_loss": 0.86071646,
      "learning_rate": 3.989275968386749e-06,
      "loss": 0.88358825,
      "num_input_tokens_seen": 11059955,
      "step": 516,
      "time_per_iteration": 3.7975122928619385
    },
    {
      "auxiliary_loss_clip": 0.01267415,
      "auxiliary_loss_mlp": 0.0104993,
      "balance_loss_clip": 1.07369089,
      "balance_loss_mlp": 1.03766906,
      "epoch": 0.06216557446041003,
      "flos": 28110621686400.0,
      "grad_norm": 2.074872396351503,
      "language_loss": 0.7654146,
      "learning_rate": 3.989195257821926e-06,
      "loss": 0.78858805,
      "num_input_tokens_seen": 11078440,
      "step": 517,
      "time_per_iteration": 3.7772343158721924
    },
    {
      "auxiliary_loss_clip": 0.01246887,
      "auxiliary_loss_mlp": 0.01045883,
      "balance_loss_clip": 1.07182837,
      "balance_loss_mlp": 1.03338385,
      "epoch": 0.06228581735104912,
      "flos": 23478181395840.0,
      "grad_norm": 2.3762103496487956,
      "language_loss": 0.84103703,
      "learning_rate": 3.989114245497765e-06,
      "loss": 0.8639648,
      "num_input_tokens_seen": 11098240,
      "step": 518,
      "time_per_iteration": 2.6965699195861816
    },
    {
      "auxiliary_loss_clip": 0.01185934,
      "auxiliary_loss_mlp": 0.01041741,
      "balance_loss_clip": 1.05212665,
      "balance_loss_mlp": 1.02915227,
      "epoch": 0.06240606024168821,
      "flos": 15195205975680.0,
      "grad_norm": 2.24936412142462,
      "language_loss": 0.949947,
      "learning_rate": 3.989032931426554e-06,
      "loss": 0.97222364,
      "num_input_tokens_seen": 11115395,
      "step": 519,
      "time_per_iteration": 2.8260133266448975
    },
    {
      "auxiliary_loss_clip": 0.0120864,
      "auxiliary_loss_mlp": 0.01045383,
      "balance_loss_clip": 1.06892312,
      "balance_loss_mlp": 1.03328288,
      "epoch": 0.06252630313232731,
      "flos": 20631829910400.0,
      "grad_norm": 2.379405475373993,
      "language_loss": 0.86131966,
      "learning_rate": 3.9889513156206295e-06,
      "loss": 0.88385987,
      "num_input_tokens_seen": 11134835,
      "step": 520,
      "time_per_iteration": 2.7457094192504883
    },
    {
      "auxiliary_loss_clip": 0.01256406,
      "auxiliary_loss_mlp": 0.01050443,
      "balance_loss_clip": 1.07201433,
      "balance_loss_mlp": 1.0379436,
      "epoch": 0.06264654602296639,
      "flos": 20778058177920.0,
      "grad_norm": 4.782520578152264,
      "language_loss": 0.73332953,
      "learning_rate": 3.988869398092371e-06,
      "loss": 0.75639802,
      "num_input_tokens_seen": 11154745,
      "step": 521,
      "time_per_iteration": 2.772616147994995
    },
    {
      "auxiliary_loss_clip": 0.01218837,
      "auxiliary_loss_mlp": 0.01048595,
      "balance_loss_clip": 1.06790996,
      "balance_loss_mlp": 1.03603625,
      "epoch": 0.06276678891360549,
      "flos": 29605798241280.0,
      "grad_norm": 2.177066364467193,
      "language_loss": 0.78800869,
      "learning_rate": 3.988787178854206e-06,
      "loss": 0.81068301,
      "num_input_tokens_seen": 11174280,
      "step": 522,
      "time_per_iteration": 2.8576719760894775
    },
    {
      "auxiliary_loss_clip": 0.01283315,
      "auxiliary_loss_mlp": 0.00765719,
      "balance_loss_clip": 1.07624555,
      "balance_loss_mlp": 1.00047755,
      "epoch": 0.06288703180424457,
      "flos": 22126288193280.0,
      "grad_norm": 2.391860791962166,
      "language_loss": 0.87386763,
      "learning_rate": 3.988704657918608e-06,
      "loss": 0.89435798,
      "num_input_tokens_seen": 11193340,
      "step": 523,
      "time_per_iteration": 2.6707231998443604
    },
    {
      "auxiliary_loss_clip": 0.01248865,
      "auxiliary_loss_mlp": 0.00765356,
      "balance_loss_clip": 1.07405424,
      "balance_loss_mlp": 1.00050592,
      "epoch": 0.06300727469488367,
      "flos": 14976689587200.0,
      "grad_norm": 2.594154521037941,
      "language_loss": 0.79494655,
      "learning_rate": 3.988621835298094e-06,
      "loss": 0.81508875,
      "num_input_tokens_seen": 11210555,
      "step": 524,
      "time_per_iteration": 2.73669695854187
    },
    {
      "auxiliary_loss_clip": 0.01174508,
      "auxiliary_loss_mlp": 0.01041143,
      "balance_loss_clip": 1.06424069,
      "balance_loss_mlp": 1.02928174,
      "epoch": 0.06312751758552275,
      "flos": 24535391420160.0,
      "grad_norm": 1.898058011534017,
      "language_loss": 0.91587555,
      "learning_rate": 3.988538711005229e-06,
      "loss": 0.93803209,
      "num_input_tokens_seen": 11230010,
      "step": 525,
      "time_per_iteration": 2.977696657180786
    },
    {
      "auxiliary_loss_clip": 0.01192532,
      "auxiliary_loss_mlp": 0.0103996,
      "balance_loss_clip": 1.06162667,
      "balance_loss_mlp": 1.02822423,
      "epoch": 0.06324776047616185,
      "flos": 21507008785920.0,
      "grad_norm": 2.2434484936186716,
      "language_loss": 0.88308305,
      "learning_rate": 3.988455285052622e-06,
      "loss": 0.90540797,
      "num_input_tokens_seen": 11246190,
      "step": 526,
      "time_per_iteration": 3.198922872543335
    },
    {
      "auxiliary_loss_clip": 0.01263344,
      "auxiliary_loss_mlp": 0.01044606,
      "balance_loss_clip": 1.07569122,
      "balance_loss_mlp": 1.03304291,
      "epoch": 0.06336800336680094,
      "flos": 21688034353920.0,
      "grad_norm": 2.3790849879221763,
      "language_loss": 0.83870161,
      "learning_rate": 3.98837155745293e-06,
      "loss": 0.86178112,
      "num_input_tokens_seen": 11264230,
      "step": 527,
      "time_per_iteration": 2.760385036468506
    },
    {
      "auxiliary_loss_clip": 0.01268151,
      "auxiliary_loss_mlp": 0.00765546,
      "balance_loss_clip": 1.07268572,
      "balance_loss_mlp": 1.00050795,
      "epoch": 0.06348824625744003,
      "flos": 19500895221120.0,
      "grad_norm": 2.1947131736288132,
      "language_loss": 0.76032692,
      "learning_rate": 3.988287528218854e-06,
      "loss": 0.78066391,
      "num_input_tokens_seen": 11283015,
      "step": 528,
      "time_per_iteration": 2.699035167694092
    },
    {
      "auxiliary_loss_clip": 0.01279226,
      "auxiliary_loss_mlp": 0.00764474,
      "balance_loss_clip": 1.07783675,
      "balance_loss_mlp": 1.00050557,
      "epoch": 0.06360848914807912,
      "flos": 15481233976320.0,
      "grad_norm": 2.097625207199049,
      "language_loss": 0.9027108,
      "learning_rate": 3.98820319736314e-06,
      "loss": 0.9231478,
      "num_input_tokens_seen": 11299630,
      "step": 529,
      "time_per_iteration": 2.654670476913452
    },
    {
      "auxiliary_loss_clip": 0.01243044,
      "auxiliary_loss_mlp": 0.01046987,
      "balance_loss_clip": 1.07123733,
      "balance_loss_mlp": 1.03500652,
      "epoch": 0.0637287320387182,
      "flos": 20593369422720.0,
      "grad_norm": 1.9227278922445414,
      "language_loss": 0.85512191,
      "learning_rate": 3.988118564898582e-06,
      "loss": 0.87802219,
      "num_input_tokens_seen": 11319170,
      "step": 530,
      "time_per_iteration": 2.701425075531006
    },
    {
      "auxiliary_loss_clip": 0.01250623,
      "auxiliary_loss_mlp": 0.01046556,
      "balance_loss_clip": 1.07008672,
      "balance_loss_mlp": 1.03415203,
      "epoch": 0.0638489749293573,
      "flos": 17412222245760.0,
      "grad_norm": 2.9794120676695584,
      "language_loss": 0.89393598,
      "learning_rate": 3.988033630838019e-06,
      "loss": 0.91690779,
      "num_input_tokens_seen": 11333210,
      "step": 531,
      "time_per_iteration": 2.6623616218566895
    },
    {
      "auxiliary_loss_clip": 0.01251646,
      "auxiliary_loss_mlp": 0.01047419,
      "balance_loss_clip": 1.0720222,
      "balance_loss_mlp": 1.0359565,
      "epoch": 0.0639692178199964,
      "flos": 23807661874560.0,
      "grad_norm": 2.5859103778946637,
      "language_loss": 0.88226765,
      "learning_rate": 3.987948395194334e-06,
      "loss": 0.9052583,
      "num_input_tokens_seen": 11355590,
      "step": 532,
      "time_per_iteration": 2.8463668823242188
    },
    {
      "auxiliary_loss_clip": 0.01187997,
      "auxiliary_loss_mlp": 0.01055604,
      "balance_loss_clip": 1.05688667,
      "balance_loss_mlp": 1.04408264,
      "epoch": 0.06408946071063548,
      "flos": 18477225521280.0,
      "grad_norm": 2.6427315206540603,
      "language_loss": 0.76472282,
      "learning_rate": 3.987862857980458e-06,
      "loss": 0.78715885,
      "num_input_tokens_seen": 11371535,
      "step": 533,
      "time_per_iteration": 2.7214324474334717
    },
    {
      "auxiliary_loss_clip": 0.01230944,
      "auxiliary_loss_mlp": 0.01054243,
      "balance_loss_clip": 1.07186401,
      "balance_loss_mlp": 1.04052794,
      "epoch": 0.06420970360127458,
      "flos": 27162220936320.0,
      "grad_norm": 7.831122919817089,
      "language_loss": 0.76731038,
      "learning_rate": 3.987777019209368e-06,
      "loss": 0.79016221,
      "num_input_tokens_seen": 11392050,
      "step": 534,
      "time_per_iteration": 2.7567026615142822
    },
    {
      "auxiliary_loss_clip": 0.01249369,
      "auxiliary_loss_mlp": 0.01044354,
      "balance_loss_clip": 1.07103801,
      "balance_loss_mlp": 1.03248644,
      "epoch": 0.06432994649191366,
      "flos": 23659673840640.0,
      "grad_norm": 1.925714893387876,
      "language_loss": 0.81112742,
      "learning_rate": 3.987690878894084e-06,
      "loss": 0.83406472,
      "num_input_tokens_seen": 11411765,
      "step": 535,
      "time_per_iteration": 2.7823357582092285
    },
    {
      "auxiliary_loss_clip": 0.01269773,
      "auxiliary_loss_mlp": 0.01045758,
      "balance_loss_clip": 1.07549667,
      "balance_loss_mlp": 1.03341413,
      "epoch": 0.06445018938255276,
      "flos": 23403953940480.0,
      "grad_norm": 2.5156527656617866,
      "language_loss": 0.84500217,
      "learning_rate": 3.987604437047673e-06,
      "loss": 0.86815751,
      "num_input_tokens_seen": 11431565,
      "step": 536,
      "time_per_iteration": 2.7140822410583496
    },
    {
      "auxiliary_loss_clip": 0.01216289,
      "auxiliary_loss_mlp": 0.01053572,
      "balance_loss_clip": 1.06984973,
      "balance_loss_mlp": 1.04199028,
      "epoch": 0.06457043227319184,
      "flos": 19646692525440.0,
      "grad_norm": 2.1924133690934013,
      "language_loss": 0.77391976,
      "learning_rate": 3.987517693683251e-06,
      "loss": 0.7966184,
      "num_input_tokens_seen": 11450140,
      "step": 537,
      "time_per_iteration": 2.875403642654419
    },
    {
      "auxiliary_loss_clip": 0.01210919,
      "auxiliary_loss_mlp": 0.01049789,
      "balance_loss_clip": 1.0653981,
      "balance_loss_mlp": 1.0377667,
      "epoch": 0.06469067516383094,
      "flos": 16978744915200.0,
      "grad_norm": 3.1589763511913858,
      "language_loss": 0.95976961,
      "learning_rate": 3.9874306488139745e-06,
      "loss": 0.98237669,
      "num_input_tokens_seen": 11465400,
      "step": 538,
      "time_per_iteration": 3.6934847831726074
    },
    {
      "auxiliary_loss_clip": 0.01202273,
      "auxiliary_loss_mlp": 0.01051257,
      "balance_loss_clip": 1.0620563,
      "balance_loss_mlp": 1.03955674,
      "epoch": 0.06481091805447003,
      "flos": 23296401642240.0,
      "grad_norm": 1.9016886796847168,
      "language_loss": 0.87890542,
      "learning_rate": 3.987343302453049e-06,
      "loss": 0.9014408,
      "num_input_tokens_seen": 11486675,
      "step": 539,
      "time_per_iteration": 3.7721173763275146
    },
    {
      "auxiliary_loss_clip": 0.01186154,
      "auxiliary_loss_mlp": 0.0104115,
      "balance_loss_clip": 1.06404042,
      "balance_loss_mlp": 1.02827549,
      "epoch": 0.06493116094510912,
      "flos": 29172356824320.0,
      "grad_norm": 1.6639389583730573,
      "language_loss": 0.82717228,
      "learning_rate": 3.987255654613724e-06,
      "loss": 0.84944528,
      "num_input_tokens_seen": 11510440,
      "step": 540,
      "time_per_iteration": 2.9513256549835205
    },
    {
      "auxiliary_loss_clip": 0.01221526,
      "auxiliary_loss_mlp": 0.01051635,
      "balance_loss_clip": 1.06765068,
      "balance_loss_mlp": 1.03839672,
      "epoch": 0.06505140383574821,
      "flos": 19865065259520.0,
      "grad_norm": 3.4513382650387205,
      "language_loss": 0.70229602,
      "learning_rate": 3.987167705309296e-06,
      "loss": 0.72502756,
      "num_input_tokens_seen": 11529715,
      "step": 541,
      "time_per_iteration": 4.130955219268799
    },
    {
      "auxiliary_loss_clip": 0.01248826,
      "auxiliary_loss_mlp": 0.01044931,
      "balance_loss_clip": 1.07442772,
      "balance_loss_mlp": 1.03225338,
      "epoch": 0.0651716467263873,
      "flos": 17924703540480.0,
      "grad_norm": 3.3033007339155236,
      "language_loss": 0.9511466,
      "learning_rate": 3.987079454553108e-06,
      "loss": 0.9740842,
      "num_input_tokens_seen": 11547665,
      "step": 542,
      "time_per_iteration": 2.733623504638672
    },
    {
      "auxiliary_loss_clip": 0.01250918,
      "auxiliary_loss_mlp": 0.01053934,
      "balance_loss_clip": 1.07198,
      "balance_loss_mlp": 1.04209089,
      "epoch": 0.0652918896170264,
      "flos": 20842840356480.0,
      "grad_norm": 2.1921960424878066,
      "language_loss": 0.91123617,
      "learning_rate": 3.986990902358546e-06,
      "loss": 0.93428469,
      "num_input_tokens_seen": 11564605,
      "step": 543,
      "time_per_iteration": 3.6900651454925537
    },
    {
      "auxiliary_loss_clip": 0.01202809,
      "auxiliary_loss_mlp": 0.01056276,
      "balance_loss_clip": 1.06423962,
      "balance_loss_mlp": 1.0433476,
      "epoch": 0.06541213250766549,
      "flos": 21872507627520.0,
      "grad_norm": 2.086152119020541,
      "language_loss": 0.93233526,
      "learning_rate": 3.986902048739045e-06,
      "loss": 0.95492607,
      "num_input_tokens_seen": 11584550,
      "step": 544,
      "time_per_iteration": 2.802401542663574
    },
    {
      "auxiliary_loss_clip": 0.01275849,
      "auxiliary_loss_mlp": 0.01052071,
      "balance_loss_clip": 1.07448757,
      "balance_loss_mlp": 1.03963161,
      "epoch": 0.06553237539830457,
      "flos": 23110743219840.0,
      "grad_norm": 3.2722802093363934,
      "language_loss": 0.80113703,
      "learning_rate": 3.986812893708082e-06,
      "loss": 0.82441628,
      "num_input_tokens_seen": 11600740,
      "step": 545,
      "time_per_iteration": 2.6449735164642334
    },
    {
      "auxiliary_loss_clip": 0.0123523,
      "auxiliary_loss_mlp": 0.0104496,
      "balance_loss_clip": 1.07100439,
      "balance_loss_mlp": 1.03315198,
      "epoch": 0.06565261828894367,
      "flos": 17923769786880.0,
      "grad_norm": 2.29457143244274,
      "language_loss": 0.8127116,
      "learning_rate": 3.9867234372791826e-06,
      "loss": 0.83551347,
      "num_input_tokens_seen": 11618695,
      "step": 546,
      "time_per_iteration": 2.8298468589782715
    },
    {
      "auxiliary_loss_clip": 0.01242883,
      "auxiliary_loss_mlp": 0.01045708,
      "balance_loss_clip": 1.06611955,
      "balance_loss_mlp": 1.03351307,
      "epoch": 0.06577286117958275,
      "flos": 22783058421120.0,
      "grad_norm": 1.6661512318348533,
      "language_loss": 0.87206882,
      "learning_rate": 3.986633679465918e-06,
      "loss": 0.8949548,
      "num_input_tokens_seen": 11638850,
      "step": 547,
      "time_per_iteration": 2.820199966430664
    },
    {
      "auxiliary_loss_clip": 0.01265993,
      "auxiliary_loss_mlp": 0.01054505,
      "balance_loss_clip": 1.07810354,
      "balance_loss_mlp": 1.0405097,
      "epoch": 0.06589310407022185,
      "flos": 23696194993920.0,
      "grad_norm": 2.476384493328963,
      "language_loss": 0.80997109,
      "learning_rate": 3.986543620281904e-06,
      "loss": 0.83317602,
      "num_input_tokens_seen": 11658500,
      "step": 548,
      "time_per_iteration": 2.7425036430358887
    },
    {
      "auxiliary_loss_clip": 0.01194374,
      "auxiliary_loss_mlp": 0.01046517,
      "balance_loss_clip": 1.06760895,
      "balance_loss_mlp": 1.03436911,
      "epoch": 0.06601334696086093,
      "flos": 26864772410880.0,
      "grad_norm": 1.752235067409458,
      "language_loss": 0.91166103,
      "learning_rate": 3.986453259740802e-06,
      "loss": 0.93406999,
      "num_input_tokens_seen": 11676670,
      "step": 549,
      "time_per_iteration": 2.959291696548462
    },
    {
      "auxiliary_loss_clip": 0.01201331,
      "auxiliary_loss_mlp": 0.01051243,
      "balance_loss_clip": 1.06391561,
      "balance_loss_mlp": 1.03879166,
      "epoch": 0.06613358985150003,
      "flos": 12567694101120.0,
      "grad_norm": 2.844893230621318,
      "language_loss": 0.79516053,
      "learning_rate": 3.986362597856319e-06,
      "loss": 0.81768626,
      "num_input_tokens_seen": 11693170,
      "step": 550,
      "time_per_iteration": 2.943969488143921
    },
    {
      "auxiliary_loss_clip": 0.01233443,
      "auxiliary_loss_mlp": 0.01043994,
      "balance_loss_clip": 1.06728077,
      "balance_loss_mlp": 1.03215015,
      "epoch": 0.06625383274213913,
      "flos": 18332505624960.0,
      "grad_norm": 2.180099677995315,
      "language_loss": 0.81458819,
      "learning_rate": 3.986271634642211e-06,
      "loss": 0.83736253,
      "num_input_tokens_seen": 11710150,
      "step": 551,
      "time_per_iteration": 2.9140915870666504
    },
    {
      "auxiliary_loss_clip": 0.01260639,
      "auxiliary_loss_mlp": 0.01050255,
      "balance_loss_clip": 1.07659626,
      "balance_loss_mlp": 1.03789937,
      "epoch": 0.06637407563277821,
      "flos": 15375585098880.0,
      "grad_norm": 2.0004002708645197,
      "language_loss": 0.81466544,
      "learning_rate": 3.986180370112274e-06,
      "loss": 0.8377744,
      "num_input_tokens_seen": 11726670,
      "step": 552,
      "time_per_iteration": 2.8319647312164307
    },
    {
      "auxiliary_loss_clip": 0.0117966,
      "auxiliary_loss_mlp": 0.01040128,
      "balance_loss_clip": 1.05857694,
      "balance_loss_mlp": 1.02807617,
      "epoch": 0.0664943185234173,
      "flos": 24025244509440.0,
      "grad_norm": 2.0404729733477875,
      "language_loss": 0.74486941,
      "learning_rate": 3.986088804280354e-06,
      "loss": 0.76706731,
      "num_input_tokens_seen": 11746400,
      "step": 553,
      "time_per_iteration": 3.1685171127319336
    },
    {
      "auxiliary_loss_clip": 0.01199705,
      "auxiliary_loss_mlp": 0.01044396,
      "balance_loss_clip": 1.06790042,
      "balance_loss_mlp": 1.03219485,
      "epoch": 0.06661456141405639,
      "flos": 20957503547520.0,
      "grad_norm": 2.2360523601390865,
      "language_loss": 0.93825519,
      "learning_rate": 3.985996937160342e-06,
      "loss": 0.96069628,
      "num_input_tokens_seen": 11765590,
      "step": 554,
      "time_per_iteration": 3.209803581237793
    },
    {
      "auxiliary_loss_clip": 0.01198135,
      "auxiliary_loss_mlp": 0.00765674,
      "balance_loss_clip": 1.06402373,
      "balance_loss_mlp": 1.00116169,
      "epoch": 0.06673480430469549,
      "flos": 52223953322880.0,
      "grad_norm": 1.9450977319532834,
      "language_loss": 0.6868822,
      "learning_rate": 3.985904768766173e-06,
      "loss": 0.70652032,
      "num_input_tokens_seen": 11788365,
      "step": 555,
      "time_per_iteration": 3.08249831199646
    },
    {
      "auxiliary_loss_clip": 0.01205625,
      "auxiliary_loss_mlp": 0.01055691,
      "balance_loss_clip": 1.06196213,
      "balance_loss_mlp": 1.04258966,
      "epoch": 0.06685504719533458,
      "flos": 16217079995520.0,
      "grad_norm": 3.4323589529389693,
      "language_loss": 0.75607371,
      "learning_rate": 3.98581229911183e-06,
      "loss": 0.77868676,
      "num_input_tokens_seen": 11807285,
      "step": 556,
      "time_per_iteration": 2.8751697540283203
    },
    {
      "auxiliary_loss_clip": 0.01229488,
      "auxiliary_loss_mlp": 0.01047933,
      "balance_loss_clip": 1.0673033,
      "balance_loss_mlp": 1.03497541,
      "epoch": 0.06697529008597367,
      "flos": 22491535639680.0,
      "grad_norm": 1.7329965347645084,
      "language_loss": 0.92096436,
      "learning_rate": 3.985719528211341e-06,
      "loss": 0.94373858,
      "num_input_tokens_seen": 11826655,
      "step": 557,
      "time_per_iteration": 2.8535451889038086
    },
    {
      "auxiliary_loss_clip": 0.01149449,
      "auxiliary_loss_mlp": 0.01012603,
      "balance_loss_clip": 1.04906499,
      "balance_loss_mlp": 1.00835884,
      "epoch": 0.06709553297661276,
      "flos": 62688216936960.0,
      "grad_norm": 0.8479931114895232,
      "language_loss": 0.6303966,
      "learning_rate": 3.985626456078777e-06,
      "loss": 0.65201712,
      "num_input_tokens_seen": 11891310,
      "step": 558,
      "time_per_iteration": 3.4326705932617188
    },
    {
      "auxiliary_loss_clip": 0.01215564,
      "auxiliary_loss_mlp": 0.01059344,
      "balance_loss_clip": 1.07150578,
      "balance_loss_mlp": 1.04683924,
      "epoch": 0.06721577586725185,
      "flos": 11216590997760.0,
      "grad_norm": 2.377401444282168,
      "language_loss": 0.86155629,
      "learning_rate": 3.985533082728259e-06,
      "loss": 0.8843053,
      "num_input_tokens_seen": 11906965,
      "step": 559,
      "time_per_iteration": 2.8315460681915283
    },
    {
      "auxiliary_loss_clip": 0.01264406,
      "auxiliary_loss_mlp": 0.00765697,
      "balance_loss_clip": 1.07338738,
      "balance_loss_mlp": 1.00107241,
      "epoch": 0.06733601875789094,
      "flos": 25922189664000.0,
      "grad_norm": 2.6851649794739334,
      "language_loss": 0.74512291,
      "learning_rate": 3.985439408173951e-06,
      "loss": 0.76542389,
      "num_input_tokens_seen": 11927190,
      "step": 560,
      "time_per_iteration": 2.7914209365844727
    },
    {
      "auxiliary_loss_clip": 0.01244686,
      "auxiliary_loss_mlp": 0.01051455,
      "balance_loss_clip": 1.07404828,
      "balance_loss_mlp": 1.03953373,
      "epoch": 0.06745626164853002,
      "flos": 20813645577600.0,
      "grad_norm": 2.1953931538012315,
      "language_loss": 0.70670712,
      "learning_rate": 3.9853454324300634e-06,
      "loss": 0.7296685,
      "num_input_tokens_seen": 11946400,
      "step": 561,
      "time_per_iteration": 2.727506637573242
    },
    {
      "auxiliary_loss_clip": 0.01222256,
      "auxiliary_loss_mlp": 0.01047631,
      "balance_loss_clip": 1.07059479,
      "balance_loss_mlp": 1.03483987,
      "epoch": 0.06757650453916912,
      "flos": 19829262378240.0,
      "grad_norm": 2.3344321599656537,
      "language_loss": 0.78105509,
      "learning_rate": 3.985251155510852e-06,
      "loss": 0.80375397,
      "num_input_tokens_seen": 11965430,
      "step": 562,
      "time_per_iteration": 2.7084105014801025
    },
    {
      "auxiliary_loss_clip": 0.01274916,
      "auxiliary_loss_mlp": 0.01052672,
      "balance_loss_clip": 1.07303667,
      "balance_loss_mlp": 1.04077482,
      "epoch": 0.06769674742980822,
      "flos": 25739224761600.0,
      "grad_norm": 1.894268491730344,
      "language_loss": 0.79861766,
      "learning_rate": 3.98515657743062e-06,
      "loss": 0.82189357,
      "num_input_tokens_seen": 11984895,
      "step": 563,
      "time_per_iteration": 3.528775691986084
    },
    {
      "auxiliary_loss_clip": 0.01260549,
      "auxiliary_loss_mlp": 0.01043579,
      "balance_loss_clip": 1.07382584,
      "balance_loss_mlp": 1.03153872,
      "epoch": 0.0678169903204473,
      "flos": 13074788355840.0,
      "grad_norm": 2.1272553757347628,
      "language_loss": 0.77985173,
      "learning_rate": 3.985061698203711e-06,
      "loss": 0.80289298,
      "num_input_tokens_seen": 12002010,
      "step": 564,
      "time_per_iteration": 2.6549177169799805
    },
    {
      "auxiliary_loss_clip": 0.01152225,
      "auxiliary_loss_mlp": 0.01009285,
      "balance_loss_clip": 1.04842925,
      "balance_loss_mlp": 1.0049454,
      "epoch": 0.0679372332110864,
      "flos": 70865830788480.0,
      "grad_norm": 0.8889283473994747,
      "language_loss": 0.63857609,
      "learning_rate": 3.984966517844523e-06,
      "loss": 0.66019124,
      "num_input_tokens_seen": 12057255,
      "step": 565,
      "time_per_iteration": 4.2005555629730225
    },
    {
      "auxiliary_loss_clip": 0.01197384,
      "auxiliary_loss_mlp": 0.01036077,
      "balance_loss_clip": 1.06091189,
      "balance_loss_mlp": 1.02358365,
      "epoch": 0.06805747610172548,
      "flos": 28256418990720.0,
      "grad_norm": 3.2271889532672433,
      "language_loss": 0.8048619,
      "learning_rate": 3.984871036367492e-06,
      "loss": 0.82719654,
      "num_input_tokens_seen": 12077280,
      "step": 566,
      "time_per_iteration": 2.8422439098358154
    },
    {
      "auxiliary_loss_clip": 0.01203309,
      "auxiliary_loss_mlp": 0.01049531,
      "balance_loss_clip": 1.06670785,
      "balance_loss_mlp": 1.03676915,
      "epoch": 0.06817771899236458,
      "flos": 20120533764480.0,
      "grad_norm": 2.161550698381199,
      "language_loss": 0.83120704,
      "learning_rate": 3.984775253787102e-06,
      "loss": 0.85373545,
      "num_input_tokens_seen": 12095570,
      "step": 567,
      "time_per_iteration": 3.7041242122650146
    },
    {
      "auxiliary_loss_clip": 0.01224188,
      "auxiliary_loss_mlp": 0.01042516,
      "balance_loss_clip": 1.06206751,
      "balance_loss_mlp": 1.03051746,
      "epoch": 0.06829796188300366,
      "flos": 17930629284480.0,
      "grad_norm": 3.4042091637508625,
      "language_loss": 0.87818724,
      "learning_rate": 3.984679170117885e-06,
      "loss": 0.90085423,
      "num_input_tokens_seen": 12111775,
      "step": 568,
      "time_per_iteration": 2.682948112487793
    },
    {
      "auxiliary_loss_clip": 0.01246224,
      "auxiliary_loss_mlp": 0.00765557,
      "balance_loss_clip": 1.06766009,
      "balance_loss_mlp": 1.001086,
      "epoch": 0.06841820477364276,
      "flos": 14501627285760.0,
      "grad_norm": 2.280006577170726,
      "language_loss": 0.78398526,
      "learning_rate": 3.984582785374415e-06,
      "loss": 0.80410314,
      "num_input_tokens_seen": 12129215,
      "step": 569,
      "time_per_iteration": 3.678678512573242
    },
    {
      "auxiliary_loss_clip": 0.0123445,
      "auxiliary_loss_mlp": 0.01042915,
      "balance_loss_clip": 1.06256127,
      "balance_loss_mlp": 1.03110695,
      "epoch": 0.06853844766428185,
      "flos": 21938474954880.0,
      "grad_norm": 2.2292399448612508,
      "language_loss": 0.80541563,
      "learning_rate": 3.9844860995713155e-06,
      "loss": 0.82818925,
      "num_input_tokens_seen": 12148755,
      "step": 570,
      "time_per_iteration": 2.7476067543029785
    },
    {
      "auxiliary_loss_clip": 0.0123895,
      "auxiliary_loss_mlp": 0.01043473,
      "balance_loss_clip": 1.07145405,
      "balance_loss_mlp": 1.03142715,
      "epoch": 0.06865869055492094,
      "flos": 16800628348800.0,
      "grad_norm": 2.4224397505099566,
      "language_loss": 0.82324165,
      "learning_rate": 3.9843891127232524e-06,
      "loss": 0.84606582,
      "num_input_tokens_seen": 12166290,
      "step": 571,
      "time_per_iteration": 2.719285726547241
    },
    {
      "auxiliary_loss_clip": 0.01206394,
      "auxiliary_loss_mlp": 0.01046816,
      "balance_loss_clip": 1.06721175,
      "balance_loss_mlp": 1.03472841,
      "epoch": 0.06877893344556003,
      "flos": 19937281553280.0,
      "grad_norm": 2.6089714676123585,
      "language_loss": 0.67101586,
      "learning_rate": 3.984291824844938e-06,
      "loss": 0.6935479,
      "num_input_tokens_seen": 12181385,
      "step": 572,
      "time_per_iteration": 2.770989179611206
    },
    {
      "auxiliary_loss_clip": 0.01231737,
      "auxiliary_loss_mlp": 0.01045707,
      "balance_loss_clip": 1.06827438,
      "balance_loss_mlp": 1.03428638,
      "epoch": 0.06889917633619912,
      "flos": 23039388852480.0,
      "grad_norm": 2.9137195669966407,
      "language_loss": 0.84698677,
      "learning_rate": 3.984194235951132e-06,
      "loss": 0.86976123,
      "num_input_tokens_seen": 12197530,
      "step": 573,
      "time_per_iteration": 2.7343716621398926
    },
    {
      "auxiliary_loss_clip": 0.01185895,
      "auxiliary_loss_mlp": 0.01040689,
      "balance_loss_clip": 1.05725205,
      "balance_loss_mlp": 1.02878606,
      "epoch": 0.06901941922683821,
      "flos": 20960556203520.0,
      "grad_norm": 3.057749371774753,
      "language_loss": 0.84424561,
      "learning_rate": 3.9840963460566375e-06,
      "loss": 0.86651146,
      "num_input_tokens_seen": 12216310,
      "step": 574,
      "time_per_iteration": 2.784069538116455
    },
    {
      "auxiliary_loss_clip": 0.01234768,
      "auxiliary_loss_mlp": 0.01046104,
      "balance_loss_clip": 1.06660581,
      "balance_loss_mlp": 1.03429008,
      "epoch": 0.06913966211747731,
      "flos": 24821850384000.0,
      "grad_norm": 1.6209812682777887,
      "language_loss": 0.89463085,
      "learning_rate": 3.983998155176305e-06,
      "loss": 0.91743958,
      "num_input_tokens_seen": 12236670,
      "step": 575,
      "time_per_iteration": 2.835955858230591
    },
    {
      "auxiliary_loss_clip": 0.01155322,
      "auxiliary_loss_mlp": 0.0100413,
      "balance_loss_clip": 1.05392718,
      "balance_loss_mlp": 1.00017214,
      "epoch": 0.06925990500811639,
      "flos": 58367446957440.0,
      "grad_norm": 0.8218476830825504,
      "language_loss": 0.57042873,
      "learning_rate": 3.9838996633250305e-06,
      "loss": 0.59202325,
      "num_input_tokens_seen": 12297185,
      "step": 576,
      "time_per_iteration": 3.3070130348205566
    },
    {
      "auxiliary_loss_clip": 0.01237885,
      "auxiliary_loss_mlp": 0.0104269,
      "balance_loss_clip": 1.06503344,
      "balance_loss_mlp": 1.03130579,
      "epoch": 0.06938014789875549,
      "flos": 12749940731520.0,
      "grad_norm": 2.5802757622249284,
      "language_loss": 0.88357341,
      "learning_rate": 3.983800870517753e-06,
      "loss": 0.90637916,
      "num_input_tokens_seen": 12313975,
      "step": 577,
      "time_per_iteration": 2.832782506942749
    },
    {
      "auxiliary_loss_clip": 0.011979,
      "auxiliary_loss_mlp": 0.01050022,
      "balance_loss_clip": 1.06249309,
      "balance_loss_mlp": 1.03856635,
      "epoch": 0.06950039078939457,
      "flos": 22820226019200.0,
      "grad_norm": 4.240980041724208,
      "language_loss": 0.78884685,
      "learning_rate": 3.983701776769463e-06,
      "loss": 0.81132603,
      "num_input_tokens_seen": 12331385,
      "step": 578,
      "time_per_iteration": 2.8131930828094482
    },
    {
      "auxiliary_loss_clip": 0.01197373,
      "auxiliary_loss_mlp": 0.01046725,
      "balance_loss_clip": 1.06720877,
      "balance_loss_mlp": 1.03508973,
      "epoch": 0.06962063368003367,
      "flos": 21941348042880.0,
      "grad_norm": 1.9886924484117616,
      "language_loss": 0.85522622,
      "learning_rate": 3.9836023820951885e-06,
      "loss": 0.87766713,
      "num_input_tokens_seen": 12350600,
      "step": 579,
      "time_per_iteration": 3.0400631427764893
    },
    {
      "auxiliary_loss_clip": 0.01249307,
      "auxiliary_loss_mlp": 0.01044439,
      "balance_loss_clip": 1.06537795,
      "balance_loss_mlp": 1.0325067,
      "epoch": 0.06974087657067275,
      "flos": 20706021452160.0,
      "grad_norm": 2.074839185995148,
      "language_loss": 0.68734109,
      "learning_rate": 3.983502686510011e-06,
      "loss": 0.71027863,
      "num_input_tokens_seen": 12371430,
      "step": 580,
      "time_per_iteration": 2.693427562713623
    },
    {
      "auxiliary_loss_clip": 0.01227398,
      "auxiliary_loss_mlp": 0.01048315,
      "balance_loss_clip": 1.06667578,
      "balance_loss_mlp": 1.03586984,
      "epoch": 0.06986111946131185,
      "flos": 22638230784000.0,
      "grad_norm": 1.8798159950082667,
      "language_loss": 0.73684669,
      "learning_rate": 3.9834026900290525e-06,
      "loss": 0.7596038,
      "num_input_tokens_seen": 12390825,
      "step": 581,
      "time_per_iteration": 2.725153923034668
    },
    {
      "auxiliary_loss_clip": 0.01208827,
      "auxiliary_loss_mlp": 0.01050812,
      "balance_loss_clip": 1.06593966,
      "balance_loss_mlp": 1.03892124,
      "epoch": 0.06998136235195095,
      "flos": 26943453152640.0,
      "grad_norm": 1.8965499659658587,
      "language_loss": 1.00054896,
      "learning_rate": 3.983302392667482e-06,
      "loss": 1.02314544,
      "num_input_tokens_seen": 12411670,
      "step": 582,
      "time_per_iteration": 2.874509572982788
    },
    {
      "auxiliary_loss_clip": 0.01215657,
      "auxiliary_loss_mlp": 0.01045385,
      "balance_loss_clip": 1.06296015,
      "balance_loss_mlp": 1.0336374,
      "epoch": 0.07010160524259003,
      "flos": 22492505306880.0,
      "grad_norm": 1.9407928546672855,
      "language_loss": 0.9378767,
      "learning_rate": 3.983201794440517e-06,
      "loss": 0.96048725,
      "num_input_tokens_seen": 12431245,
      "step": 583,
      "time_per_iteration": 2.7837202548980713
    },
    {
      "auxiliary_loss_clip": 0.01272468,
      "auxiliary_loss_mlp": 0.01044986,
      "balance_loss_clip": 1.07253122,
      "balance_loss_mlp": 1.03300512,
      "epoch": 0.07022184813322913,
      "flos": 18332541538560.0,
      "grad_norm": 1.9033391039338037,
      "language_loss": 0.67480874,
      "learning_rate": 3.9831008953634165e-06,
      "loss": 0.69798326,
      "num_input_tokens_seen": 12450535,
      "step": 584,
      "time_per_iteration": 2.6220948696136475
    },
    {
      "auxiliary_loss_clip": 0.01259994,
      "auxiliary_loss_mlp": 0.01052713,
      "balance_loss_clip": 1.0692606,
      "balance_loss_mlp": 1.04102433,
      "epoch": 0.07034209102386821,
      "flos": 24675550289280.0,
      "grad_norm": 3.053311280318837,
      "language_loss": 0.81094968,
      "learning_rate": 3.9829996954514864e-06,
      "loss": 0.83407676,
      "num_input_tokens_seen": 12469675,
      "step": 585,
      "time_per_iteration": 2.721475124359131
    },
    {
      "auxiliary_loss_clip": 0.01165023,
      "auxiliary_loss_mlp": 0.01041273,
      "balance_loss_clip": 1.05764079,
      "balance_loss_mlp": 1.02835095,
      "epoch": 0.0704623339145073,
      "flos": 25995878415360.0,
      "grad_norm": 2.1745990924999283,
      "language_loss": 0.84281027,
      "learning_rate": 3.982898194720079e-06,
      "loss": 0.86487317,
      "num_input_tokens_seen": 12490405,
      "step": 586,
      "time_per_iteration": 3.0298781394958496
    },
    {
      "auxiliary_loss_clip": 0.01232205,
      "auxiliary_loss_mlp": 0.0104431,
      "balance_loss_clip": 1.06764162,
      "balance_loss_mlp": 1.03160238,
      "epoch": 0.0705825768051464,
      "flos": 25338318088320.0,
      "grad_norm": 2.094014614823836,
      "language_loss": 0.8235786,
      "learning_rate": 3.982796393184592e-06,
      "loss": 0.84634376,
      "num_input_tokens_seen": 12509485,
      "step": 587,
      "time_per_iteration": 3.021771192550659
    },
    {
      "auxiliary_loss_clip": 0.01169723,
      "auxiliary_loss_mlp": 0.01004392,
      "balance_loss_clip": 1.05671477,
      "balance_loss_mlp": 1.00060129,
      "epoch": 0.07070281969578548,
      "flos": 66047552507520.0,
      "grad_norm": 0.7896606635108009,
      "language_loss": 0.62605387,
      "learning_rate": 3.98269429086047e-06,
      "loss": 0.64779502,
      "num_input_tokens_seen": 12567325,
      "step": 588,
      "time_per_iteration": 3.2227704524993896
    },
    {
      "auxiliary_loss_clip": 0.01209828,
      "auxiliary_loss_mlp": 0.01048125,
      "balance_loss_clip": 1.06355906,
      "balance_loss_mlp": 1.03584647,
      "epoch": 0.07082306258642458,
      "flos": 23653568528640.0,
      "grad_norm": 2.5234893557185556,
      "language_loss": 0.86399806,
      "learning_rate": 3.982591887763199e-06,
      "loss": 0.88657761,
      "num_input_tokens_seen": 12584785,
      "step": 589,
      "time_per_iteration": 3.935974597930908
    },
    {
      "auxiliary_loss_clip": 0.01217896,
      "auxiliary_loss_mlp": 0.01045864,
      "balance_loss_clip": 1.06524634,
      "balance_loss_mlp": 1.03403211,
      "epoch": 0.07094330547706366,
      "flos": 13880049408000.0,
      "grad_norm": 2.2736804654672698,
      "language_loss": 0.81953847,
      "learning_rate": 3.982489183908316e-06,
      "loss": 0.84217608,
      "num_input_tokens_seen": 12601205,
      "step": 590,
      "time_per_iteration": 3.571317434310913
    },
    {
      "auxiliary_loss_clip": 0.01247909,
      "auxiliary_loss_mlp": 0.0104362,
      "balance_loss_clip": 1.06996584,
      "balance_loss_mlp": 1.03156805,
      "epoch": 0.07106354836770276,
      "flos": 24645098534400.0,
      "grad_norm": 1.8491565064679454,
      "language_loss": 0.84544909,
      "learning_rate": 3.982386179311399e-06,
      "loss": 0.86836433,
      "num_input_tokens_seen": 12621725,
      "step": 591,
      "time_per_iteration": 2.710681915283203
    },
    {
      "auxiliary_loss_clip": 0.01261031,
      "auxiliary_loss_mlp": 0.01038623,
      "balance_loss_clip": 1.07046509,
      "balance_loss_mlp": 1.02658248,
      "epoch": 0.07118379125834184,
      "flos": 16217223649920.0,
      "grad_norm": 2.4857701852251686,
      "language_loss": 0.87258887,
      "learning_rate": 3.982282873988075e-06,
      "loss": 0.89558542,
      "num_input_tokens_seen": 12639600,
      "step": 592,
      "time_per_iteration": 2.6381425857543945
    },
    {
      "auxiliary_loss_clip": 0.01226121,
      "auxiliary_loss_mlp": 0.01040593,
      "balance_loss_clip": 1.06600177,
      "balance_loss_mlp": 1.02852941,
      "epoch": 0.07130403414898094,
      "flos": 19719986227200.0,
      "grad_norm": 3.4798417580294325,
      "language_loss": 0.87028968,
      "learning_rate": 3.982179267954016e-06,
      "loss": 0.89295679,
      "num_input_tokens_seen": 12660030,
      "step": 593,
      "time_per_iteration": 3.6259536743164062
    },
    {
      "auxiliary_loss_clip": 0.01230899,
      "auxiliary_loss_mlp": 0.01043108,
      "balance_loss_clip": 1.06574273,
      "balance_loss_mlp": 1.0307821,
      "epoch": 0.07142427703962004,
      "flos": 21871933009920.0,
      "grad_norm": 14.64581398888859,
      "language_loss": 0.96350008,
      "learning_rate": 3.982075361224937e-06,
      "loss": 0.98624021,
      "num_input_tokens_seen": 12678395,
      "step": 594,
      "time_per_iteration": 2.659471035003662
    },
    {
      "auxiliary_loss_clip": 0.01202641,
      "auxiliary_loss_mlp": 0.00765438,
      "balance_loss_clip": 1.0641284,
      "balance_loss_mlp": 1.00075483,
      "epoch": 0.07154451993025912,
      "flos": 18296595002880.0,
      "grad_norm": 1.821260855450604,
      "language_loss": 0.88337958,
      "learning_rate": 3.981971153816602e-06,
      "loss": 0.90306038,
      "num_input_tokens_seen": 12696000,
      "step": 595,
      "time_per_iteration": 2.757218837738037
    },
    {
      "auxiliary_loss_clip": 0.01241433,
      "auxiliary_loss_mlp": 0.01043124,
      "balance_loss_clip": 1.06908393,
      "balance_loss_mlp": 1.03207374,
      "epoch": 0.07166476282089822,
      "flos": 22160690444160.0,
      "grad_norm": 1.7469836944106585,
      "language_loss": 0.96149433,
      "learning_rate": 3.981866645744819e-06,
      "loss": 0.98433989,
      "num_input_tokens_seen": 12716715,
      "step": 596,
      "time_per_iteration": 4.1318182945251465
    },
    {
      "auxiliary_loss_clip": 0.01259159,
      "auxiliary_loss_mlp": 0.01045563,
      "balance_loss_clip": 1.06965959,
      "balance_loss_mlp": 1.03318858,
      "epoch": 0.0717850057115373,
      "flos": 14136343925760.0,
      "grad_norm": 2.8021764022267743,
      "language_loss": 0.81631577,
      "learning_rate": 3.9817618370254416e-06,
      "loss": 0.83936304,
      "num_input_tokens_seen": 12733370,
      "step": 597,
      "time_per_iteration": 2.641650438308716
    },
    {
      "auxiliary_loss_clip": 0.01257703,
      "auxiliary_loss_mlp": 0.01041444,
      "balance_loss_clip": 1.06639063,
      "balance_loss_mlp": 1.02960038,
      "epoch": 0.0719052486021764,
      "flos": 30917794412160.0,
      "grad_norm": 2.8674089847243422,
      "language_loss": 0.87726855,
      "learning_rate": 3.9816567276743684e-06,
      "loss": 0.90026009,
      "num_input_tokens_seen": 12753235,
      "step": 598,
      "time_per_iteration": 2.8279225826263428
    },
    {
      "auxiliary_loss_clip": 0.01208453,
      "auxiliary_loss_mlp": 0.01044361,
      "balance_loss_clip": 1.06506872,
      "balance_loss_mlp": 1.03228486,
      "epoch": 0.0720254914928155,
      "flos": 21287019939840.0,
      "grad_norm": 2.041619128041891,
      "language_loss": 0.77295321,
      "learning_rate": 3.9815513177075466e-06,
      "loss": 0.79548132,
      "num_input_tokens_seen": 12772020,
      "step": 599,
      "time_per_iteration": 2.7407066822052
    },
    {
      "auxiliary_loss_clip": 0.01236182,
      "auxiliary_loss_mlp": 0.00764438,
      "balance_loss_clip": 1.07066846,
      "balance_loss_mlp": 1.00071287,
      "epoch": 0.07214573438345458,
      "flos": 27819170732160.0,
      "grad_norm": 1.8590189621645046,
      "language_loss": 0.69975054,
      "learning_rate": 3.9814456071409646e-06,
      "loss": 0.71975672,
      "num_input_tokens_seen": 12792555,
      "step": 600,
      "time_per_iteration": 2.7428464889526367
    },
    {
      "auxiliary_loss_clip": 0.0120764,
      "auxiliary_loss_mlp": 0.01051465,
      "balance_loss_clip": 1.0635376,
      "balance_loss_mlp": 1.03899539,
      "epoch": 0.07226597727409367,
      "flos": 25483576688640.0,
      "grad_norm": 2.735359036362914,
      "language_loss": 0.85217196,
      "learning_rate": 3.981339595990659e-06,
      "loss": 0.87476301,
      "num_input_tokens_seen": 12811085,
      "step": 601,
      "time_per_iteration": 2.8044188022613525
    },
    {
      "auxiliary_loss_clip": 0.01223775,
      "auxiliary_loss_mlp": 0.01047576,
      "balance_loss_clip": 1.06838036,
      "balance_loss_mlp": 1.03525591,
      "epoch": 0.07238622016473276,
      "flos": 23513840622720.0,
      "grad_norm": 2.0291265857174015,
      "language_loss": 0.81211948,
      "learning_rate": 3.981233284272713e-06,
      "loss": 0.83483303,
      "num_input_tokens_seen": 12830830,
      "step": 602,
      "time_per_iteration": 2.782904624938965
    },
    {
      "auxiliary_loss_clip": 0.01237209,
      "auxiliary_loss_mlp": 0.01046618,
      "balance_loss_clip": 1.06793237,
      "balance_loss_mlp": 1.03583574,
      "epoch": 0.07250646305537185,
      "flos": 25453519983360.0,
      "grad_norm": 2.7330197497330952,
      "language_loss": 0.90195215,
      "learning_rate": 3.981126672003253e-06,
      "loss": 0.92479044,
      "num_input_tokens_seen": 12853505,
      "step": 603,
      "time_per_iteration": 2.840543270111084
    },
    {
      "auxiliary_loss_clip": 0.01247288,
      "auxiliary_loss_mlp": 0.01045229,
      "balance_loss_clip": 1.06806612,
      "balance_loss_mlp": 1.03386879,
      "epoch": 0.07262670594601094,
      "flos": 27155038216320.0,
      "grad_norm": 2.700268743016048,
      "language_loss": 0.77888298,
      "learning_rate": 3.981019759198451e-06,
      "loss": 0.80180812,
      "num_input_tokens_seen": 12872455,
      "step": 604,
      "time_per_iteration": 2.712303876876831
    },
    {
      "auxiliary_loss_clip": 0.01234207,
      "auxiliary_loss_mlp": 0.01039811,
      "balance_loss_clip": 1.06718588,
      "balance_loss_mlp": 1.02862942,
      "epoch": 0.07274694883665003,
      "flos": 26651607148800.0,
      "grad_norm": 2.319479110856928,
      "language_loss": 0.84231746,
      "learning_rate": 3.980912545874528e-06,
      "loss": 0.86505771,
      "num_input_tokens_seen": 12892620,
      "step": 605,
      "time_per_iteration": 2.8592605590820312
    },
    {
      "auxiliary_loss_clip": 0.01260489,
      "auxiliary_loss_mlp": 0.01038379,
      "balance_loss_clip": 1.07255316,
      "balance_loss_mlp": 1.02652395,
      "epoch": 0.07286719172728913,
      "flos": 29862344154240.0,
      "grad_norm": 3.024631864408276,
      "language_loss": 0.85400486,
      "learning_rate": 3.980805032047746e-06,
      "loss": 0.8769936,
      "num_input_tokens_seen": 12914090,
      "step": 606,
      "time_per_iteration": 2.7431564331054688
    },
    {
      "auxiliary_loss_clip": 0.01214022,
      "auxiliary_loss_mlp": 0.01046132,
      "balance_loss_clip": 1.06259549,
      "balance_loss_mlp": 1.03341842,
      "epoch": 0.07298743461792821,
      "flos": 17382057799680.0,
      "grad_norm": 1.9790326726480278,
      "language_loss": 0.81206286,
      "learning_rate": 3.980697217734415e-06,
      "loss": 0.83466434,
      "num_input_tokens_seen": 12931830,
      "step": 607,
      "time_per_iteration": 2.7150204181671143
    },
    {
      "auxiliary_loss_clip": 0.01220408,
      "auxiliary_loss_mlp": 0.01037413,
      "balance_loss_clip": 1.06850481,
      "balance_loss_mlp": 1.02553928,
      "epoch": 0.07310767750856731,
      "flos": 19498201701120.0,
      "grad_norm": 1.8669418696192024,
      "language_loss": 0.9164803,
      "learning_rate": 3.980589102950891e-06,
      "loss": 0.93905854,
      "num_input_tokens_seen": 12949995,
      "step": 608,
      "time_per_iteration": 2.7593629360198975
    },
    {
      "auxiliary_loss_clip": 0.01226869,
      "auxiliary_loss_mlp": 0.01052001,
      "balance_loss_clip": 1.06528926,
      "balance_loss_mlp": 1.03968024,
      "epoch": 0.07322792039920639,
      "flos": 29168693637120.0,
      "grad_norm": 2.7910210573704055,
      "language_loss": 0.7598232,
      "learning_rate": 3.9804806877135755e-06,
      "loss": 0.78261185,
      "num_input_tokens_seen": 12968040,
      "step": 609,
      "time_per_iteration": 2.729274272918701
    },
    {
      "auxiliary_loss_clip": 0.01247036,
      "auxiliary_loss_mlp": 0.00765929,
      "balance_loss_clip": 1.06828082,
      "balance_loss_mlp": 1.00065541,
      "epoch": 0.07334816328984549,
      "flos": 23477822259840.0,
      "grad_norm": 2.384786032005252,
      "language_loss": 0.86178041,
      "learning_rate": 3.980371972038915e-06,
      "loss": 0.88191009,
      "num_input_tokens_seen": 12988530,
      "step": 610,
      "time_per_iteration": 2.7808494567871094
    },
    {
      "auxiliary_loss_clip": 0.01246448,
      "auxiliary_loss_mlp": 0.01052788,
      "balance_loss_clip": 1.07251251,
      "balance_loss_mlp": 1.04059863,
      "epoch": 0.07346840618048459,
      "flos": 22962467877120.0,
      "grad_norm": 2.153470184835652,
      "language_loss": 0.8431437,
      "learning_rate": 3.980262955943399e-06,
      "loss": 0.86613607,
      "num_input_tokens_seen": 13008195,
      "step": 611,
      "time_per_iteration": 2.7578516006469727
    },
    {
      "auxiliary_loss_clip": 0.01251924,
      "auxiliary_loss_mlp": 0.00764932,
      "balance_loss_clip": 1.07042646,
      "balance_loss_mlp": 1.00061524,
      "epoch": 0.07358864907112367,
      "flos": 17673903803520.0,
      "grad_norm": 2.37758242512816,
      "language_loss": 0.86917543,
      "learning_rate": 3.980153639443569e-06,
      "loss": 0.88934398,
      "num_input_tokens_seen": 13024180,
      "step": 612,
      "time_per_iteration": 2.6975042819976807
    },
    {
      "auxiliary_loss_clip": 0.01277694,
      "auxiliary_loss_mlp": 0.01045521,
      "balance_loss_clip": 1.07610679,
      "balance_loss_mlp": 1.0331527,
      "epoch": 0.07370889196176277,
      "flos": 24097029840000.0,
      "grad_norm": 2.022729607692615,
      "language_loss": 0.80167735,
      "learning_rate": 3.980044022556005e-06,
      "loss": 0.82490945,
      "num_input_tokens_seen": 13043865,
      "step": 613,
      "time_per_iteration": 2.669982671737671
    },
    {
      "auxiliary_loss_clip": 0.01196107,
      "auxiliary_loss_mlp": 0.01052241,
      "balance_loss_clip": 1.06260204,
      "balance_loss_mlp": 1.03946781,
      "epoch": 0.07382913485240185,
      "flos": 25885919905920.0,
      "grad_norm": 2.5599948372318235,
      "language_loss": 0.72994971,
      "learning_rate": 3.9799341052973375e-06,
      "loss": 0.75243318,
      "num_input_tokens_seen": 13063700,
      "step": 614,
      "time_per_iteration": 2.822356939315796
    },
    {
      "auxiliary_loss_clip": 0.01243909,
      "auxiliary_loss_mlp": 0.01049006,
      "balance_loss_clip": 1.07273412,
      "balance_loss_mlp": 1.03632212,
      "epoch": 0.07394937774304094,
      "flos": 16873850223360.0,
      "grad_norm": 2.6836400587955374,
      "language_loss": 0.7538473,
      "learning_rate": 3.979823887684241e-06,
      "loss": 0.77677643,
      "num_input_tokens_seen": 13082640,
      "step": 615,
      "time_per_iteration": 3.696251630783081
    },
    {
      "auxiliary_loss_clip": 0.01226643,
      "auxiliary_loss_mlp": 0.01054761,
      "balance_loss_clip": 1.06521058,
      "balance_loss_mlp": 1.04203546,
      "epoch": 0.07406962063368003,
      "flos": 20703471586560.0,
      "grad_norm": 2.903973524446849,
      "language_loss": 0.84505206,
      "learning_rate": 3.979713369733434e-06,
      "loss": 0.8678661,
      "num_input_tokens_seen": 13100505,
      "step": 616,
      "time_per_iteration": 3.6192123889923096
    },
    {
      "auxiliary_loss_clip": 0.0120486,
      "auxiliary_loss_mlp": 0.01055688,
      "balance_loss_clip": 1.06227469,
      "balance_loss_mlp": 1.04146671,
      "epoch": 0.07418986352431912,
      "flos": 21430985650560.0,
      "grad_norm": 2.1077540259755376,
      "language_loss": 0.84925747,
      "learning_rate": 3.979602551461683e-06,
      "loss": 0.87186301,
      "num_input_tokens_seen": 13121285,
      "step": 617,
      "time_per_iteration": 2.784346103668213
    },
    {
      "auxiliary_loss_clip": 0.01238601,
      "auxiliary_loss_mlp": 0.0104195,
      "balance_loss_clip": 1.06497753,
      "balance_loss_mlp": 1.03004706,
      "epoch": 0.07431010641495822,
      "flos": 12021133777920.0,
      "grad_norm": 2.633343848149125,
      "language_loss": 0.91413927,
      "learning_rate": 3.979491432885799e-06,
      "loss": 0.93694484,
      "num_input_tokens_seen": 13137550,
      "step": 618,
      "time_per_iteration": 2.6806142330169678
    },
    {
      "auxiliary_loss_clip": 0.01197699,
      "auxiliary_loss_mlp": 0.01041143,
      "balance_loss_clip": 1.0629741,
      "balance_loss_mlp": 1.02927589,
      "epoch": 0.0744303493055973,
      "flos": 20957575374720.0,
      "grad_norm": 2.437391330959691,
      "language_loss": 0.83031648,
      "learning_rate": 3.97938001402264e-06,
      "loss": 0.85270488,
      "num_input_tokens_seen": 13156675,
      "step": 619,
      "time_per_iteration": 3.76682710647583
    },
    {
      "auxiliary_loss_clip": 0.01275137,
      "auxiliary_loss_mlp": 0.01046452,
      "balance_loss_clip": 1.07344544,
      "balance_loss_mlp": 1.03474522,
      "epoch": 0.0745505921962364,
      "flos": 16253134272000.0,
      "grad_norm": 6.248499905082643,
      "language_loss": 0.79955256,
      "learning_rate": 3.979268294889105e-06,
      "loss": 0.82276845,
      "num_input_tokens_seen": 13172225,
      "step": 620,
      "time_per_iteration": 2.6797285079956055
    },
    {
      "auxiliary_loss_clip": 0.01225198,
      "auxiliary_loss_mlp": 0.01043698,
      "balance_loss_clip": 1.06931496,
      "balance_loss_mlp": 1.03077567,
      "epoch": 0.07467083508687548,
      "flos": 50944635550080.0,
      "grad_norm": 1.7775191980152718,
      "language_loss": 0.74326718,
      "learning_rate": 3.979156275502143e-06,
      "loss": 0.7659561,
      "num_input_tokens_seen": 13195885,
      "step": 621,
      "time_per_iteration": 3.041588306427002
    },
    {
      "auxiliary_loss_clip": 0.01242409,
      "auxiliary_loss_mlp": 0.0076513,
      "balance_loss_clip": 1.068573,
      "balance_loss_mlp": 1.00055027,
      "epoch": 0.07479107797751458,
      "flos": 17529686697600.0,
      "grad_norm": 2.9412825866100687,
      "language_loss": 0.91613424,
      "learning_rate": 3.979043955878749e-06,
      "loss": 0.93620968,
      "num_input_tokens_seen": 13213730,
      "step": 622,
      "time_per_iteration": 3.7438700199127197
    },
    {
      "auxiliary_loss_clip": 0.01203169,
      "auxiliary_loss_mlp": 0.0104903,
      "balance_loss_clip": 1.06836772,
      "balance_loss_mlp": 1.03648353,
      "epoch": 0.07491132086815366,
      "flos": 23473943591040.0,
      "grad_norm": 1.904861424529706,
      "language_loss": 0.83460808,
      "learning_rate": 3.978931336035959e-06,
      "loss": 0.85713005,
      "num_input_tokens_seen": 13232540,
      "step": 623,
      "time_per_iteration": 2.855055332183838
    },
    {
      "auxiliary_loss_clip": 0.01218546,
      "auxiliary_loss_mlp": 0.0076474,
      "balance_loss_clip": 1.06841135,
      "balance_loss_mlp": 1.00057721,
      "epoch": 0.07503156375879276,
      "flos": 20157557708160.0,
      "grad_norm": 3.2606850275221153,
      "language_loss": 0.82835734,
      "learning_rate": 3.9788184159908595e-06,
      "loss": 0.84819019,
      "num_input_tokens_seen": 13249670,
      "step": 624,
      "time_per_iteration": 2.7680704593658447
    },
    {
      "auxiliary_loss_clip": 0.01200072,
      "auxiliary_loss_mlp": 0.01047288,
      "balance_loss_clip": 1.06015754,
      "balance_loss_mlp": 1.03642786,
      "epoch": 0.07515180664943186,
      "flos": 15115519653120.0,
      "grad_norm": 4.284523299701527,
      "language_loss": 0.82478976,
      "learning_rate": 3.97870519576058e-06,
      "loss": 0.84726334,
      "num_input_tokens_seen": 13266095,
      "step": 625,
      "time_per_iteration": 2.793337106704712
    },
    {
      "auxiliary_loss_clip": 0.01208386,
      "auxiliary_loss_mlp": 0.00764738,
      "balance_loss_clip": 1.05975258,
      "balance_loss_mlp": 1.00067973,
      "epoch": 0.07527204954007094,
      "flos": 21287702298240.0,
      "grad_norm": 2.7363188253860544,
      "language_loss": 0.81452644,
      "learning_rate": 3.978591675362295e-06,
      "loss": 0.83425772,
      "num_input_tokens_seen": 13284810,
      "step": 626,
      "time_per_iteration": 2.8018057346343994
    },
    {
      "auxiliary_loss_clip": 0.01253205,
      "auxiliary_loss_mlp": 0.01039297,
      "balance_loss_clip": 1.07201135,
      "balance_loss_mlp": 1.02768016,
      "epoch": 0.07539229243071004,
      "flos": 21324187537920.0,
      "grad_norm": 2.108183105600102,
      "language_loss": 0.87689835,
      "learning_rate": 3.978477854813226e-06,
      "loss": 0.89982343,
      "num_input_tokens_seen": 13304150,
      "step": 627,
      "time_per_iteration": 2.6955528259277344
    },
    {
      "auxiliary_loss_clip": 0.01201127,
      "auxiliary_loss_mlp": 0.01045857,
      "balance_loss_clip": 1.06132579,
      "balance_loss_mlp": 1.03403711,
      "epoch": 0.07551253532134912,
      "flos": 13042540920960.0,
      "grad_norm": 3.047895472754248,
      "language_loss": 0.82292295,
      "learning_rate": 3.97836373413064e-06,
      "loss": 0.84539282,
      "num_input_tokens_seen": 13322205,
      "step": 628,
      "time_per_iteration": 2.7342066764831543
    },
    {
      "auxiliary_loss_clip": 0.01189198,
      "auxiliary_loss_mlp": 0.00765089,
      "balance_loss_clip": 1.05454803,
      "balance_loss_mlp": 1.00052655,
      "epoch": 0.07563277821198822,
      "flos": 19208761908480.0,
      "grad_norm": 2.3137676914005194,
      "language_loss": 0.74404138,
      "learning_rate": 3.978249313331848e-06,
      "loss": 0.76358426,
      "num_input_tokens_seen": 13340435,
      "step": 629,
      "time_per_iteration": 2.8635952472686768
    },
    {
      "auxiliary_loss_clip": 0.01257866,
      "auxiliary_loss_mlp": 0.01044302,
      "balance_loss_clip": 1.07239354,
      "balance_loss_mlp": 1.0318265,
      "epoch": 0.07575302110262731,
      "flos": 19537200892800.0,
      "grad_norm": 2.8533301159375144,
      "language_loss": 0.61618102,
      "learning_rate": 3.978134592434208e-06,
      "loss": 0.63920271,
      "num_input_tokens_seen": 13358185,
      "step": 630,
      "time_per_iteration": 3.0580978393554688
    },
    {
      "auxiliary_loss_clip": 0.01177992,
      "auxiliary_loss_mlp": 0.01004392,
      "balance_loss_clip": 1.05571914,
      "balance_loss_mlp": 1.00002885,
      "epoch": 0.0758732639932664,
      "flos": 67961808017280.0,
      "grad_norm": 1.008827848601039,
      "language_loss": 0.59417611,
      "learning_rate": 3.978019571455123e-06,
      "loss": 0.61599994,
      "num_input_tokens_seen": 13410130,
      "step": 631,
      "time_per_iteration": 3.315131187438965
    },
    {
      "auxiliary_loss_clip": 0.01233712,
      "auxiliary_loss_mlp": 0.00765059,
      "balance_loss_clip": 1.067065,
      "balance_loss_mlp": 1.00049245,
      "epoch": 0.07599350688390549,
      "flos": 18989204025600.0,
      "grad_norm": 2.0481622294515,
      "language_loss": 0.84038639,
      "learning_rate": 3.977904250412042e-06,
      "loss": 0.86037409,
      "num_input_tokens_seen": 13429085,
      "step": 632,
      "time_per_iteration": 2.7544379234313965
    },
    {
      "auxiliary_loss_clip": 0.01205202,
      "auxiliary_loss_mlp": 0.01051452,
      "balance_loss_clip": 1.06298423,
      "balance_loss_mlp": 1.03952503,
      "epoch": 0.07611374977454458,
      "flos": 21069006341760.0,
      "grad_norm": 2.3780791918256887,
      "language_loss": 0.8577776,
      "learning_rate": 3.97778862932246e-06,
      "loss": 0.88034415,
      "num_input_tokens_seen": 13446250,
      "step": 633,
      "time_per_iteration": 2.7744345664978027
    },
    {
      "auxiliary_loss_clip": 0.01237175,
      "auxiliary_loss_mlp": 0.01044299,
      "balance_loss_clip": 1.06440163,
      "balance_loss_mlp": 1.03268838,
      "epoch": 0.07623399266518367,
      "flos": 18514536773760.0,
      "grad_norm": 2.503477117913484,
      "language_loss": 0.94179893,
      "learning_rate": 3.9776727082039144e-06,
      "loss": 0.96461368,
      "num_input_tokens_seen": 13463220,
      "step": 634,
      "time_per_iteration": 2.692260503768921
    },
    {
      "auxiliary_loss_clip": 0.01178931,
      "auxiliary_loss_mlp": 0.01003427,
      "balance_loss_clip": 1.0545404,
      "balance_loss_mlp": 0.99932605,
      "epoch": 0.07635423555582276,
      "flos": 44663036077440.0,
      "grad_norm": 0.803817422789986,
      "language_loss": 0.55483758,
      "learning_rate": 3.977556487073991e-06,
      "loss": 0.57666117,
      "num_input_tokens_seen": 13517775,
      "step": 635,
      "time_per_iteration": 3.194889783859253
    },
    {
      "auxiliary_loss_clip": 0.01235775,
      "auxiliary_loss_mlp": 0.01048871,
      "balance_loss_clip": 1.06370485,
      "balance_loss_mlp": 1.03687263,
      "epoch": 0.07647447844646185,
      "flos": 21761148487680.0,
      "grad_norm": 1.7146814617461108,
      "language_loss": 0.81834447,
      "learning_rate": 3.97743996595032e-06,
      "loss": 0.84119093,
      "num_input_tokens_seen": 13537815,
      "step": 636,
      "time_per_iteration": 2.752675771713257
    },
    {
      "auxiliary_loss_clip": 0.01229375,
      "auxiliary_loss_mlp": 0.01044938,
      "balance_loss_clip": 1.06623518,
      "balance_loss_mlp": 1.03264201,
      "epoch": 0.07659472133710095,
      "flos": 23806799948160.0,
      "grad_norm": 1.6697436443983626,
      "language_loss": 0.81886601,
      "learning_rate": 3.9773231448505804e-06,
      "loss": 0.84160912,
      "num_input_tokens_seen": 13559605,
      "step": 637,
      "time_per_iteration": 2.7066802978515625
    },
    {
      "auxiliary_loss_clip": 0.01190152,
      "auxiliary_loss_mlp": 0.01041249,
      "balance_loss_clip": 1.05871463,
      "balance_loss_mlp": 1.02931643,
      "epoch": 0.07671496422774003,
      "flos": 21469984842240.0,
      "grad_norm": 3.249828694791789,
      "language_loss": 0.78277349,
      "learning_rate": 3.977206023792491e-06,
      "loss": 0.80508757,
      "num_input_tokens_seen": 13579495,
      "step": 638,
      "time_per_iteration": 2.8496875762939453
    },
    {
      "auxiliary_loss_clip": 0.01232111,
      "auxiliary_loss_mlp": 0.01054099,
      "balance_loss_clip": 1.06630635,
      "balance_loss_mlp": 1.04158854,
      "epoch": 0.07683520711837913,
      "flos": 16980971558400.0,
      "grad_norm": 2.4294207493321487,
      "language_loss": 0.81057286,
      "learning_rate": 3.97708860279382e-06,
      "loss": 0.83343494,
      "num_input_tokens_seen": 13597605,
      "step": 639,
      "time_per_iteration": 2.70994234085083
    },
    {
      "auxiliary_loss_clip": 0.01229052,
      "auxiliary_loss_mlp": 0.01042122,
      "balance_loss_clip": 1.06611156,
      "balance_loss_mlp": 1.03074384,
      "epoch": 0.07695545000901821,
      "flos": 23476744851840.0,
      "grad_norm": 1.8415533550929835,
      "language_loss": 0.77996045,
      "learning_rate": 3.97697088187238e-06,
      "loss": 0.80267227,
      "num_input_tokens_seen": 13618120,
      "step": 640,
      "time_per_iteration": 2.7542877197265625
    },
    {
      "auxiliary_loss_clip": 0.01240412,
      "auxiliary_loss_mlp": 0.01046206,
      "balance_loss_clip": 1.06613076,
      "balance_loss_mlp": 1.03505397,
      "epoch": 0.07707569289965731,
      "flos": 17634258167040.0,
      "grad_norm": 2.863783042913633,
      "language_loss": 0.91886061,
      "learning_rate": 3.976852861046029e-06,
      "loss": 0.9417268,
      "num_input_tokens_seen": 13634735,
      "step": 641,
      "time_per_iteration": 3.675323247909546
    },
    {
      "auxiliary_loss_clip": 0.01257901,
      "auxiliary_loss_mlp": 0.01052055,
      "balance_loss_clip": 1.07220209,
      "balance_loss_mlp": 1.04061663,
      "epoch": 0.0771959357902964,
      "flos": 25775674087680.0,
      "grad_norm": 1.885677702898158,
      "language_loss": 0.80191356,
      "learning_rate": 3.97673454033267e-06,
      "loss": 0.8250131,
      "num_input_tokens_seen": 13656835,
      "step": 642,
      "time_per_iteration": 2.7264585494995117
    },
    {
      "auxiliary_loss_clip": 0.01247032,
      "auxiliary_loss_mlp": 0.01046415,
      "balance_loss_clip": 1.06794977,
      "balance_loss_mlp": 1.03489304,
      "epoch": 0.07731617868093549,
      "flos": 19828651847040.0,
      "grad_norm": 2.063436541408446,
      "language_loss": 0.8278383,
      "learning_rate": 3.976615919750254e-06,
      "loss": 0.8507728,
      "num_input_tokens_seen": 13674535,
      "step": 643,
      "time_per_iteration": 2.751925230026245
    },
    {
      "auxiliary_loss_clip": 0.012384,
      "auxiliary_loss_mlp": 0.01048092,
      "balance_loss_clip": 1.06817102,
      "balance_loss_mlp": 1.03652894,
      "epoch": 0.07743642157157458,
      "flos": 21324654414720.0,
      "grad_norm": 1.8697262330304358,
      "language_loss": 0.86876953,
      "learning_rate": 3.976496999316775e-06,
      "loss": 0.89163452,
      "num_input_tokens_seen": 13693290,
      "step": 644,
      "time_per_iteration": 2.69700288772583
    },
    {
      "auxiliary_loss_clip": 0.01197787,
      "auxiliary_loss_mlp": 0.01044876,
      "balance_loss_clip": 1.05900645,
      "balance_loss_mlp": 1.03153086,
      "epoch": 0.07755666446221367,
      "flos": 19969133938560.0,
      "grad_norm": 2.6563386825620463,
      "language_loss": 0.83825344,
      "learning_rate": 3.976377779050271e-06,
      "loss": 0.86068004,
      "num_input_tokens_seen": 13711420,
      "step": 645,
      "time_per_iteration": 3.70819091796875
    },
    {
      "auxiliary_loss_clip": 0.01271536,
      "auxiliary_loss_mlp": 0.01052598,
      "balance_loss_clip": 1.07069182,
      "balance_loss_mlp": 1.0400691,
      "epoch": 0.07767690735285276,
      "flos": 23623224514560.0,
      "grad_norm": 2.3881957718837024,
      "language_loss": 0.84484065,
      "learning_rate": 3.976258258968831e-06,
      "loss": 0.86808199,
      "num_input_tokens_seen": 13729965,
      "step": 646,
      "time_per_iteration": 2.701023817062378
    },
    {
      "auxiliary_loss_clip": 0.01220532,
      "auxiliary_loss_mlp": 0.0104112,
      "balance_loss_clip": 1.06303239,
      "balance_loss_mlp": 1.02965188,
      "epoch": 0.07779715024349185,
      "flos": 22236246702720.0,
      "grad_norm": 2.199887088164321,
      "language_loss": 0.74001765,
      "learning_rate": 3.976138439090583e-06,
      "loss": 0.76263416,
      "num_input_tokens_seen": 13748045,
      "step": 647,
      "time_per_iteration": 2.781933069229126
    },
    {
      "auxiliary_loss_clip": 0.01223575,
      "auxiliary_loss_mlp": 0.01037858,
      "balance_loss_clip": 1.06610537,
      "balance_loss_mlp": 1.02446485,
      "epoch": 0.07791739313413094,
      "flos": 20955097336320.0,
      "grad_norm": 2.464218140580502,
      "language_loss": 0.8521772,
      "learning_rate": 3.976018319433706e-06,
      "loss": 0.87479156,
      "num_input_tokens_seen": 13765590,
      "step": 648,
      "time_per_iteration": 3.67681884765625
    },
    {
      "auxiliary_loss_clip": 0.01255038,
      "auxiliary_loss_mlp": 0.01046219,
      "balance_loss_clip": 1.07127476,
      "balance_loss_mlp": 1.03360057,
      "epoch": 0.07803763602477004,
      "flos": 19312327797120.0,
      "grad_norm": 2.503846722390071,
      "language_loss": 0.91356802,
      "learning_rate": 3.9758979000164205e-06,
      "loss": 0.93658054,
      "num_input_tokens_seen": 13782410,
      "step": 649,
      "time_per_iteration": 2.6701207160949707
    },
    {
      "auxiliary_loss_clip": 0.01261037,
      "auxiliary_loss_mlp": 0.01043784,
      "balance_loss_clip": 1.07374811,
      "balance_loss_mlp": 1.03149295,
      "epoch": 0.07815787891540912,
      "flos": 22710806213760.0,
      "grad_norm": 1.7999464756657582,
      "language_loss": 0.71888119,
      "learning_rate": 3.975777180856995e-06,
      "loss": 0.74192941,
      "num_input_tokens_seen": 13801530,
      "step": 650,
      "time_per_iteration": 2.6976633071899414
    },
    {
      "auxiliary_loss_clip": 0.01237295,
      "auxiliary_loss_mlp": 0.01043992,
      "balance_loss_clip": 1.0664525,
      "balance_loss_mlp": 1.03080142,
      "epoch": 0.07827812180604822,
      "flos": 22711129436160.0,
      "grad_norm": 2.0238144074644064,
      "language_loss": 0.86517888,
      "learning_rate": 3.975656161973742e-06,
      "loss": 0.88799179,
      "num_input_tokens_seen": 13820615,
      "step": 651,
      "time_per_iteration": 2.7228567600250244
    },
    {
      "auxiliary_loss_clip": 0.01187925,
      "auxiliary_loss_mlp": 0.01042157,
      "balance_loss_clip": 1.06136298,
      "balance_loss_mlp": 1.02939606,
      "epoch": 0.0783983646966873,
      "flos": 21725597001600.0,
      "grad_norm": 2.2773217531824885,
      "language_loss": 0.89070278,
      "learning_rate": 3.9755348433850194e-06,
      "loss": 0.91300362,
      "num_input_tokens_seen": 13835955,
      "step": 652,
      "time_per_iteration": 2.7503139972686768
    },
    {
      "auxiliary_loss_clip": 0.01113667,
      "auxiliary_loss_mlp": 0.01014087,
      "balance_loss_clip": 1.03575695,
      "balance_loss_mlp": 1.01005769,
      "epoch": 0.0785186075873264,
      "flos": 60640877537280.0,
      "grad_norm": 0.9740718228222455,
      "language_loss": 0.63713819,
      "learning_rate": 3.975413225109232e-06,
      "loss": 0.65841568,
      "num_input_tokens_seen": 13896505,
      "step": 653,
      "time_per_iteration": 3.4016127586364746
    },
    {
      "auxiliary_loss_clip": 0.01257813,
      "auxiliary_loss_mlp": 0.01040468,
      "balance_loss_clip": 1.06880677,
      "balance_loss_mlp": 1.02756321,
      "epoch": 0.0786388504779655,
      "flos": 23877902920320.0,
      "grad_norm": 3.879827586418496,
      "language_loss": 0.93185699,
      "learning_rate": 3.975291307164829e-06,
      "loss": 0.95483977,
      "num_input_tokens_seen": 13915150,
      "step": 654,
      "time_per_iteration": 2.7756288051605225
    },
    {
      "auxiliary_loss_clip": 0.01244414,
      "auxiliary_loss_mlp": 0.01043519,
      "balance_loss_clip": 1.06705654,
      "balance_loss_mlp": 1.03218198,
      "epoch": 0.07875909336860458,
      "flos": 15158684822400.0,
      "grad_norm": 2.6880796789301376,
      "language_loss": 0.85330009,
      "learning_rate": 3.975169089570306e-06,
      "loss": 0.87617946,
      "num_input_tokens_seen": 13933525,
      "step": 655,
      "time_per_iteration": 2.7885332107543945
    },
    {
      "auxiliary_loss_clip": 0.0126707,
      "auxiliary_loss_mlp": 0.01044651,
      "balance_loss_clip": 1.07049799,
      "balance_loss_mlp": 1.03256941,
      "epoch": 0.07887933625924368,
      "flos": 22236857233920.0,
      "grad_norm": 2.6857075397535173,
      "language_loss": 0.91566646,
      "learning_rate": 3.975046572344202e-06,
      "loss": 0.93878365,
      "num_input_tokens_seen": 13949985,
      "step": 656,
      "time_per_iteration": 2.655366897583008
    },
    {
      "auxiliary_loss_clip": 0.01221409,
      "auxiliary_loss_mlp": 0.01043759,
      "balance_loss_clip": 1.06354237,
      "balance_loss_mlp": 1.0310576,
      "epoch": 0.07899957914988276,
      "flos": 20777734955520.0,
      "grad_norm": 6.982016344485925,
      "language_loss": 0.71145278,
      "learning_rate": 3.974923755505103e-06,
      "loss": 0.73410445,
      "num_input_tokens_seen": 13969215,
      "step": 657,
      "time_per_iteration": 2.75472354888916
    },
    {
      "auxiliary_loss_clip": 0.01239395,
      "auxiliary_loss_mlp": 0.01053223,
      "balance_loss_clip": 1.06495976,
      "balance_loss_mlp": 1.04069448,
      "epoch": 0.07911982204052186,
      "flos": 23003047267200.0,
      "grad_norm": 1.7648982397761483,
      "language_loss": 0.91024214,
      "learning_rate": 3.974800639071641e-06,
      "loss": 0.93316841,
      "num_input_tokens_seen": 13989935,
      "step": 658,
      "time_per_iteration": 2.786238193511963
    },
    {
      "auxiliary_loss_clip": 0.01132458,
      "auxiliary_loss_mlp": 0.01044587,
      "balance_loss_clip": 1.05495787,
      "balance_loss_mlp": 1.03280962,
      "epoch": 0.07924006493116094,
      "flos": 23111389664640.0,
      "grad_norm": 2.328959275374011,
      "language_loss": 1.00518072,
      "learning_rate": 3.974677223062492e-06,
      "loss": 1.02695107,
      "num_input_tokens_seen": 14007150,
      "step": 659,
      "time_per_iteration": 3.1076600551605225
    },
    {
      "auxiliary_loss_clip": 0.01244747,
      "auxiliary_loss_mlp": 0.01041284,
      "balance_loss_clip": 1.06597924,
      "balance_loss_mlp": 1.02928567,
      "epoch": 0.07936030782180004,
      "flos": 16472153450880.0,
      "grad_norm": 2.5585254001148137,
      "language_loss": 0.74897248,
      "learning_rate": 3.974553507496378e-06,
      "loss": 0.77183276,
      "num_input_tokens_seen": 14025725,
      "step": 660,
      "time_per_iteration": 2.9521210193634033
    },
    {
      "auxiliary_loss_clip": 0.01242171,
      "auxiliary_loss_mlp": 0.01043331,
      "balance_loss_clip": 1.07160902,
      "balance_loss_mlp": 1.03118896,
      "epoch": 0.07948055071243913,
      "flos": 23733290764800.0,
      "grad_norm": 2.6383792739148033,
      "language_loss": 0.89223242,
      "learning_rate": 3.974429492392068e-06,
      "loss": 0.9150874,
      "num_input_tokens_seen": 14045750,
      "step": 661,
      "time_per_iteration": 2.777139902114868
    },
    {
      "auxiliary_loss_clip": 0.01232509,
      "auxiliary_loss_mlp": 0.01042314,
      "balance_loss_clip": 1.06894732,
      "balance_loss_mlp": 1.02988005,
      "epoch": 0.07960079360307822,
      "flos": 19573326996480.0,
      "grad_norm": 2.380956127432338,
      "language_loss": 0.90974975,
      "learning_rate": 3.974305177768373e-06,
      "loss": 0.93249798,
      "num_input_tokens_seen": 14063960,
      "step": 662,
      "time_per_iteration": 2.688359022140503
    },
    {
      "auxiliary_loss_clip": 0.0121887,
      "auxiliary_loss_mlp": 0.0104826,
      "balance_loss_clip": 1.06458712,
      "balance_loss_mlp": 1.03623223,
      "epoch": 0.07972103649371731,
      "flos": 23513409659520.0,
      "grad_norm": 2.4881055496953595,
      "language_loss": 0.86381161,
      "learning_rate": 3.974180563644152e-06,
      "loss": 0.88648295,
      "num_input_tokens_seen": 14082525,
      "step": 663,
      "time_per_iteration": 2.7511723041534424
    },
    {
      "auxiliary_loss_clip": 0.01242497,
      "auxiliary_loss_mlp": 0.01045159,
      "balance_loss_clip": 1.06681776,
      "balance_loss_mlp": 1.03345847,
      "epoch": 0.0798412793843564,
      "flos": 16726867770240.0,
      "grad_norm": 2.475336996238335,
      "language_loss": 0.89013839,
      "learning_rate": 3.97405565003831e-06,
      "loss": 0.91301495,
      "num_input_tokens_seen": 14098610,
      "step": 664,
      "time_per_iteration": 2.742305278778076
    },
    {
      "auxiliary_loss_clip": 0.01264979,
      "auxiliary_loss_mlp": 0.01051239,
      "balance_loss_clip": 1.0701412,
      "balance_loss_mlp": 1.03975892,
      "epoch": 0.07996152227499549,
      "flos": 18223337214720.0,
      "grad_norm": 7.639981606399918,
      "language_loss": 0.78188622,
      "learning_rate": 3.973930436969794e-06,
      "loss": 0.80504847,
      "num_input_tokens_seen": 14117065,
      "step": 665,
      "time_per_iteration": 2.679553985595703
    },
    {
      "auxiliary_loss_clip": 0.01226422,
      "auxiliary_loss_mlp": 0.0076528,
      "balance_loss_clip": 1.06533861,
      "balance_loss_mlp": 1.00047612,
      "epoch": 0.08008176516563459,
      "flos": 20594877793920.0,
      "grad_norm": 2.097676442224994,
      "language_loss": 0.86034334,
      "learning_rate": 3.973804924457602e-06,
      "loss": 0.88026035,
      "num_input_tokens_seen": 14135145,
      "step": 666,
      "time_per_iteration": 2.66218900680542
    },
    {
      "auxiliary_loss_clip": 0.01241627,
      "auxiliary_loss_mlp": 0.01049038,
      "balance_loss_clip": 1.06850123,
      "balance_loss_mlp": 1.0368247,
      "epoch": 0.08020200805627367,
      "flos": 31834306863360.0,
      "grad_norm": 2.347258358593638,
      "language_loss": 0.85555542,
      "learning_rate": 3.973679112520771e-06,
      "loss": 0.87846202,
      "num_input_tokens_seen": 14156860,
      "step": 667,
      "time_per_iteration": 4.122810363769531
    },
    {
      "auxiliary_loss_clip": 0.01230356,
      "auxiliary_loss_mlp": 0.01045993,
      "balance_loss_clip": 1.06184542,
      "balance_loss_mlp": 1.0340426,
      "epoch": 0.08032225094691277,
      "flos": 17783503176960.0,
      "grad_norm": 1.95086315975534,
      "language_loss": 0.99160695,
      "learning_rate": 3.973553001178389e-06,
      "loss": 1.01437056,
      "num_input_tokens_seen": 14174365,
      "step": 668,
      "time_per_iteration": 2.7246711254119873
    },
    {
      "auxiliary_loss_clip": 0.01267811,
      "auxiliary_loss_mlp": 0.0104538,
      "balance_loss_clip": 1.0719142,
      "balance_loss_mlp": 1.03323805,
      "epoch": 0.08044249383755185,
      "flos": 24061693835520.0,
      "grad_norm": 2.3534705429665097,
      "language_loss": 0.75515103,
      "learning_rate": 3.973426590449585e-06,
      "loss": 0.77828294,
      "num_input_tokens_seen": 14192320,
      "step": 669,
      "time_per_iteration": 2.6912992000579834
    },
    {
      "auxiliary_loss_clip": 0.0124453,
      "auxiliary_loss_mlp": 0.01049789,
      "balance_loss_clip": 1.06824398,
      "balance_loss_mlp": 1.03700423,
      "epoch": 0.08056273672819095,
      "flos": 18223624523520.0,
      "grad_norm": 2.161434855193005,
      "language_loss": 0.75297123,
      "learning_rate": 3.9732998803535364e-06,
      "loss": 0.77591443,
      "num_input_tokens_seen": 14210380,
      "step": 670,
      "time_per_iteration": 2.6965038776397705
    },
    {
      "auxiliary_loss_clip": 0.01186069,
      "auxiliary_loss_mlp": 0.01047989,
      "balance_loss_clip": 1.0537467,
      "balance_loss_mlp": 1.03577054,
      "epoch": 0.08068297961883003,
      "flos": 19676856971520.0,
      "grad_norm": 3.142205933179083,
      "language_loss": 0.85570449,
      "learning_rate": 3.973172870909465e-06,
      "loss": 0.87804508,
      "num_input_tokens_seen": 14225145,
      "step": 671,
      "time_per_iteration": 3.6533305644989014
    },
    {
      "auxiliary_loss_clip": 0.01198908,
      "auxiliary_loss_mlp": 0.01051346,
      "balance_loss_clip": 1.06098294,
      "balance_loss_mlp": 1.03856087,
      "epoch": 0.08080322250946913,
      "flos": 23148736830720.0,
      "grad_norm": 2.426954540679722,
      "language_loss": 0.80360776,
      "learning_rate": 3.973045562136638e-06,
      "loss": 0.8261103,
      "num_input_tokens_seen": 14241960,
      "step": 672,
      "time_per_iteration": 2.781479835510254
    },
    {
      "auxiliary_loss_clip": 0.01248918,
      "auxiliary_loss_mlp": 0.01046882,
      "balance_loss_clip": 1.07208157,
      "balance_loss_mlp": 1.03539622,
      "epoch": 0.08092346540010822,
      "flos": 21763626526080.0,
      "grad_norm": 3.2631904143831747,
      "language_loss": 0.91398287,
      "learning_rate": 3.972917954054368e-06,
      "loss": 0.93694085,
      "num_input_tokens_seen": 14260515,
      "step": 673,
      "time_per_iteration": 2.646489143371582
    },
    {
      "auxiliary_loss_clip": 0.01264255,
      "auxiliary_loss_mlp": 0.01042548,
      "balance_loss_clip": 1.07519674,
      "balance_loss_mlp": 1.03029287,
      "epoch": 0.08104370829074731,
      "flos": 21032485188480.0,
      "grad_norm": 2.470461745647926,
      "language_loss": 0.82223493,
      "learning_rate": 3.972790046682013e-06,
      "loss": 0.845303,
      "num_input_tokens_seen": 14279190,
      "step": 674,
      "time_per_iteration": 3.713736057281494
    },
    {
      "auxiliary_loss_clip": 0.01213698,
      "auxiliary_loss_mlp": 0.01047574,
      "balance_loss_clip": 1.05834198,
      "balance_loss_mlp": 1.03505754,
      "epoch": 0.0811639511813864,
      "flos": 20083186598400.0,
      "grad_norm": 2.1580151859139645,
      "language_loss": 0.79051387,
      "learning_rate": 3.972661840038977e-06,
      "loss": 0.81312662,
      "num_input_tokens_seen": 14299480,
      "step": 675,
      "time_per_iteration": 2.754770040512085
    },
    {
      "auxiliary_loss_clip": 0.01217327,
      "auxiliary_loss_mlp": 0.01053457,
      "balance_loss_clip": 1.06613564,
      "balance_loss_mlp": 1.04125583,
      "epoch": 0.08128419407202549,
      "flos": 16836718538880.0,
      "grad_norm": 2.440606031109393,
      "language_loss": 0.83474255,
      "learning_rate": 3.972533334144707e-06,
      "loss": 0.85745037,
      "num_input_tokens_seen": 14316405,
      "step": 676,
      "time_per_iteration": 2.654292345046997
    },
    {
      "auxiliary_loss_clip": 0.01253706,
      "auxiliary_loss_mlp": 0.01047429,
      "balance_loss_clip": 1.06820357,
      "balance_loss_mlp": 1.03416121,
      "epoch": 0.08140443696266458,
      "flos": 23769273214080.0,
      "grad_norm": 2.2483551435759384,
      "language_loss": 0.78757006,
      "learning_rate": 3.972404529018699e-06,
      "loss": 0.81058139,
      "num_input_tokens_seen": 14336265,
      "step": 677,
      "time_per_iteration": 2.656214714050293
    },
    {
      "auxiliary_loss_clip": 0.01205416,
      "auxiliary_loss_mlp": 0.01041608,
      "balance_loss_clip": 1.05760443,
      "balance_loss_mlp": 1.02946615,
      "epoch": 0.08152467985330367,
      "flos": 24390132819840.0,
      "grad_norm": 1.8600175900797367,
      "language_loss": 0.85350752,
      "learning_rate": 3.972275424680493e-06,
      "loss": 0.87597775,
      "num_input_tokens_seen": 14356375,
      "step": 678,
      "time_per_iteration": 2.7209877967834473
    },
    {
      "auxiliary_loss_clip": 0.01198686,
      "auxiliary_loss_mlp": 0.01042667,
      "balance_loss_clip": 1.06079972,
      "balance_loss_mlp": 1.03082919,
      "epoch": 0.08164492274394276,
      "flos": 19317750750720.0,
      "grad_norm": 2.2750923723078365,
      "language_loss": 0.9160254,
      "learning_rate": 3.972146021149673e-06,
      "loss": 0.93843901,
      "num_input_tokens_seen": 14374650,
      "step": 679,
      "time_per_iteration": 2.7246086597442627
    },
    {
      "auxiliary_loss_clip": 0.0121573,
      "auxiliary_loss_mlp": 0.01047386,
      "balance_loss_clip": 1.06287289,
      "balance_loss_mlp": 1.03623974,
      "epoch": 0.08176516563458186,
      "flos": 14830461319680.0,
      "grad_norm": 2.182526438262483,
      "language_loss": 0.78744769,
      "learning_rate": 3.972016318445868e-06,
      "loss": 0.81007886,
      "num_input_tokens_seen": 14392650,
      "step": 680,
      "time_per_iteration": 2.698295831680298
    },
    {
      "auxiliary_loss_clip": 0.01241588,
      "auxiliary_loss_mlp": 0.01041782,
      "balance_loss_clip": 1.06628489,
      "balance_loss_mlp": 1.02831757,
      "epoch": 0.08188540852522094,
      "flos": 22602320161920.0,
      "grad_norm": 2.1507748341823287,
      "language_loss": 0.92042553,
      "learning_rate": 3.971886316588757e-06,
      "loss": 0.94325924,
      "num_input_tokens_seen": 14413155,
      "step": 681,
      "time_per_iteration": 2.695688247680664
    },
    {
      "auxiliary_loss_clip": 0.01209818,
      "auxiliary_loss_mlp": 0.01046007,
      "balance_loss_clip": 1.0645175,
      "balance_loss_mlp": 1.03340042,
      "epoch": 0.08200565141586004,
      "flos": 19463727623040.0,
      "grad_norm": 2.525768757550836,
      "language_loss": 0.7336095,
      "learning_rate": 3.9717560155980595e-06,
      "loss": 0.75616777,
      "num_input_tokens_seen": 14428805,
      "step": 682,
      "time_per_iteration": 2.7242660522460938
    },
    {
      "auxiliary_loss_clip": 0.01264162,
      "auxiliary_loss_mlp": 0.01047355,
      "balance_loss_clip": 1.06932533,
      "balance_loss_mlp": 1.03515959,
      "epoch": 0.08212589430649912,
      "flos": 20594662312320.0,
      "grad_norm": 2.6641062001657874,
      "language_loss": 0.92070282,
      "learning_rate": 3.971625415493542e-06,
      "loss": 0.94381791,
      "num_input_tokens_seen": 14447125,
      "step": 683,
      "time_per_iteration": 2.6173059940338135
    },
    {
      "auxiliary_loss_clip": 0.01216478,
      "auxiliary_loss_mlp": 0.01045553,
      "balance_loss_clip": 1.06240058,
      "balance_loss_mlp": 1.03280973,
      "epoch": 0.08224613719713822,
      "flos": 25953611086080.0,
      "grad_norm": 1.8863575467518532,
      "language_loss": 0.87627685,
      "learning_rate": 3.971494516295017e-06,
      "loss": 0.89889717,
      "num_input_tokens_seen": 14466575,
      "step": 684,
      "time_per_iteration": 2.7631607055664062
    },
    {
      "auxiliary_loss_clip": 0.01231443,
      "auxiliary_loss_mlp": 0.01048579,
      "balance_loss_clip": 1.06858528,
      "balance_loss_mlp": 1.03655684,
      "epoch": 0.08236638008777732,
      "flos": 23768734510080.0,
      "grad_norm": 1.88756708783247,
      "language_loss": 0.85411727,
      "learning_rate": 3.971363318022341e-06,
      "loss": 0.87691748,
      "num_input_tokens_seen": 14487915,
      "step": 685,
      "time_per_iteration": 2.7501144409179688
    },
    {
      "auxiliary_loss_clip": 0.01231221,
      "auxiliary_loss_mlp": 0.01043256,
      "balance_loss_clip": 1.06502724,
      "balance_loss_mlp": 1.03141212,
      "epoch": 0.0824866229784164,
      "flos": 38799144887040.0,
      "grad_norm": 1.761159305762775,
      "language_loss": 0.68768191,
      "learning_rate": 3.971231820695417e-06,
      "loss": 0.71042669,
      "num_input_tokens_seen": 14511530,
      "step": 686,
      "time_per_iteration": 2.9122443199157715
    },
    {
      "auxiliary_loss_clip": 0.01252981,
      "auxiliary_loss_mlp": 0.0104586,
      "balance_loss_clip": 1.06786084,
      "balance_loss_mlp": 1.03392756,
      "epoch": 0.0826068658690555,
      "flos": 23107762391040.0,
      "grad_norm": 2.0542398667931803,
      "language_loss": 0.81396675,
      "learning_rate": 3.971100024334193e-06,
      "loss": 0.83695513,
      "num_input_tokens_seen": 14529050,
      "step": 687,
      "time_per_iteration": 2.722506046295166
    },
    {
      "auxiliary_loss_clip": 0.01244022,
      "auxiliary_loss_mlp": 0.01045888,
      "balance_loss_clip": 1.0663681,
      "balance_loss_mlp": 1.03418815,
      "epoch": 0.08272710875969458,
      "flos": 21136374299520.0,
      "grad_norm": 3.568744456721291,
      "language_loss": 0.86065859,
      "learning_rate": 3.970967928958663e-06,
      "loss": 0.88355774,
      "num_input_tokens_seen": 14546165,
      "step": 688,
      "time_per_iteration": 2.676602363586426
    },
    {
      "auxiliary_loss_clip": 0.01247996,
      "auxiliary_loss_mlp": 0.01048114,
      "balance_loss_clip": 1.07196701,
      "balance_loss_mlp": 1.0360682,
      "epoch": 0.08284735165033368,
      "flos": 19063000517760.0,
      "grad_norm": 2.2064983831922773,
      "language_loss": 0.83411163,
      "learning_rate": 3.970835534588865e-06,
      "loss": 0.85707271,
      "num_input_tokens_seen": 14563660,
      "step": 689,
      "time_per_iteration": 2.64440655708313
    },
    {
      "auxiliary_loss_clip": 0.01235675,
      "auxiliary_loss_mlp": 0.01048755,
      "balance_loss_clip": 1.06802762,
      "balance_loss_mlp": 1.03714991,
      "epoch": 0.08296759454097276,
      "flos": 16727442387840.0,
      "grad_norm": 1.898658770042709,
      "language_loss": 0.85789168,
      "learning_rate": 3.970702841244883e-06,
      "loss": 0.88073593,
      "num_input_tokens_seen": 14581980,
      "step": 690,
      "time_per_iteration": 2.6947004795074463
    },
    {
      "auxiliary_loss_clip": 0.01252877,
      "auxiliary_loss_mlp": 0.01045879,
      "balance_loss_clip": 1.07178259,
      "balance_loss_mlp": 1.03395796,
      "epoch": 0.08308783743161186,
      "flos": 18004928567040.0,
      "grad_norm": 2.8404753557348914,
      "language_loss": 0.82631266,
      "learning_rate": 3.970569848946847e-06,
      "loss": 0.84930021,
      "num_input_tokens_seen": 14601795,
      "step": 691,
      "time_per_iteration": 2.6343977451324463
    },
    {
      "auxiliary_loss_clip": 0.01148309,
      "auxiliary_loss_mlp": 0.00764748,
      "balance_loss_clip": 1.054968,
      "balance_loss_mlp": 1.00061011,
      "epoch": 0.08320808032225095,
      "flos": 15079788599040.0,
      "grad_norm": 2.3652926735661444,
      "language_loss": 0.8260864,
      "learning_rate": 3.970436557714932e-06,
      "loss": 0.84521699,
      "num_input_tokens_seen": 14618315,
      "step": 692,
      "time_per_iteration": 2.8094089031219482
    },
    {
      "auxiliary_loss_clip": 0.01267521,
      "auxiliary_loss_mlp": 0.01036535,
      "balance_loss_clip": 1.06976199,
      "balance_loss_mlp": 1.02382755,
      "epoch": 0.08332832321289003,
      "flos": 22383085501440.0,
      "grad_norm": 2.710530995215425,
      "language_loss": 0.86514622,
      "learning_rate": 3.970302967569358e-06,
      "loss": 0.88818681,
      "num_input_tokens_seen": 14636905,
      "step": 693,
      "time_per_iteration": 4.478849411010742
    },
    {
      "auxiliary_loss_clip": 0.01186974,
      "auxiliary_loss_mlp": 0.01042543,
      "balance_loss_clip": 1.06294894,
      "balance_loss_mlp": 1.03006136,
      "epoch": 0.08344856610352913,
      "flos": 24717386655360.0,
      "grad_norm": 2.642591927421962,
      "language_loss": 0.68201745,
      "learning_rate": 3.9701690785303896e-06,
      "loss": 0.70431262,
      "num_input_tokens_seen": 14656100,
      "step": 694,
      "time_per_iteration": 2.801844596862793
    },
    {
      "auxiliary_loss_clip": 0.01238583,
      "auxiliary_loss_mlp": 0.0105113,
      "balance_loss_clip": 1.06816244,
      "balance_loss_mlp": 1.03948331,
      "epoch": 0.08356880899416821,
      "flos": 25370206387200.0,
      "grad_norm": 3.3729047024133085,
      "language_loss": 0.88310385,
      "learning_rate": 3.970034890618339e-06,
      "loss": 0.90600097,
      "num_input_tokens_seen": 14675790,
      "step": 695,
      "time_per_iteration": 2.7061355113983154
    },
    {
      "auxiliary_loss_clip": 0.01231214,
      "auxiliary_loss_mlp": 0.01039628,
      "balance_loss_clip": 1.0671128,
      "balance_loss_mlp": 1.02820146,
      "epoch": 0.08368905188480731,
      "flos": 24353072962560.0,
      "grad_norm": 2.9414764158676934,
      "language_loss": 0.8818357,
      "learning_rate": 3.969900403853562e-06,
      "loss": 0.90454412,
      "num_input_tokens_seen": 14694830,
      "step": 696,
      "time_per_iteration": 2.6733052730560303
    },
    {
      "auxiliary_loss_clip": 0.01234442,
      "auxiliary_loss_mlp": 0.01048021,
      "balance_loss_clip": 1.07224727,
      "balance_loss_mlp": 1.03634465,
      "epoch": 0.08380929477544641,
      "flos": 18037319656320.0,
      "grad_norm": 2.7057675387883977,
      "language_loss": 0.78032041,
      "learning_rate": 3.96976561825646e-06,
      "loss": 0.80314505,
      "num_input_tokens_seen": 14711920,
      "step": 697,
      "time_per_iteration": 3.6131269931793213
    },
    {
      "auxiliary_loss_clip": 0.01232143,
      "auxiliary_loss_mlp": 0.01045259,
      "balance_loss_clip": 1.0708437,
      "balance_loss_mlp": 1.03346944,
      "epoch": 0.08392953766608549,
      "flos": 26286287875200.0,
      "grad_norm": 2.8179104465955236,
      "language_loss": 0.87282151,
      "learning_rate": 3.969630533847479e-06,
      "loss": 0.89559549,
      "num_input_tokens_seen": 14730880,
      "step": 698,
      "time_per_iteration": 2.7224044799804688
    },
    {
      "auxiliary_loss_clip": 0.0120676,
      "auxiliary_loss_mlp": 0.01041001,
      "balance_loss_clip": 1.06260478,
      "balance_loss_mlp": 1.02964652,
      "epoch": 0.08404978055672459,
      "flos": 22492146170880.0,
      "grad_norm": 2.039741311336977,
      "language_loss": 0.8400808,
      "learning_rate": 3.969495150647113e-06,
      "loss": 0.86255848,
      "num_input_tokens_seen": 14749050,
      "step": 699,
      "time_per_iteration": 2.726409673690796
    },
    {
      "auxiliary_loss_clip": 0.01236239,
      "auxiliary_loss_mlp": 0.01045671,
      "balance_loss_clip": 1.06920242,
      "balance_loss_mlp": 1.03394628,
      "epoch": 0.08417002344736367,
      "flos": 24826878288000.0,
      "grad_norm": 1.8059604245993996,
      "language_loss": 0.76638073,
      "learning_rate": 3.969359468675899e-06,
      "loss": 0.78919983,
      "num_input_tokens_seen": 14769180,
      "step": 700,
      "time_per_iteration": 3.7343268394470215
    },
    {
      "auxiliary_loss_clip": 0.01214102,
      "auxiliary_loss_mlp": 0.01038626,
      "balance_loss_clip": 1.05808473,
      "balance_loss_mlp": 1.02618027,
      "epoch": 0.08429026633800277,
      "flos": 16945922862720.0,
      "grad_norm": 1.8798002861368084,
      "language_loss": 0.88798165,
      "learning_rate": 3.969223487954418e-06,
      "loss": 0.91050899,
      "num_input_tokens_seen": 14786640,
      "step": 701,
      "time_per_iteration": 2.7208199501037598
    },
    {
      "auxiliary_loss_clip": 0.01178241,
      "auxiliary_loss_mlp": 0.00764876,
      "balance_loss_clip": 1.05592155,
      "balance_loss_mlp": 1.00063396,
      "epoch": 0.08441050922864185,
      "flos": 23841920471040.0,
      "grad_norm": 2.3173308737042158,
      "language_loss": 0.82572961,
      "learning_rate": 3.969087208503301e-06,
      "loss": 0.84516078,
      "num_input_tokens_seen": 14806720,
      "step": 702,
      "time_per_iteration": 2.857052803039551
    },
    {
      "auxiliary_loss_clip": 0.01238871,
      "auxiliary_loss_mlp": 0.01038004,
      "balance_loss_clip": 1.07133698,
      "balance_loss_mlp": 1.02586794,
      "epoch": 0.08453075211928095,
      "flos": 25520205582720.0,
      "grad_norm": 3.189415470762739,
      "language_loss": 0.84500206,
      "learning_rate": 3.968950630343219e-06,
      "loss": 0.86777079,
      "num_input_tokens_seen": 14823705,
      "step": 703,
      "time_per_iteration": 2.7181830406188965
    },
    {
      "auxiliary_loss_clip": 0.01232579,
      "auxiliary_loss_mlp": 0.01037156,
      "balance_loss_clip": 1.06407583,
      "balance_loss_mlp": 1.02511644,
      "epoch": 0.08465099500992004,
      "flos": 19532496211200.0,
      "grad_norm": 2.5129263124859103,
      "language_loss": 0.9342283,
      "learning_rate": 3.968813753494892e-06,
      "loss": 0.95692563,
      "num_input_tokens_seen": 14841865,
      "step": 704,
      "time_per_iteration": 2.626840114593506
    },
    {
      "auxiliary_loss_clip": 0.01183531,
      "auxiliary_loss_mlp": 0.00765154,
      "balance_loss_clip": 1.05903435,
      "balance_loss_mlp": 1.00055218,
      "epoch": 0.08477123790055913,
      "flos": 29351299403520.0,
      "grad_norm": 3.7910765189673694,
      "language_loss": 0.75391722,
      "learning_rate": 3.968676577979084e-06,
      "loss": 0.77340412,
      "num_input_tokens_seen": 14861415,
      "step": 705,
      "time_per_iteration": 2.8336901664733887
    },
    {
      "auxiliary_loss_clip": 0.01187131,
      "auxiliary_loss_mlp": 0.01047629,
      "balance_loss_clip": 1.06039,
      "balance_loss_mlp": 1.03508234,
      "epoch": 0.08489148079119822,
      "flos": 18624495283200.0,
      "grad_norm": 2.1592333305706886,
      "language_loss": 0.77932382,
      "learning_rate": 3.968539103816605e-06,
      "loss": 0.80167139,
      "num_input_tokens_seen": 14879215,
      "step": 706,
      "time_per_iteration": 2.7318620681762695
    },
    {
      "auxiliary_loss_clip": 0.012496,
      "auxiliary_loss_mlp": 0.007645,
      "balance_loss_clip": 1.07176948,
      "balance_loss_mlp": 1.00067806,
      "epoch": 0.0850117236818373,
      "flos": 23471393725440.0,
      "grad_norm": 2.263737771735796,
      "language_loss": 0.89459217,
      "learning_rate": 3.9684013310283085e-06,
      "loss": 0.91473317,
      "num_input_tokens_seen": 14897900,
      "step": 707,
      "time_per_iteration": 2.751319646835327
    },
    {
      "auxiliary_loss_clip": 0.01201685,
      "auxiliary_loss_mlp": 0.01045135,
      "balance_loss_clip": 1.06541216,
      "balance_loss_mlp": 1.03264201,
      "epoch": 0.0851319665724764,
      "flos": 40625058896640.0,
      "grad_norm": 2.2149022075689766,
      "language_loss": 0.64212894,
      "learning_rate": 3.9682632596350956e-06,
      "loss": 0.66459715,
      "num_input_tokens_seen": 14919065,
      "step": 708,
      "time_per_iteration": 2.8771755695343018
    },
    {
      "auxiliary_loss_clip": 0.01224742,
      "auxiliary_loss_mlp": 0.01044684,
      "balance_loss_clip": 1.06172228,
      "balance_loss_mlp": 1.03337741,
      "epoch": 0.0852522094631155,
      "flos": 15879554870400.0,
      "grad_norm": 2.0985334720863915,
      "language_loss": 0.78575742,
      "learning_rate": 3.968124889657911e-06,
      "loss": 0.80845165,
      "num_input_tokens_seen": 14934165,
      "step": 709,
      "time_per_iteration": 2.6554219722747803
    },
    {
      "auxiliary_loss_clip": 0.01196478,
      "auxiliary_loss_mlp": 0.01042985,
      "balance_loss_clip": 1.06222844,
      "balance_loss_mlp": 1.03054583,
      "epoch": 0.08537245235375458,
      "flos": 14567091822720.0,
      "grad_norm": 2.5062092953112676,
      "language_loss": 0.90665293,
      "learning_rate": 3.967986221117746e-06,
      "loss": 0.92904752,
      "num_input_tokens_seen": 14950105,
      "step": 710,
      "time_per_iteration": 2.7636613845825195
    },
    {
      "auxiliary_loss_clip": 0.01247455,
      "auxiliary_loss_mlp": 0.01039978,
      "balance_loss_clip": 1.0670476,
      "balance_loss_mlp": 1.02813506,
      "epoch": 0.08549269524439368,
      "flos": 26468929555200.0,
      "grad_norm": 2.0729407614634447,
      "language_loss": 0.86309278,
      "learning_rate": 3.967847254035635e-06,
      "loss": 0.88596702,
      "num_input_tokens_seen": 14969490,
      "step": 711,
      "time_per_iteration": 2.7108051776885986
    },
    {
      "auxiliary_loss_clip": 0.01195835,
      "auxiliary_loss_mlp": 0.01043966,
      "balance_loss_clip": 1.05964231,
      "balance_loss_mlp": 1.0316335,
      "epoch": 0.08561293813503276,
      "flos": 13590214565760.0,
      "grad_norm": 3.530129166242322,
      "language_loss": 0.86260515,
      "learning_rate": 3.967707988432661e-06,
      "loss": 0.88500321,
      "num_input_tokens_seen": 14987195,
      "step": 712,
      "time_per_iteration": 2.7264649868011475
    },
    {
      "auxiliary_loss_clip": 0.01200601,
      "auxiliary_loss_mlp": 0.01042249,
      "balance_loss_clip": 1.05823898,
      "balance_loss_mlp": 1.03020287,
      "epoch": 0.08573318102567186,
      "flos": 26943524979840.0,
      "grad_norm": 2.5856641116695127,
      "language_loss": 0.87603682,
      "learning_rate": 3.967568424329949e-06,
      "loss": 0.89846528,
      "num_input_tokens_seen": 15007620,
      "step": 713,
      "time_per_iteration": 2.793138265609741
    },
    {
      "auxiliary_loss_clip": 0.01105437,
      "auxiliary_loss_mlp": 0.01006869,
      "balance_loss_clip": 1.03985548,
      "balance_loss_mlp": 1.00338793,
      "epoch": 0.08585342391631094,
      "flos": 67302739319040.0,
      "grad_norm": 0.8455214695874186,
      "language_loss": 0.55497974,
      "learning_rate": 3.967428561748671e-06,
      "loss": 0.57610285,
      "num_input_tokens_seen": 15075590,
      "step": 714,
      "time_per_iteration": 3.608048677444458
    },
    {
      "auxiliary_loss_clip": 0.01251215,
      "auxiliary_loss_mlp": 0.01043775,
      "balance_loss_clip": 1.06624317,
      "balance_loss_mlp": 1.03147233,
      "epoch": 0.08597366680695004,
      "flos": 22456594684800.0,
      "grad_norm": 2.144727581768476,
      "language_loss": 0.87505186,
      "learning_rate": 3.967288400710045e-06,
      "loss": 0.89800179,
      "num_input_tokens_seen": 15095055,
      "step": 715,
      "time_per_iteration": 2.9701473712921143
    },
    {
      "auxiliary_loss_clip": 0.01181723,
      "auxiliary_loss_mlp": 0.01051649,
      "balance_loss_clip": 1.05945337,
      "balance_loss_mlp": 1.04056811,
      "epoch": 0.08609390969758914,
      "flos": 23550505430400.0,
      "grad_norm": 1.965278686742079,
      "language_loss": 0.88580489,
      "learning_rate": 3.9671479412353335e-06,
      "loss": 0.90813857,
      "num_input_tokens_seen": 15113520,
      "step": 716,
      "time_per_iteration": 3.2133312225341797
    },
    {
      "auxiliary_loss_clip": 0.01261524,
      "auxiliary_loss_mlp": 0.01044077,
      "balance_loss_clip": 1.06968451,
      "balance_loss_mlp": 1.03117287,
      "epoch": 0.08621415258822822,
      "flos": 25885848078720.0,
      "grad_norm": 2.2346447076990583,
      "language_loss": 0.73921961,
      "learning_rate": 3.967007183345843e-06,
      "loss": 0.76227564,
      "num_input_tokens_seen": 15133375,
      "step": 717,
      "time_per_iteration": 2.71862530708313
    },
    {
      "auxiliary_loss_clip": 0.01182008,
      "auxiliary_loss_mlp": 0.01044121,
      "balance_loss_clip": 1.06043267,
      "balance_loss_mlp": 1.0322175,
      "epoch": 0.08633439547886732,
      "flos": 13589568120960.0,
      "grad_norm": 2.825644460345985,
      "language_loss": 0.89199787,
      "learning_rate": 3.966866127062927e-06,
      "loss": 0.9142592,
      "num_input_tokens_seen": 15150500,
      "step": 718,
      "time_per_iteration": 2.7264485359191895
    },
    {
      "auxiliary_loss_clip": 0.01126944,
      "auxiliary_loss_mlp": 0.01005423,
      "balance_loss_clip": 1.04105854,
      "balance_loss_mlp": 1.00189483,
      "epoch": 0.0864546383695064,
      "flos": 57767342434560.0,
      "grad_norm": 0.8840166423488564,
      "language_loss": 0.62709439,
      "learning_rate": 3.966724772407982e-06,
      "loss": 0.64841807,
      "num_input_tokens_seen": 15208015,
      "step": 719,
      "time_per_iteration": 5.081363677978516
    },
    {
      "auxiliary_loss_clip": 0.01224135,
      "auxiliary_loss_mlp": 0.0105226,
      "balance_loss_clip": 1.06482005,
      "balance_loss_mlp": 1.04063678,
      "epoch": 0.0865748812601455,
      "flos": 20046952753920.0,
      "grad_norm": 2.304138526332285,
      "language_loss": 0.89000297,
      "learning_rate": 3.966583119402454e-06,
      "loss": 0.91276693,
      "num_input_tokens_seen": 15224780,
      "step": 720,
      "time_per_iteration": 2.7490594387054443
    },
    {
      "auxiliary_loss_clip": 0.01193768,
      "auxiliary_loss_mlp": 0.00765123,
      "balance_loss_clip": 1.06088519,
      "balance_loss_mlp": 1.00076222,
      "epoch": 0.08669512415078459,
      "flos": 35262446935680.0,
      "grad_norm": 1.6981794649940107,
      "language_loss": 0.82001364,
      "learning_rate": 3.9664411680678305e-06,
      "loss": 0.83960259,
      "num_input_tokens_seen": 15246535,
      "step": 721,
      "time_per_iteration": 2.9023845195770264
    },
    {
      "auxiliary_loss_clip": 0.01166533,
      "auxiliary_loss_mlp": 0.01003986,
      "balance_loss_clip": 1.0493207,
      "balance_loss_mlp": 1.00048125,
      "epoch": 0.08681536704142367,
      "flos": 65654870048640.0,
      "grad_norm": 0.8485582029594358,
      "language_loss": 0.61446196,
      "learning_rate": 3.966298918425644e-06,
      "loss": 0.63616711,
      "num_input_tokens_seen": 15304025,
      "step": 722,
      "time_per_iteration": 3.0913403034210205
    },
    {
      "auxiliary_loss_clip": 0.01199584,
      "auxiliary_loss_mlp": 0.01045259,
      "balance_loss_clip": 1.06366897,
      "balance_loss_mlp": 1.03281403,
      "epoch": 0.08693560993206277,
      "flos": 34529940881280.0,
      "grad_norm": 2.109510759518133,
      "language_loss": 0.82734478,
      "learning_rate": 3.966156370497476e-06,
      "loss": 0.8497932,
      "num_input_tokens_seen": 15327635,
      "step": 723,
      "time_per_iteration": 3.811575174331665
    },
    {
      "auxiliary_loss_clip": 0.01203938,
      "auxiliary_loss_mlp": 0.01041289,
      "balance_loss_clip": 1.0622859,
      "balance_loss_mlp": 1.02937984,
      "epoch": 0.08705585282270185,
      "flos": 23149419189120.0,
      "grad_norm": 2.1740645794341233,
      "language_loss": 0.8891257,
      "learning_rate": 3.96601352430495e-06,
      "loss": 0.91157794,
      "num_input_tokens_seen": 15347405,
      "step": 724,
      "time_per_iteration": 2.718278169631958
    },
    {
      "auxiliary_loss_clip": 0.01243198,
      "auxiliary_loss_mlp": 0.0104691,
      "balance_loss_clip": 1.06997895,
      "balance_loss_mlp": 1.03509688,
      "epoch": 0.08717609571334095,
      "flos": 29497599498240.0,
      "grad_norm": 1.6956604299831302,
      "language_loss": 0.82921636,
      "learning_rate": 3.965870379869735e-06,
      "loss": 0.85211736,
      "num_input_tokens_seen": 15369450,
      "step": 725,
      "time_per_iteration": 3.7477147579193115
    },
    {
      "auxiliary_loss_clip": 0.01239163,
      "auxiliary_loss_mlp": 0.01048652,
      "balance_loss_clip": 1.0649631,
      "balance_loss_mlp": 1.03742862,
      "epoch": 0.08729633860398003,
      "flos": 20667489137280.0,
      "grad_norm": 2.735364830501616,
      "language_loss": 0.8701787,
      "learning_rate": 3.965726937213547e-06,
      "loss": 0.89305687,
      "num_input_tokens_seen": 15388085,
      "step": 726,
      "time_per_iteration": 2.6924595832824707
    },
    {
      "auxiliary_loss_clip": 0.01246228,
      "auxiliary_loss_mlp": 0.01041043,
      "balance_loss_clip": 1.06766248,
      "balance_loss_mlp": 1.02915788,
      "epoch": 0.08741658149461913,
      "flos": 18369493655040.0,
      "grad_norm": 2.2650615832141208,
      "language_loss": 0.81357771,
      "learning_rate": 3.965583196358144e-06,
      "loss": 0.8364504,
      "num_input_tokens_seen": 15407120,
      "step": 727,
      "time_per_iteration": 2.663308620452881
    },
    {
      "auxiliary_loss_clip": 0.01236162,
      "auxiliary_loss_mlp": 0.01044894,
      "balance_loss_clip": 1.06791198,
      "balance_loss_mlp": 1.03196597,
      "epoch": 0.08753682438525823,
      "flos": 18729677283840.0,
      "grad_norm": 3.5194004353310526,
      "language_loss": 0.74449861,
      "learning_rate": 3.965439157325335e-06,
      "loss": 0.76730919,
      "num_input_tokens_seen": 15424485,
      "step": 728,
      "time_per_iteration": 2.7452521324157715
    },
    {
      "auxiliary_loss_clip": 0.01231983,
      "auxiliary_loss_mlp": 0.01044667,
      "balance_loss_clip": 1.06306577,
      "balance_loss_mlp": 1.0325973,
      "epoch": 0.08765706727589731,
      "flos": 27776113303680.0,
      "grad_norm": 3.1092236787649603,
      "language_loss": 0.75388861,
      "learning_rate": 3.965294820136968e-06,
      "loss": 0.77665508,
      "num_input_tokens_seen": 15446285,
      "step": 729,
      "time_per_iteration": 2.909882068634033
    },
    {
      "auxiliary_loss_clip": 0.01213317,
      "auxiliary_loss_mlp": 0.0104747,
      "balance_loss_clip": 1.06588674,
      "balance_loss_mlp": 1.03548408,
      "epoch": 0.08777731016653641,
      "flos": 24389127239040.0,
      "grad_norm": 2.1870024630783815,
      "language_loss": 0.87179518,
      "learning_rate": 3.965150184814938e-06,
      "loss": 0.89440304,
      "num_input_tokens_seen": 15465770,
      "step": 730,
      "time_per_iteration": 2.7604048252105713
    },
    {
      "auxiliary_loss_clip": 0.01254751,
      "auxiliary_loss_mlp": 0.01046624,
      "balance_loss_clip": 1.07321072,
      "balance_loss_mlp": 1.03435695,
      "epoch": 0.08789755305717549,
      "flos": 21981855605760.0,
      "grad_norm": 2.4188834264594994,
      "language_loss": 0.76511037,
      "learning_rate": 3.965005251381189e-06,
      "loss": 0.78812408,
      "num_input_tokens_seen": 15483705,
      "step": 731,
      "time_per_iteration": 2.624833822250366
    },
    {
      "auxiliary_loss_clip": 0.01125048,
      "auxiliary_loss_mlp": 0.01008968,
      "balance_loss_clip": 1.03749394,
      "balance_loss_mlp": 1.0051769,
      "epoch": 0.08801779594781459,
      "flos": 58360120583040.0,
      "grad_norm": 0.8930467562666728,
      "language_loss": 0.64548004,
      "learning_rate": 3.964860019857705e-06,
      "loss": 0.66682029,
      "num_input_tokens_seen": 15548620,
      "step": 732,
      "time_per_iteration": 3.2876298427581787
    },
    {
      "auxiliary_loss_clip": 0.01225802,
      "auxiliary_loss_mlp": 0.01050449,
      "balance_loss_clip": 1.06533575,
      "balance_loss_mlp": 1.03895688,
      "epoch": 0.08813803883845367,
      "flos": 23294785530240.0,
      "grad_norm": 1.7610471285972713,
      "language_loss": 0.84077227,
      "learning_rate": 3.964714490266518e-06,
      "loss": 0.86353481,
      "num_input_tokens_seen": 15569265,
      "step": 733,
      "time_per_iteration": 2.6083693504333496
    },
    {
      "auxiliary_loss_clip": 0.01172543,
      "auxiliary_loss_mlp": 0.01006071,
      "balance_loss_clip": 1.04523623,
      "balance_loss_mlp": 1.00235128,
      "epoch": 0.08825828172909277,
      "flos": 63424924882560.0,
      "grad_norm": 1.3833702459708264,
      "language_loss": 0.64589143,
      "learning_rate": 3.964568662629706e-06,
      "loss": 0.66767764,
      "num_input_tokens_seen": 15630570,
      "step": 734,
      "time_per_iteration": 3.0399959087371826
    },
    {
      "auxiliary_loss_clip": 0.01199663,
      "auxiliary_loss_mlp": 0.01040408,
      "balance_loss_clip": 1.06345391,
      "balance_loss_mlp": 1.02780128,
      "epoch": 0.08837852461973186,
      "flos": 26720986268160.0,
      "grad_norm": 5.198937467638228,
      "language_loss": 0.84419781,
      "learning_rate": 3.9644225369693895e-06,
      "loss": 0.86659849,
      "num_input_tokens_seen": 15650870,
      "step": 735,
      "time_per_iteration": 2.6632845401763916
    },
    {
      "auxiliary_loss_clip": 0.01229356,
      "auxiliary_loss_mlp": 0.01043352,
      "balance_loss_clip": 1.06716812,
      "balance_loss_mlp": 1.03115118,
      "epoch": 0.08849876751037095,
      "flos": 27265427688960.0,
      "grad_norm": 2.2734751094640595,
      "language_loss": 0.8689456,
      "learning_rate": 3.964276113307735e-06,
      "loss": 0.89167261,
      "num_input_tokens_seen": 15670835,
      "step": 736,
      "time_per_iteration": 2.741910219192505
    },
    {
      "auxiliary_loss_clip": 0.01244666,
      "auxiliary_loss_mlp": 0.01047993,
      "balance_loss_clip": 1.0689311,
      "balance_loss_mlp": 1.03546369,
      "epoch": 0.08861901040101004,
      "flos": 19828759587840.0,
      "grad_norm": 2.341978435101914,
      "language_loss": 0.80522442,
      "learning_rate": 3.9641293916669574e-06,
      "loss": 0.82815099,
      "num_input_tokens_seen": 15689795,
      "step": 737,
      "time_per_iteration": 2.6336917877197266
    },
    {
      "auxiliary_loss_clip": 0.01223374,
      "auxiliary_loss_mlp": 0.01049315,
      "balance_loss_clip": 1.06792164,
      "balance_loss_mlp": 1.0360589,
      "epoch": 0.08873925329164913,
      "flos": 23658704173440.0,
      "grad_norm": 2.4016583141281878,
      "language_loss": 0.83089554,
      "learning_rate": 3.9639823720693115e-06,
      "loss": 0.85362244,
      "num_input_tokens_seen": 15711650,
      "step": 738,
      "time_per_iteration": 2.6980602741241455
    },
    {
      "auxiliary_loss_clip": 0.01103976,
      "auxiliary_loss_mlp": 0.01012525,
      "balance_loss_clip": 1.03965425,
      "balance_loss_mlp": 1.00944984,
      "epoch": 0.08885949618228822,
      "flos": 71831541893760.0,
      "grad_norm": 0.8323119202617051,
      "language_loss": 0.60012054,
      "learning_rate": 3.963835054537102e-06,
      "loss": 0.62128556,
      "num_input_tokens_seen": 15780615,
      "step": 739,
      "time_per_iteration": 3.4278883934020996
    },
    {
      "auxiliary_loss_clip": 0.01228734,
      "auxiliary_loss_mlp": 0.01042002,
      "balance_loss_clip": 1.06510484,
      "balance_loss_mlp": 1.02951515,
      "epoch": 0.08897973907292732,
      "flos": 22346169298560.0,
      "grad_norm": 2.242931741232289,
      "language_loss": 0.60895097,
      "learning_rate": 3.963687439092676e-06,
      "loss": 0.63165832,
      "num_input_tokens_seen": 15801300,
      "step": 740,
      "time_per_iteration": 3.0269393920898438
    },
    {
      "auxiliary_loss_clip": 0.01177806,
      "auxiliary_loss_mlp": 0.01041426,
      "balance_loss_clip": 1.05715656,
      "balance_loss_mlp": 1.02930844,
      "epoch": 0.0890999819635664,
      "flos": 21251827589760.0,
      "grad_norm": 4.007119571122445,
      "language_loss": 0.8010143,
      "learning_rate": 3.963539525758427e-06,
      "loss": 0.8232066,
      "num_input_tokens_seen": 15820860,
      "step": 741,
      "time_per_iteration": 2.7772903442382812
    },
    {
      "auxiliary_loss_clip": 0.01197704,
      "auxiliary_loss_mlp": 0.01040724,
      "balance_loss_clip": 1.06526303,
      "balance_loss_mlp": 1.02770042,
      "epoch": 0.0892202248542055,
      "flos": 25370888745600.0,
      "grad_norm": 1.929654767836682,
      "language_loss": 0.67589986,
      "learning_rate": 3.9633913145567925e-06,
      "loss": 0.69828415,
      "num_input_tokens_seen": 15841350,
      "step": 742,
      "time_per_iteration": 2.8339710235595703
    },
    {
      "auxiliary_loss_clip": 0.01232695,
      "auxiliary_loss_mlp": 0.01038657,
      "balance_loss_clip": 1.06808221,
      "balance_loss_mlp": 1.02740979,
      "epoch": 0.08934046774484458,
      "flos": 24457895827200.0,
      "grad_norm": 2.0293083234038387,
      "language_loss": 0.81409675,
      "learning_rate": 3.9632428055102575e-06,
      "loss": 0.83681029,
      "num_input_tokens_seen": 15861360,
      "step": 743,
      "time_per_iteration": 2.7395942211151123
    },
    {
      "auxiliary_loss_clip": 0.01268259,
      "auxiliary_loss_mlp": 0.01043697,
      "balance_loss_clip": 1.07448769,
      "balance_loss_mlp": 1.03172803,
      "epoch": 0.08946071063548368,
      "flos": 35772773414400.0,
      "grad_norm": 2.5320127187766204,
      "language_loss": 0.66962594,
      "learning_rate": 3.9630939986413495e-06,
      "loss": 0.69274545,
      "num_input_tokens_seen": 15883160,
      "step": 744,
      "time_per_iteration": 2.724064826965332
    },
    {
      "auxiliary_loss_clip": 0.01229303,
      "auxiliary_loss_mlp": 0.01048464,
      "balance_loss_clip": 1.06673384,
      "balance_loss_mlp": 1.03682292,
      "epoch": 0.08958095352612276,
      "flos": 14356584167040.0,
      "grad_norm": 1.8770557140797437,
      "language_loss": 0.77875578,
      "learning_rate": 3.962944893972643e-06,
      "loss": 0.80153346,
      "num_input_tokens_seen": 15901610,
      "step": 745,
      "time_per_iteration": 3.835167646408081
    },
    {
      "auxiliary_loss_clip": 0.01171853,
      "auxiliary_loss_mlp": 0.01040983,
      "balance_loss_clip": 1.05747318,
      "balance_loss_mlp": 1.02902675,
      "epoch": 0.08970119641676186,
      "flos": 17853277345920.0,
      "grad_norm": 2.7675935670243423,
      "language_loss": 0.91079009,
      "learning_rate": 3.962795491526756e-06,
      "loss": 0.93291843,
      "num_input_tokens_seen": 15918770,
      "step": 746,
      "time_per_iteration": 3.723522663116455
    },
    {
      "auxiliary_loss_clip": 0.01231221,
      "auxiliary_loss_mlp": 0.01053688,
      "balance_loss_clip": 1.06976914,
      "balance_loss_mlp": 1.04158247,
      "epoch": 0.08982143930740095,
      "flos": 20811670329600.0,
      "grad_norm": 2.2166998151027966,
      "language_loss": 0.89419347,
      "learning_rate": 3.962645791326354e-06,
      "loss": 0.91704255,
      "num_input_tokens_seen": 15938025,
      "step": 747,
      "time_per_iteration": 2.6909613609313965
    },
    {
      "auxiliary_loss_clip": 0.01257721,
      "auxiliary_loss_mlp": 0.01046358,
      "balance_loss_clip": 1.07061934,
      "balance_loss_mlp": 1.03478873,
      "epoch": 0.08994168219804004,
      "flos": 24097712198400.0,
      "grad_norm": 2.3410163509893835,
      "language_loss": 0.83014798,
      "learning_rate": 3.962495793394146e-06,
      "loss": 0.85318875,
      "num_input_tokens_seen": 15957215,
      "step": 748,
      "time_per_iteration": 2.6479406356811523
    },
    {
      "auxiliary_loss_clip": 0.01140505,
      "auxiliary_loss_mlp": 0.01007794,
      "balance_loss_clip": 1.04306591,
      "balance_loss_mlp": 1.00443256,
      "epoch": 0.09006192508867913,
      "flos": 57188893812480.0,
      "grad_norm": 0.7475160422045899,
      "language_loss": 0.61229533,
      "learning_rate": 3.9623454977528864e-06,
      "loss": 0.63377839,
      "num_input_tokens_seen": 16015870,
      "step": 749,
      "time_per_iteration": 4.044859409332275
    },
    {
      "auxiliary_loss_clip": 0.0122693,
      "auxiliary_loss_mlp": 0.0103775,
      "balance_loss_clip": 1.06504011,
      "balance_loss_mlp": 1.0255965,
      "epoch": 0.09018216797931822,
      "flos": 20487505063680.0,
      "grad_norm": 1.7848719698516478,
      "language_loss": 0.85268617,
      "learning_rate": 3.962194904425375e-06,
      "loss": 0.87533295,
      "num_input_tokens_seen": 16036500,
      "step": 750,
      "time_per_iteration": 2.9428136348724365
    },
    {
      "auxiliary_loss_clip": 0.01195383,
      "auxiliary_loss_mlp": 0.01054011,
      "balance_loss_clip": 1.06404829,
      "balance_loss_mlp": 1.04182231,
      "epoch": 0.09030241086995731,
      "flos": 22638123043200.0,
      "grad_norm": 2.462874611237399,
      "language_loss": 0.67820901,
      "learning_rate": 3.9620440134344566e-06,
      "loss": 0.70070291,
      "num_input_tokens_seen": 16054655,
      "step": 751,
      "time_per_iteration": 3.842817544937134
    },
    {
      "auxiliary_loss_clip": 0.0125147,
      "auxiliary_loss_mlp": 0.01042624,
      "balance_loss_clip": 1.06892276,
      "balance_loss_mlp": 1.03070855,
      "epoch": 0.09042265376059641,
      "flos": 21871502046720.0,
      "grad_norm": 3.2477499393271665,
      "language_loss": 0.82635999,
      "learning_rate": 3.9618928248030215e-06,
      "loss": 0.84930086,
      "num_input_tokens_seen": 16074165,
      "step": 752,
      "time_per_iteration": 2.7667884826660156
    },
    {
      "auxiliary_loss_clip": 0.01231843,
      "auxiliary_loss_mlp": 0.01044341,
      "balance_loss_clip": 1.06710887,
      "balance_loss_mlp": 1.0321337,
      "epoch": 0.0905428966512355,
      "flos": 24316192673280.0,
      "grad_norm": 2.76542093855688,
      "language_loss": 0.82775009,
      "learning_rate": 3.961741338554005e-06,
      "loss": 0.85051197,
      "num_input_tokens_seen": 16092505,
      "step": 753,
      "time_per_iteration": 2.7124264240264893
    },
    {
      "auxiliary_loss_clip": 0.01237899,
      "auxiliary_loss_mlp": 0.01039241,
      "balance_loss_clip": 1.07048512,
      "balance_loss_mlp": 1.02654529,
      "epoch": 0.09066313954187459,
      "flos": 35845061535360.0,
      "grad_norm": 2.4557218736496442,
      "language_loss": 0.75384927,
      "learning_rate": 3.9615895547103865e-06,
      "loss": 0.77662063,
      "num_input_tokens_seen": 16116150,
      "step": 754,
      "time_per_iteration": 2.8565475940704346
    },
    {
      "auxiliary_loss_clip": 0.01188119,
      "auxiliary_loss_mlp": 0.01049539,
      "balance_loss_clip": 1.06203103,
      "balance_loss_mlp": 1.03691459,
      "epoch": 0.09078338243251367,
      "flos": 29168729550720.0,
      "grad_norm": 1.8932351571269108,
      "language_loss": 0.77664167,
      "learning_rate": 3.961437473295193e-06,
      "loss": 0.79901826,
      "num_input_tokens_seen": 16136295,
      "step": 755,
      "time_per_iteration": 2.843766212463379
    },
    {
      "auxiliary_loss_clip": 0.01208244,
      "auxiliary_loss_mlp": 0.01039727,
      "balance_loss_clip": 1.06344533,
      "balance_loss_mlp": 1.02923703,
      "epoch": 0.09090362532315277,
      "flos": 21907699977600.0,
      "grad_norm": 2.630951569424829,
      "language_loss": 0.72269344,
      "learning_rate": 3.961285094331495e-06,
      "loss": 0.74517316,
      "num_input_tokens_seen": 16154210,
      "step": 756,
      "time_per_iteration": 2.815603017807007
    },
    {
      "auxiliary_loss_clip": 0.01207544,
      "auxiliary_loss_mlp": 0.0104629,
      "balance_loss_clip": 1.06227946,
      "balance_loss_mlp": 1.03458977,
      "epoch": 0.09102386821379185,
      "flos": 27344503480320.0,
      "grad_norm": 2.2351552100611456,
      "language_loss": 0.85645479,
      "learning_rate": 3.961132417842406e-06,
      "loss": 0.87899315,
      "num_input_tokens_seen": 16173995,
      "step": 757,
      "time_per_iteration": 2.7559654712677
    },
    {
      "auxiliary_loss_clip": 0.01192333,
      "auxiliary_loss_mlp": 0.01041768,
      "balance_loss_clip": 1.06593633,
      "balance_loss_mlp": 1.02956092,
      "epoch": 0.09114411110443095,
      "flos": 20813501923200.0,
      "grad_norm": 2.578062810486569,
      "language_loss": 0.75526977,
      "learning_rate": 3.960979443851089e-06,
      "loss": 0.77761078,
      "num_input_tokens_seen": 16191020,
      "step": 758,
      "time_per_iteration": 2.732771396636963
    },
    {
      "auxiliary_loss_clip": 0.01186893,
      "auxiliary_loss_mlp": 0.00765683,
      "balance_loss_clip": 1.0636332,
      "balance_loss_mlp": 1.00116491,
      "epoch": 0.09126435399507005,
      "flos": 26145949438080.0,
      "grad_norm": 2.637157930277463,
      "language_loss": 0.78987849,
      "learning_rate": 3.96082617238075e-06,
      "loss": 0.80940425,
      "num_input_tokens_seen": 16213645,
      "step": 759,
      "time_per_iteration": 2.8481314182281494
    },
    {
      "auxiliary_loss_clip": 0.01179142,
      "auxiliary_loss_mlp": 0.01041833,
      "balance_loss_clip": 1.05646372,
      "balance_loss_mlp": 1.03037715,
      "epoch": 0.09138459688570913,
      "flos": 24388911757440.0,
      "grad_norm": 2.820188992832352,
      "language_loss": 0.79552698,
      "learning_rate": 3.960672603454639e-06,
      "loss": 0.81773674,
      "num_input_tokens_seen": 16233625,
      "step": 760,
      "time_per_iteration": 2.8200881481170654
    },
    {
      "auxiliary_loss_clip": 0.01206279,
      "auxiliary_loss_mlp": 0.01036126,
      "balance_loss_clip": 1.06136107,
      "balance_loss_mlp": 1.02365661,
      "epoch": 0.09150483977634823,
      "flos": 21032664756480.0,
      "grad_norm": 3.0453003578021205,
      "language_loss": 0.76769364,
      "learning_rate": 3.960518737096054e-06,
      "loss": 0.79011762,
      "num_input_tokens_seen": 16253255,
      "step": 761,
      "time_per_iteration": 2.890014410018921
    },
    {
      "auxiliary_loss_clip": 0.0125816,
      "auxiliary_loss_mlp": 0.01046689,
      "balance_loss_clip": 1.07194185,
      "balance_loss_mlp": 1.03472638,
      "epoch": 0.09162508266698731,
      "flos": 22856998567680.0,
      "grad_norm": 3.0936389686040413,
      "language_loss": 0.7249397,
      "learning_rate": 3.960364573328334e-06,
      "loss": 0.74798816,
      "num_input_tokens_seen": 16272580,
      "step": 762,
      "time_per_iteration": 2.6499125957489014
    },
    {
      "auxiliary_loss_clip": 0.01246159,
      "auxiliary_loss_mlp": 0.01039631,
      "balance_loss_clip": 1.07099068,
      "balance_loss_mlp": 1.0285567,
      "epoch": 0.0917453255576264,
      "flos": 21724411852800.0,
      "grad_norm": 2.1533458336488462,
      "language_loss": 0.88720888,
      "learning_rate": 3.9602101121748675e-06,
      "loss": 0.91006678,
      "num_input_tokens_seen": 16293075,
      "step": 763,
      "time_per_iteration": 2.6609740257263184
    },
    {
      "auxiliary_loss_clip": 0.01140797,
      "auxiliary_loss_mlp": 0.01047776,
      "balance_loss_clip": 1.05403495,
      "balance_loss_mlp": 1.0359503,
      "epoch": 0.0918655684482655,
      "flos": 14609215497600.0,
      "grad_norm": 2.194238221440174,
      "language_loss": 0.72452831,
      "learning_rate": 3.960055353659085e-06,
      "loss": 0.74641401,
      "num_input_tokens_seen": 16310185,
      "step": 764,
      "time_per_iteration": 2.799625873565674
    },
    {
      "auxiliary_loss_clip": 0.01257008,
      "auxiliary_loss_mlp": 0.01036111,
      "balance_loss_clip": 1.07306314,
      "balance_loss_mlp": 1.02495313,
      "epoch": 0.09198581133890459,
      "flos": 23435016226560.0,
      "grad_norm": 1.888571296516867,
      "language_loss": 0.83760387,
      "learning_rate": 3.959900297804465e-06,
      "loss": 0.86053503,
      "num_input_tokens_seen": 16330355,
      "step": 765,
      "time_per_iteration": 2.6377811431884766
    },
    {
      "auxiliary_loss_clip": 0.01191309,
      "auxiliary_loss_mlp": 0.01039048,
      "balance_loss_clip": 1.05946803,
      "balance_loss_mlp": 1.02792025,
      "epoch": 0.09210605422954368,
      "flos": 16795887753600.0,
      "grad_norm": 3.0221248223196837,
      "language_loss": 0.77422285,
      "learning_rate": 3.9597449446345276e-06,
      "loss": 0.79652643,
      "num_input_tokens_seen": 16347600,
      "step": 766,
      "time_per_iteration": 2.772815227508545
    },
    {
      "auxiliary_loss_clip": 0.01214258,
      "auxiliary_loss_mlp": 0.01047137,
      "balance_loss_clip": 1.0642153,
      "balance_loss_mlp": 1.03557992,
      "epoch": 0.09222629712018277,
      "flos": 22674249146880.0,
      "grad_norm": 2.436566877915815,
      "language_loss": 0.83335924,
      "learning_rate": 3.95958929417284e-06,
      "loss": 0.85597324,
      "num_input_tokens_seen": 16365755,
      "step": 767,
      "time_per_iteration": 2.6932523250579834
    },
    {
      "auxiliary_loss_clip": 0.01120316,
      "auxiliary_loss_mlp": 0.01006517,
      "balance_loss_clip": 1.04208577,
      "balance_loss_mlp": 1.00320292,
      "epoch": 0.09234654001082186,
      "flos": 69976756327680.0,
      "grad_norm": 0.7408888675965851,
      "language_loss": 0.58775818,
      "learning_rate": 3.9594333464430145e-06,
      "loss": 0.60902649,
      "num_input_tokens_seen": 16435245,
      "step": 768,
      "time_per_iteration": 3.6268184185028076
    },
    {
      "auxiliary_loss_clip": 0.01238785,
      "auxiliary_loss_mlp": 0.01039561,
      "balance_loss_clip": 1.07094121,
      "balance_loss_mlp": 1.02869463,
      "epoch": 0.09246678290146094,
      "flos": 20011437181440.0,
      "grad_norm": 2.0094586755412402,
      "language_loss": 0.87899256,
      "learning_rate": 3.959277101468709e-06,
      "loss": 0.90177596,
      "num_input_tokens_seen": 16454795,
      "step": 769,
      "time_per_iteration": 2.8988940715789795
    },
    {
      "auxiliary_loss_clip": 0.01207298,
      "auxiliary_loss_mlp": 0.00764084,
      "balance_loss_clip": 1.07161975,
      "balance_loss_mlp": 1.0009768,
      "epoch": 0.09258702579210004,
      "flos": 17747448900480.0,
      "grad_norm": 2.6228740803568082,
      "language_loss": 0.78512323,
      "learning_rate": 3.959120559273624e-06,
      "loss": 0.80483705,
      "num_input_tokens_seen": 16472580,
      "step": 770,
      "time_per_iteration": 3.7239251136779785
    },
    {
      "auxiliary_loss_clip": 0.0123175,
      "auxiliary_loss_mlp": 0.00764982,
      "balance_loss_clip": 1.06985712,
      "balance_loss_mlp": 1.00097156,
      "epoch": 0.09270726868273914,
      "flos": 20886544229760.0,
      "grad_norm": 2.302666982020081,
      "language_loss": 0.83388376,
      "learning_rate": 3.958963719881509e-06,
      "loss": 0.85385108,
      "num_input_tokens_seen": 16490670,
      "step": 771,
      "time_per_iteration": 2.7952115535736084
    },
    {
      "auxiliary_loss_clip": 0.01213855,
      "auxiliary_loss_mlp": 0.01053455,
      "balance_loss_clip": 1.06585562,
      "balance_loss_mlp": 1.04208255,
      "epoch": 0.09282751157337822,
      "flos": 17015697031680.0,
      "grad_norm": 2.5233796731183977,
      "language_loss": 0.93846607,
      "learning_rate": 3.958806583316154e-06,
      "loss": 0.96113914,
      "num_input_tokens_seen": 16508640,
      "step": 772,
      "time_per_iteration": 3.699009895324707
    },
    {
      "auxiliary_loss_clip": 0.01194865,
      "auxiliary_loss_mlp": 0.01042319,
      "balance_loss_clip": 1.0618031,
      "balance_loss_mlp": 1.03092909,
      "epoch": 0.09294775446401732,
      "flos": 32523647748480.0,
      "grad_norm": 2.815759678452415,
      "language_loss": 0.78820097,
      "learning_rate": 3.9586491496013985e-06,
      "loss": 0.81057286,
      "num_input_tokens_seen": 16531035,
      "step": 773,
      "time_per_iteration": 2.7469842433929443
    },
    {
      "auxiliary_loss_clip": 0.01229611,
      "auxiliary_loss_mlp": 0.01049441,
      "balance_loss_clip": 1.07157397,
      "balance_loss_mlp": 1.03768659,
      "epoch": 0.0930679973546564,
      "flos": 18259750627200.0,
      "grad_norm": 2.12654865696361,
      "language_loss": 0.82946897,
      "learning_rate": 3.958491418761124e-06,
      "loss": 0.85225952,
      "num_input_tokens_seen": 16548605,
      "step": 774,
      "time_per_iteration": 3.700348138809204
    },
    {
      "auxiliary_loss_clip": 0.01229319,
      "auxiliary_loss_mlp": 0.01042407,
      "balance_loss_clip": 1.06840813,
      "balance_loss_mlp": 1.03113604,
      "epoch": 0.0931882402452955,
      "flos": 21099745405440.0,
      "grad_norm": 2.5199470407268496,
      "language_loss": 0.72523189,
      "learning_rate": 3.958333390819258e-06,
      "loss": 0.74794918,
      "num_input_tokens_seen": 16565535,
      "step": 775,
      "time_per_iteration": 2.690011501312256
    },
    {
      "auxiliary_loss_clip": 0.01240622,
      "auxiliary_loss_mlp": 0.01043794,
      "balance_loss_clip": 1.07255006,
      "balance_loss_mlp": 1.03343499,
      "epoch": 0.0933084831359346,
      "flos": 24207275658240.0,
      "grad_norm": 2.233343871383868,
      "language_loss": 0.79870123,
      "learning_rate": 3.9581750657997754e-06,
      "loss": 0.82154536,
      "num_input_tokens_seen": 16584900,
      "step": 776,
      "time_per_iteration": 2.667844533920288
    },
    {
      "auxiliary_loss_clip": 0.01198937,
      "auxiliary_loss_mlp": 0.01039535,
      "balance_loss_clip": 1.05666828,
      "balance_loss_mlp": 1.02873445,
      "epoch": 0.09342872602657368,
      "flos": 25480272637440.0,
      "grad_norm": 2.229888297526721,
      "language_loss": 0.89551646,
      "learning_rate": 3.95801644372669e-06,
      "loss": 0.91790116,
      "num_input_tokens_seen": 16604805,
      "step": 777,
      "time_per_iteration": 3.714669704437256
    },
    {
      "auxiliary_loss_clip": 0.01241254,
      "auxiliary_loss_mlp": 0.01037598,
      "balance_loss_clip": 1.07092571,
      "balance_loss_mlp": 1.02645159,
      "epoch": 0.09354896891721277,
      "flos": 23149060053120.0,
      "grad_norm": 2.2630239807823442,
      "language_loss": 0.84390724,
      "learning_rate": 3.957857524624068e-06,
      "loss": 0.86669576,
      "num_input_tokens_seen": 16623685,
      "step": 778,
      "time_per_iteration": 2.614175796508789
    },
    {
      "auxiliary_loss_clip": 0.01242831,
      "auxiliary_loss_mlp": 0.01040151,
      "balance_loss_clip": 1.06981349,
      "balance_loss_mlp": 1.02876651,
      "epoch": 0.09366921180785186,
      "flos": 24279563779200.0,
      "grad_norm": 1.7686370966172624,
      "language_loss": 0.89628834,
      "learning_rate": 3.957698308516016e-06,
      "loss": 0.91911817,
      "num_input_tokens_seen": 16644985,
      "step": 779,
      "time_per_iteration": 2.7097535133361816
    },
    {
      "auxiliary_loss_clip": 0.01196133,
      "auxiliary_loss_mlp": 0.01048322,
      "balance_loss_clip": 1.06446552,
      "balance_loss_mlp": 1.03745067,
      "epoch": 0.09378945469849095,
      "flos": 18730036419840.0,
      "grad_norm": 2.2606879635244845,
      "language_loss": 0.82494384,
      "learning_rate": 3.957538795426688e-06,
      "loss": 0.84738839,
      "num_input_tokens_seen": 16662410,
      "step": 780,
      "time_per_iteration": 2.7804014682769775
    },
    {
      "auxiliary_loss_clip": 0.01208444,
      "auxiliary_loss_mlp": 0.01042239,
      "balance_loss_clip": 1.06475687,
      "balance_loss_mlp": 1.03085482,
      "epoch": 0.09390969758913004,
      "flos": 23218834222080.0,
      "grad_norm": 3.189679253240508,
      "language_loss": 0.77159017,
      "learning_rate": 3.9573789853802804e-06,
      "loss": 0.79409701,
      "num_input_tokens_seen": 16680885,
      "step": 781,
      "time_per_iteration": 2.7750344276428223
    },
    {
      "auxiliary_loss_clip": 0.0117366,
      "auxiliary_loss_mlp": 0.0103908,
      "balance_loss_clip": 1.05717182,
      "balance_loss_mlp": 1.02732015,
      "epoch": 0.09402994047976913,
      "flos": 19646728439040.0,
      "grad_norm": 3.3324361888238165,
      "language_loss": 0.74603301,
      "learning_rate": 3.957218878401037e-06,
      "loss": 0.7681604,
      "num_input_tokens_seen": 16699375,
      "step": 782,
      "time_per_iteration": 2.841315269470215
    },
    {
      "auxiliary_loss_clip": 0.01261326,
      "auxiliary_loss_mlp": 0.01041539,
      "balance_loss_clip": 1.07383609,
      "balance_loss_mlp": 1.02905822,
      "epoch": 0.09415018337040823,
      "flos": 29420463041280.0,
      "grad_norm": 2.2082920913419977,
      "language_loss": 0.8928709,
      "learning_rate": 3.957058474513246e-06,
      "loss": 0.91589952,
      "num_input_tokens_seen": 16719230,
      "step": 783,
      "time_per_iteration": 2.709139347076416
    },
    {
      "auxiliary_loss_clip": 0.01210224,
      "auxiliary_loss_mlp": 0.01040526,
      "balance_loss_clip": 1.06557608,
      "balance_loss_mlp": 1.02953458,
      "epoch": 0.09427042626104731,
      "flos": 24572092141440.0,
      "grad_norm": 1.8291056147921962,
      "language_loss": 0.78447253,
      "learning_rate": 3.956897773741241e-06,
      "loss": 0.80698007,
      "num_input_tokens_seen": 16738220,
      "step": 784,
      "time_per_iteration": 2.7862167358398438
    },
    {
      "auxiliary_loss_clip": 0.01130963,
      "auxiliary_loss_mlp": 0.01037976,
      "balance_loss_clip": 1.05532312,
      "balance_loss_mlp": 1.02606738,
      "epoch": 0.09439066915168641,
      "flos": 26359581576960.0,
      "grad_norm": 1.809266866706107,
      "language_loss": 0.7152254,
      "learning_rate": 3.956736776109398e-06,
      "loss": 0.73691481,
      "num_input_tokens_seen": 16759395,
      "step": 785,
      "time_per_iteration": 2.9162817001342773
    },
    {
      "auxiliary_loss_clip": 0.01236448,
      "auxiliary_loss_mlp": 0.01033235,
      "balance_loss_clip": 1.06882429,
      "balance_loss_mlp": 1.02261961,
      "epoch": 0.09451091204232549,
      "flos": 19427278296960.0,
      "grad_norm": 2.0104308751631828,
      "language_loss": 0.83761066,
      "learning_rate": 3.956575481642143e-06,
      "loss": 0.86030751,
      "num_input_tokens_seen": 16778285,
      "step": 786,
      "time_per_iteration": 3.3635671138763428
    },
    {
      "auxiliary_loss_clip": 0.01228702,
      "auxiliary_loss_mlp": 0.01040707,
      "balance_loss_clip": 1.06599092,
      "balance_loss_mlp": 1.02891088,
      "epoch": 0.09463115493296459,
      "flos": 25368051571200.0,
      "grad_norm": 2.7176483242102103,
      "language_loss": 0.74652159,
      "learning_rate": 3.956413890363943e-06,
      "loss": 0.7692157,
      "num_input_tokens_seen": 16795265,
      "step": 787,
      "time_per_iteration": 2.667904853820801
    },
    {
      "auxiliary_loss_clip": 0.01258438,
      "auxiliary_loss_mlp": 0.01046991,
      "balance_loss_clip": 1.07107544,
      "balance_loss_mlp": 1.03554702,
      "epoch": 0.09475139782360369,
      "flos": 10123254869760.0,
      "grad_norm": 2.212686951774838,
      "language_loss": 0.8197794,
      "learning_rate": 3.956252002299312e-06,
      "loss": 0.8428337,
      "num_input_tokens_seen": 16811165,
      "step": 788,
      "time_per_iteration": 2.8402774333953857
    },
    {
      "auxiliary_loss_clip": 0.01235535,
      "auxiliary_loss_mlp": 0.01037898,
      "balance_loss_clip": 1.06888759,
      "balance_loss_mlp": 1.02617407,
      "epoch": 0.09487164071424277,
      "flos": 17231088936960.0,
      "grad_norm": 1.979427030836698,
      "language_loss": 0.90861887,
      "learning_rate": 3.956089817472807e-06,
      "loss": 0.93135321,
      "num_input_tokens_seen": 16828470,
      "step": 789,
      "time_per_iteration": 2.6758711338043213
    },
    {
      "auxiliary_loss_clip": 0.01230113,
      "auxiliary_loss_mlp": 0.01034729,
      "balance_loss_clip": 1.06539977,
      "balance_loss_mlp": 1.02348781,
      "epoch": 0.09499188360488187,
      "flos": 30849564528000.0,
      "grad_norm": 2.255059855473432,
      "language_loss": 0.8525418,
      "learning_rate": 3.955927335909032e-06,
      "loss": 0.87519026,
      "num_input_tokens_seen": 16851680,
      "step": 790,
      "time_per_iteration": 2.7788097858428955
    },
    {
      "auxiliary_loss_clip": 0.01179224,
      "auxiliary_loss_mlp": 0.0104053,
      "balance_loss_clip": 1.06005311,
      "balance_loss_mlp": 1.02999759,
      "epoch": 0.09511212649552095,
      "flos": 29351694453120.0,
      "grad_norm": 2.440208888465043,
      "language_loss": 0.75818264,
      "learning_rate": 3.955764557632634e-06,
      "loss": 0.78038013,
      "num_input_tokens_seen": 16871490,
      "step": 791,
      "time_per_iteration": 2.871212959289551
    },
    {
      "auxiliary_loss_clip": 0.01186814,
      "auxiliary_loss_mlp": 0.0104359,
      "balance_loss_clip": 1.05733585,
      "balance_loss_mlp": 1.03221726,
      "epoch": 0.09523236938616005,
      "flos": 10378687461120.0,
      "grad_norm": 2.2305285378686377,
      "language_loss": 0.94835413,
      "learning_rate": 3.955601482668309e-06,
      "loss": 0.97065812,
      "num_input_tokens_seen": 16889350,
      "step": 792,
      "time_per_iteration": 2.704331398010254
    },
    {
      "auxiliary_loss_clip": 0.01218495,
      "auxiliary_loss_mlp": 0.01034703,
      "balance_loss_clip": 1.06598043,
      "balance_loss_mlp": 1.02338409,
      "epoch": 0.09535261227679913,
      "flos": 19061815368960.0,
      "grad_norm": 1.9839803760649894,
      "language_loss": 0.8840642,
      "learning_rate": 3.955438111040794e-06,
      "loss": 0.90659618,
      "num_input_tokens_seen": 16907625,
      "step": 793,
      "time_per_iteration": 2.754643440246582
    },
    {
      "auxiliary_loss_clip": 0.01235984,
      "auxiliary_loss_mlp": 0.01031896,
      "balance_loss_clip": 1.06721544,
      "balance_loss_mlp": 1.02073812,
      "epoch": 0.09547285516743823,
      "flos": 20922993555840.0,
      "grad_norm": 1.9926546255775897,
      "language_loss": 0.80135125,
      "learning_rate": 3.955274442774873e-06,
      "loss": 0.8240301,
      "num_input_tokens_seen": 16926205,
      "step": 794,
      "time_per_iteration": 2.670908212661743
    },
    {
      "auxiliary_loss_clip": 0.0124404,
      "auxiliary_loss_mlp": 0.01042432,
      "balance_loss_clip": 1.07341683,
      "balance_loss_mlp": 1.03073716,
      "epoch": 0.09559309805807732,
      "flos": 30154405639680.0,
      "grad_norm": 2.608019333155879,
      "language_loss": 0.70601368,
      "learning_rate": 3.9551104778953725e-06,
      "loss": 0.72887838,
      "num_input_tokens_seen": 16946500,
      "step": 795,
      "time_per_iteration": 2.7498340606689453
    },
    {
      "auxiliary_loss_clip": 0.01255072,
      "auxiliary_loss_mlp": 0.01035191,
      "balance_loss_clip": 1.06891489,
      "balance_loss_mlp": 1.0234369,
      "epoch": 0.0957133409487164,
      "flos": 21066743784960.0,
      "grad_norm": 1.9417488937166247,
      "language_loss": 0.85216105,
      "learning_rate": 3.954946216427167e-06,
      "loss": 0.87506372,
      "num_input_tokens_seen": 16966960,
      "step": 796,
      "time_per_iteration": 3.6037991046905518
    },
    {
      "auxiliary_loss_clip": 0.01114436,
      "auxiliary_loss_mlp": 0.01010798,
      "balance_loss_clip": 1.04110789,
      "balance_loss_mlp": 1.00734091,
      "epoch": 0.0958335838393555,
      "flos": 71297979315840.0,
      "grad_norm": 0.879623715017303,
      "language_loss": 0.61550283,
      "learning_rate": 3.954781658395176e-06,
      "loss": 0.63675523,
      "num_input_tokens_seen": 17023215,
      "step": 797,
      "time_per_iteration": 3.332825183868408
    },
    {
      "auxiliary_loss_clip": 0.0124084,
      "auxiliary_loss_mlp": 0.0103717,
      "balance_loss_clip": 1.0709511,
      "balance_loss_mlp": 1.02560079,
      "epoch": 0.09595382672999458,
      "flos": 21872974504320.0,
      "grad_norm": 2.0747212153639647,
      "language_loss": 0.92141771,
      "learning_rate": 3.95461680382436e-06,
      "loss": 0.94419783,
      "num_input_tokens_seen": 17042140,
      "step": 798,
      "time_per_iteration": 3.6332335472106934
    },
    {
      "auxiliary_loss_clip": 0.01216145,
      "auxiliary_loss_mlp": 0.00765503,
      "balance_loss_clip": 1.06689596,
      "balance_loss_mlp": 1.0013088,
      "epoch": 0.09607406962063368,
      "flos": 18695562341760.0,
      "grad_norm": 2.882883760037335,
      "language_loss": 0.86362046,
      "learning_rate": 3.9544516527397295e-06,
      "loss": 0.88343692,
      "num_input_tokens_seen": 17058490,
      "step": 799,
      "time_per_iteration": 2.7005677223205566
    },
    {
      "auxiliary_loss_clip": 0.01183597,
      "auxiliary_loss_mlp": 0.01044407,
      "balance_loss_clip": 1.05781198,
      "balance_loss_mlp": 1.0324682,
      "epoch": 0.09619431251127276,
      "flos": 22568456615040.0,
      "grad_norm": 1.8682557032993272,
      "language_loss": 0.80630112,
      "learning_rate": 3.954286205166338e-06,
      "loss": 0.82858121,
      "num_input_tokens_seen": 17079655,
      "step": 800,
      "time_per_iteration": 3.7603538036346436
    },
    {
      "auxiliary_loss_clip": 0.0120255,
      "auxiliary_loss_mlp": 0.00765348,
      "balance_loss_clip": 1.06343615,
      "balance_loss_mlp": 1.00128388,
      "epoch": 0.09631455540191186,
      "flos": 14246230608000.0,
      "grad_norm": 4.237870559771355,
      "language_loss": 0.83922505,
      "learning_rate": 3.954120461129282e-06,
      "loss": 0.858904,
      "num_input_tokens_seen": 17097065,
      "step": 801,
      "time_per_iteration": 2.735997200012207
    },
    {
      "auxiliary_loss_clip": 0.01177644,
      "auxiliary_loss_mlp": 0.01039197,
      "balance_loss_clip": 1.05932677,
      "balance_loss_mlp": 1.02746081,
      "epoch": 0.09643479829255096,
      "flos": 20740387789440.0,
      "grad_norm": 2.53794951563969,
      "language_loss": 0.84056884,
      "learning_rate": 3.953954420653706e-06,
      "loss": 0.8627373,
      "num_input_tokens_seen": 17114090,
      "step": 802,
      "time_per_iteration": 3.7952117919921875
    },
    {
      "auxiliary_loss_clip": 0.0120583,
      "auxiliary_loss_mlp": 0.01044055,
      "balance_loss_clip": 1.06392145,
      "balance_loss_mlp": 1.03177595,
      "epoch": 0.09655504118319004,
      "flos": 24420476833920.0,
      "grad_norm": 2.0083106686516174,
      "language_loss": 0.88098556,
      "learning_rate": 3.953788083764798e-06,
      "loss": 0.9034844,
      "num_input_tokens_seen": 17133325,
      "step": 803,
      "time_per_iteration": 2.7643072605133057
    },
    {
      "auxiliary_loss_clip": 0.01180817,
      "auxiliary_loss_mlp": 0.0104342,
      "balance_loss_clip": 1.05962074,
      "balance_loss_mlp": 1.03200579,
      "epoch": 0.09667528407382914,
      "flos": 18441961344000.0,
      "grad_norm": 2.54832600449551,
      "language_loss": 0.92054343,
      "learning_rate": 3.953621450487792e-06,
      "loss": 0.9427858,
      "num_input_tokens_seen": 17151945,
      "step": 804,
      "time_per_iteration": 2.7721176147460938
    },
    {
      "auxiliary_loss_clip": 0.01152356,
      "auxiliary_loss_mlp": 0.01002885,
      "balance_loss_clip": 1.04312229,
      "balance_loss_mlp": 0.99938059,
      "epoch": 0.09679552696446822,
      "flos": 70816455544320.0,
      "grad_norm": 0.8400030516587856,
      "language_loss": 0.6118924,
      "learning_rate": 3.953454520847964e-06,
      "loss": 0.63344479,
      "num_input_tokens_seen": 17216790,
      "step": 805,
      "time_per_iteration": 3.379054307937622
    },
    {
      "auxiliary_loss_clip": 0.01203422,
      "auxiliary_loss_mlp": 0.01046007,
      "balance_loss_clip": 1.06440091,
      "balance_loss_mlp": 1.03264987,
      "epoch": 0.09691576985510732,
      "flos": 21945514020480.0,
      "grad_norm": 2.567525217232246,
      "language_loss": 0.73865843,
      "learning_rate": 3.9532872948706395e-06,
      "loss": 0.76115268,
      "num_input_tokens_seen": 17236285,
      "step": 806,
      "time_per_iteration": 2.76947283744812
    },
    {
      "auxiliary_loss_clip": 0.01235009,
      "auxiliary_loss_mlp": 0.0104991,
      "balance_loss_clip": 1.06625628,
      "balance_loss_mlp": 1.03840685,
      "epoch": 0.09703601274574641,
      "flos": 17965211103360.0,
      "grad_norm": 2.686819613709095,
      "language_loss": 0.82935905,
      "learning_rate": 3.9531197725811845e-06,
      "loss": 0.85220826,
      "num_input_tokens_seen": 17251670,
      "step": 807,
      "time_per_iteration": 2.673630475997925
    },
    {
      "auxiliary_loss_clip": 0.01204348,
      "auxiliary_loss_mlp": 0.00765024,
      "balance_loss_clip": 1.07013905,
      "balance_loss_mlp": 1.00118864,
      "epoch": 0.0971562556363855,
      "flos": 22162162901760.0,
      "grad_norm": 1.7328838170177063,
      "language_loss": 0.87836361,
      "learning_rate": 3.952951954005013e-06,
      "loss": 0.89805734,
      "num_input_tokens_seen": 17271355,
      "step": 808,
      "time_per_iteration": 2.7688395977020264
    },
    {
      "auxiliary_loss_clip": 0.01223772,
      "auxiliary_loss_mlp": 0.01045168,
      "balance_loss_clip": 1.07094026,
      "balance_loss_mlp": 1.03363442,
      "epoch": 0.0972764985270246,
      "flos": 25848716394240.0,
      "grad_norm": 1.8159603727233125,
      "language_loss": 0.84784389,
      "learning_rate": 3.952783839167584e-06,
      "loss": 0.87053323,
      "num_input_tokens_seen": 17291400,
      "step": 809,
      "time_per_iteration": 2.7807745933532715
    },
    {
      "auxiliary_loss_clip": 0.01248776,
      "auxiliary_loss_mlp": 0.0105276,
      "balance_loss_clip": 1.06844807,
      "balance_loss_mlp": 1.04141092,
      "epoch": 0.09739674141766368,
      "flos": 20339373375360.0,
      "grad_norm": 3.155415363296611,
      "language_loss": 0.74243653,
      "learning_rate": 3.952615428094398e-06,
      "loss": 0.76545191,
      "num_input_tokens_seen": 17310920,
      "step": 810,
      "time_per_iteration": 2.714893341064453
    },
    {
      "auxiliary_loss_clip": 0.01200271,
      "auxiliary_loss_mlp": 0.01044466,
      "balance_loss_clip": 1.06191599,
      "balance_loss_mlp": 1.0324676,
      "epoch": 0.09751698430830277,
      "flos": 15743059188480.0,
      "grad_norm": 1.7856697431861013,
      "language_loss": 0.73307681,
      "learning_rate": 3.952446720811004e-06,
      "loss": 0.75552416,
      "num_input_tokens_seen": 17329245,
      "step": 811,
      "time_per_iteration": 2.7329516410827637
    },
    {
      "auxiliary_loss_clip": 0.01116946,
      "auxiliary_loss_mlp": 0.01002246,
      "balance_loss_clip": 1.03623104,
      "balance_loss_mlp": 0.99874097,
      "epoch": 0.09763722719894186,
      "flos": 63716806800000.0,
      "grad_norm": 0.8408178564428109,
      "language_loss": 0.63565826,
      "learning_rate": 3.952277717342995e-06,
      "loss": 0.65685016,
      "num_input_tokens_seen": 17395680,
      "step": 812,
      "time_per_iteration": 3.3864805698394775
    },
    {
      "auxiliary_loss_clip": 0.01223818,
      "auxiliary_loss_mlp": 0.01038134,
      "balance_loss_clip": 1.06641161,
      "balance_loss_mlp": 1.02564692,
      "epoch": 0.09775747008958095,
      "flos": 22090916275200.0,
      "grad_norm": 2.0414177520033405,
      "language_loss": 0.85318661,
      "learning_rate": 3.952108417716009e-06,
      "loss": 0.87580609,
      "num_input_tokens_seen": 17415135,
      "step": 813,
      "time_per_iteration": 2.7075788974761963
    },
    {
      "auxiliary_loss_clip": 0.01156625,
      "auxiliary_loss_mlp": 0.01044249,
      "balance_loss_clip": 1.06608152,
      "balance_loss_mlp": 1.03023601,
      "epoch": 0.09787771298022005,
      "flos": 21286050272640.0,
      "grad_norm": 1.9409691608860036,
      "language_loss": 0.85082096,
      "learning_rate": 3.951938821955727e-06,
      "loss": 0.87282968,
      "num_input_tokens_seen": 17434535,
      "step": 814,
      "time_per_iteration": 2.9245986938476562
    },
    {
      "auxiliary_loss_clip": 0.0120074,
      "auxiliary_loss_mlp": 0.01043453,
      "balance_loss_clip": 1.06607461,
      "balance_loss_mlp": 1.03149676,
      "epoch": 0.09799795587085913,
      "flos": 22054574689920.0,
      "grad_norm": 1.634037680675284,
      "language_loss": 0.76570815,
      "learning_rate": 3.9517689300878786e-06,
      "loss": 0.78815007,
      "num_input_tokens_seen": 17454270,
      "step": 815,
      "time_per_iteration": 3.056788682937622
    },
    {
      "auxiliary_loss_clip": 0.01242449,
      "auxiliary_loss_mlp": 0.01050282,
      "balance_loss_clip": 1.06678677,
      "balance_loss_mlp": 1.03830159,
      "epoch": 0.09811819876149823,
      "flos": 22163743100160.0,
      "grad_norm": 1.7554967779842434,
      "language_loss": 0.78886276,
      "learning_rate": 3.951598742138236e-06,
      "loss": 0.81179005,
      "num_input_tokens_seen": 17472995,
      "step": 816,
      "time_per_iteration": 2.676743745803833
    },
    {
      "auxiliary_loss_clip": 0.01247103,
      "auxiliary_loss_mlp": 0.0104276,
      "balance_loss_clip": 1.07103896,
      "balance_loss_mlp": 1.03116655,
      "epoch": 0.09823844165213731,
      "flos": 22231111057920.0,
      "grad_norm": 2.0084266418768695,
      "language_loss": 0.79502797,
      "learning_rate": 3.951428258132615e-06,
      "loss": 0.81792665,
      "num_input_tokens_seen": 17491115,
      "step": 817,
      "time_per_iteration": 2.701756000518799
    },
    {
      "auxiliary_loss_clip": 0.01206814,
      "auxiliary_loss_mlp": 0.0076575,
      "balance_loss_clip": 1.0646987,
      "balance_loss_mlp": 1.00102365,
      "epoch": 0.09835868454277641,
      "flos": 22487728798080.0,
      "grad_norm": 2.0141114091048804,
      "language_loss": 0.84532785,
      "learning_rate": 3.951257478096879e-06,
      "loss": 0.86505347,
      "num_input_tokens_seen": 17509480,
      "step": 818,
      "time_per_iteration": 2.796668529510498
    },
    {
      "auxiliary_loss_clip": 0.01213218,
      "auxiliary_loss_mlp": 0.00765854,
      "balance_loss_clip": 1.06505489,
      "balance_loss_mlp": 1.00109935,
      "epoch": 0.0984789274334155,
      "flos": 16362554077440.0,
      "grad_norm": 2.8210357275869784,
      "language_loss": 0.68684852,
      "learning_rate": 3.951086402056936e-06,
      "loss": 0.70663923,
      "num_input_tokens_seen": 17524080,
      "step": 819,
      "time_per_iteration": 2.7870256900787354
    },
    {
      "auxiliary_loss_clip": 0.01182206,
      "auxiliary_loss_mlp": 0.01049855,
      "balance_loss_clip": 1.05865347,
      "balance_loss_mlp": 1.03861344,
      "epoch": 0.09859917032405459,
      "flos": 24243545416320.0,
      "grad_norm": 2.36836589483705,
      "language_loss": 0.83493483,
      "learning_rate": 3.950915030038735e-06,
      "loss": 0.85725546,
      "num_input_tokens_seen": 17543875,
      "step": 820,
      "time_per_iteration": 2.863797903060913
    },
    {
      "auxiliary_loss_clip": 0.01194744,
      "auxiliary_loss_mlp": 0.01038797,
      "balance_loss_clip": 1.05924833,
      "balance_loss_mlp": 1.02694726,
      "epoch": 0.09871941321469369,
      "flos": 17420195064960.0,
      "grad_norm": 2.549714972177991,
      "language_loss": 0.835361,
      "learning_rate": 3.9507433620682765e-06,
      "loss": 0.85769641,
      "num_input_tokens_seen": 17560810,
      "step": 821,
      "time_per_iteration": 2.750748634338379
    },
    {
      "auxiliary_loss_clip": 0.01260228,
      "auxiliary_loss_mlp": 0.01043429,
      "balance_loss_clip": 1.07364929,
      "balance_loss_mlp": 1.03141868,
      "epoch": 0.09883965610533277,
      "flos": 28477341590400.0,
      "grad_norm": 1.92069080884834,
      "language_loss": 0.88145888,
      "learning_rate": 3.9505713981716e-06,
      "loss": 0.90449548,
      "num_input_tokens_seen": 17583640,
      "step": 822,
      "time_per_iteration": 3.713313341140747
    },
    {
      "auxiliary_loss_clip": 0.01220914,
      "auxiliary_loss_mlp": 0.0076443,
      "balance_loss_clip": 1.06933129,
      "balance_loss_mlp": 1.00102532,
      "epoch": 0.09895989899597187,
      "flos": 23693932437120.0,
      "grad_norm": 1.8055485946871768,
      "language_loss": 0.81233001,
      "learning_rate": 3.950399138374795e-06,
      "loss": 0.83218348,
      "num_input_tokens_seen": 17602720,
      "step": 823,
      "time_per_iteration": 2.7772958278656006
    },
    {
      "auxiliary_loss_clip": 0.01202437,
      "auxiliary_loss_mlp": 0.01042124,
      "balance_loss_clip": 1.06388104,
      "balance_loss_mlp": 1.03047097,
      "epoch": 0.09908014188661095,
      "flos": 24679608526080.0,
      "grad_norm": 8.159035722177368,
      "language_loss": 0.74526125,
      "learning_rate": 3.95022658270399e-06,
      "loss": 0.76770687,
      "num_input_tokens_seen": 17623085,
      "step": 824,
      "time_per_iteration": 3.7653884887695312
    },
    {
      "auxiliary_loss_clip": 0.01207931,
      "auxiliary_loss_mlp": 0.01050139,
      "balance_loss_clip": 1.06859553,
      "balance_loss_mlp": 1.03856397,
      "epoch": 0.09920038477725004,
      "flos": 14064307200000.0,
      "grad_norm": 2.0744973053451687,
      "language_loss": 0.78184563,
      "learning_rate": 3.9500537311853635e-06,
      "loss": 0.80442631,
      "num_input_tokens_seen": 17641040,
      "step": 825,
      "time_per_iteration": 2.811979055404663
    },
    {
      "auxiliary_loss_clip": 0.01183038,
      "auxiliary_loss_mlp": 0.00765774,
      "balance_loss_clip": 1.05667889,
      "balance_loss_mlp": 1.0009048,
      "epoch": 0.09932062766788914,
      "flos": 13407070095360.0,
      "grad_norm": 2.5099876393958485,
      "language_loss": 0.83185947,
      "learning_rate": 3.949880583845136e-06,
      "loss": 0.85134757,
      "num_input_tokens_seen": 17659115,
      "step": 826,
      "time_per_iteration": 3.7744903564453125
    },
    {
      "auxiliary_loss_clip": 0.01224792,
      "auxiliary_loss_mlp": 0.01044024,
      "balance_loss_clip": 1.06755471,
      "balance_loss_mlp": 1.03153658,
      "epoch": 0.09944087055852822,
      "flos": 19500751566720.0,
      "grad_norm": 1.7707350857240407,
      "language_loss": 0.81429732,
      "learning_rate": 3.949707140709575e-06,
      "loss": 0.83698553,
      "num_input_tokens_seen": 17678845,
      "step": 827,
      "time_per_iteration": 3.6667263507843018
    },
    {
      "auxiliary_loss_clip": 0.01261966,
      "auxiliary_loss_mlp": 0.01047758,
      "balance_loss_clip": 1.07394958,
      "balance_loss_mlp": 1.03537858,
      "epoch": 0.09956111344916732,
      "flos": 17749100926080.0,
      "grad_norm": 3.0117258070951602,
      "language_loss": 0.83521938,
      "learning_rate": 3.949533401804991e-06,
      "loss": 0.85831666,
      "num_input_tokens_seen": 17695750,
      "step": 828,
      "time_per_iteration": 2.62721848487854
    },
    {
      "auxiliary_loss_clip": 0.01218084,
      "auxiliary_loss_mlp": 0.0104559,
      "balance_loss_clip": 1.06654942,
      "balance_loss_mlp": 1.03289998,
      "epoch": 0.0996813563398064,
      "flos": 17967581400960.0,
      "grad_norm": 2.3511730754978943,
      "language_loss": 0.90795547,
      "learning_rate": 3.949359367157739e-06,
      "loss": 0.9305923,
      "num_input_tokens_seen": 17714445,
      "step": 829,
      "time_per_iteration": 2.732100009918213
    },
    {
      "auxiliary_loss_clip": 0.01200916,
      "auxiliary_loss_mlp": 0.01046665,
      "balance_loss_clip": 1.06298018,
      "balance_loss_mlp": 1.03387952,
      "epoch": 0.0998015992304455,
      "flos": 17457039440640.0,
      "grad_norm": 2.6511180760411963,
      "language_loss": 0.75676137,
      "learning_rate": 3.949185036794222e-06,
      "loss": 0.77923715,
      "num_input_tokens_seen": 17732455,
      "step": 830,
      "time_per_iteration": 2.7167558670043945
    },
    {
      "auxiliary_loss_clip": 0.01237402,
      "auxiliary_loss_mlp": 0.01047265,
      "balance_loss_clip": 1.06784248,
      "balance_loss_mlp": 1.03566647,
      "epoch": 0.0999218421210846,
      "flos": 25888757080320.0,
      "grad_norm": 1.7149304500211973,
      "language_loss": 0.79006481,
      "learning_rate": 3.949010410740884e-06,
      "loss": 0.81291151,
      "num_input_tokens_seen": 17755280,
      "step": 831,
      "time_per_iteration": 2.719242572784424
    },
    {
      "auxiliary_loss_clip": 0.01262069,
      "auxiliary_loss_mlp": 0.01043064,
      "balance_loss_clip": 1.07366097,
      "balance_loss_mlp": 1.03118467,
      "epoch": 0.10004208501172368,
      "flos": 21215916967680.0,
      "grad_norm": 1.879465373832776,
      "language_loss": 0.86511189,
      "learning_rate": 3.948835489024216e-06,
      "loss": 0.88816321,
      "num_input_tokens_seen": 17775015,
      "step": 832,
      "time_per_iteration": 2.6643288135528564
    },
    {
      "auxiliary_loss_clip": 0.01194558,
      "auxiliary_loss_mlp": 0.01047906,
      "balance_loss_clip": 1.06090808,
      "balance_loss_mlp": 1.03624725,
      "epoch": 0.10016232790236278,
      "flos": 17348409734400.0,
      "grad_norm": 2.0826829651860095,
      "language_loss": 0.90377605,
      "learning_rate": 3.948660271670755e-06,
      "loss": 0.92620069,
      "num_input_tokens_seen": 17792165,
      "step": 833,
      "time_per_iteration": 2.7432425022125244
    },
    {
      "auxiliary_loss_clip": 0.0120818,
      "auxiliary_loss_mlp": 0.0104233,
      "balance_loss_clip": 1.06715357,
      "balance_loss_mlp": 1.03058791,
      "epoch": 0.10028257079300186,
      "flos": 25666541591040.0,
      "grad_norm": 2.232794503103066,
      "language_loss": 0.84390932,
      "learning_rate": 3.948484758707079e-06,
      "loss": 0.86641443,
      "num_input_tokens_seen": 17811765,
      "step": 834,
      "time_per_iteration": 2.777538299560547
    },
    {
      "auxiliary_loss_clip": 0.01228176,
      "auxiliary_loss_mlp": 0.01046354,
      "balance_loss_clip": 1.06832004,
      "balance_loss_mlp": 1.03390241,
      "epoch": 0.10040281368364096,
      "flos": 25156035544320.0,
      "grad_norm": 2.322044389564842,
      "language_loss": 0.83567065,
      "learning_rate": 3.948308950159815e-06,
      "loss": 0.85841596,
      "num_input_tokens_seen": 17830445,
      "step": 835,
      "time_per_iteration": 2.840890884399414
    },
    {
      "auxiliary_loss_clip": 0.01195938,
      "auxiliary_loss_mlp": 0.01041189,
      "balance_loss_clip": 1.06646538,
      "balance_loss_mlp": 1.02868438,
      "epoch": 0.10052305657428004,
      "flos": 17603303621760.0,
      "grad_norm": 2.4826514564229645,
      "language_loss": 0.76105332,
      "learning_rate": 3.9481328460556326e-06,
      "loss": 0.78342462,
      "num_input_tokens_seen": 17847665,
      "step": 836,
      "time_per_iteration": 2.806753635406494
    },
    {
      "auxiliary_loss_clip": 0.01181371,
      "auxiliary_loss_mlp": 0.01037745,
      "balance_loss_clip": 1.05804491,
      "balance_loss_mlp": 1.02603853,
      "epoch": 0.10064329946491914,
      "flos": 18660154510080.0,
      "grad_norm": 2.049086624022854,
      "language_loss": 0.89565498,
      "learning_rate": 3.9479564464212455e-06,
      "loss": 0.91784614,
      "num_input_tokens_seen": 17866825,
      "step": 837,
      "time_per_iteration": 2.8701860904693604
    },
    {
      "auxiliary_loss_clip": 0.01212376,
      "auxiliary_loss_mlp": 0.01038519,
      "balance_loss_clip": 1.06564569,
      "balance_loss_mlp": 1.02621698,
      "epoch": 0.10076354235555823,
      "flos": 17199056983680.0,
      "grad_norm": 2.5711469662816486,
      "language_loss": 0.76091975,
      "learning_rate": 3.947779751283414e-06,
      "loss": 0.78342873,
      "num_input_tokens_seen": 17883995,
      "step": 838,
      "time_per_iteration": 2.72456693649292
    },
    {
      "auxiliary_loss_clip": 0.01231702,
      "auxiliary_loss_mlp": 0.01054793,
      "balance_loss_clip": 1.07030118,
      "balance_loss_mlp": 1.04172814,
      "epoch": 0.10088378524619732,
      "flos": 22962252395520.0,
      "grad_norm": 1.781020286149406,
      "language_loss": 0.76066703,
      "learning_rate": 3.947602760668944e-06,
      "loss": 0.78353196,
      "num_input_tokens_seen": 17903785,
      "step": 839,
      "time_per_iteration": 2.644556760787964
    },
    {
      "auxiliary_loss_clip": 0.01230081,
      "auxiliary_loss_mlp": 0.01039701,
      "balance_loss_clip": 1.07170558,
      "balance_loss_mlp": 1.02770281,
      "epoch": 0.10100402813683641,
      "flos": 37885828746240.0,
      "grad_norm": 4.092531724220394,
      "language_loss": 0.71408957,
      "learning_rate": 3.947425474604684e-06,
      "loss": 0.73678744,
      "num_input_tokens_seen": 17927720,
      "step": 840,
      "time_per_iteration": 2.8449835777282715
    },
    {
      "auxiliary_loss_clip": 0.01199929,
      "auxiliary_loss_mlp": 0.01044385,
      "balance_loss_clip": 1.06477952,
      "balance_loss_mlp": 1.03227973,
      "epoch": 0.1011242710274755,
      "flos": 21543458112000.0,
      "grad_norm": 2.1901002439142823,
      "language_loss": 0.92389607,
      "learning_rate": 3.947247893117528e-06,
      "loss": 0.94633913,
      "num_input_tokens_seen": 17946225,
      "step": 841,
      "time_per_iteration": 2.721627712249756
    },
    {
      "auxiliary_loss_clip": 0.01219896,
      "auxiliary_loss_mlp": 0.01043676,
      "balance_loss_clip": 1.06637096,
      "balance_loss_mlp": 1.03075409,
      "epoch": 0.10124451391811459,
      "flos": 13621456419840.0,
      "grad_norm": 3.734100729867583,
      "language_loss": 0.6915009,
      "learning_rate": 3.947070016234413e-06,
      "loss": 0.7141366,
      "num_input_tokens_seen": 17962015,
      "step": 842,
      "time_per_iteration": 2.723480463027954
    },
    {
      "auxiliary_loss_clip": 0.0121377,
      "auxiliary_loss_mlp": 0.01044129,
      "balance_loss_clip": 1.06984603,
      "balance_loss_mlp": 1.03137982,
      "epoch": 0.10136475680875369,
      "flos": 16649228522880.0,
      "grad_norm": 2.7618491785767256,
      "language_loss": 0.74976104,
      "learning_rate": 3.946891843982326e-06,
      "loss": 0.77234006,
      "num_input_tokens_seen": 17979680,
      "step": 843,
      "time_per_iteration": 2.7643344402313232
    },
    {
      "auxiliary_loss_clip": 0.01155111,
      "auxiliary_loss_mlp": 0.01047196,
      "balance_loss_clip": 1.06061816,
      "balance_loss_mlp": 1.03436875,
      "epoch": 0.10148499969939277,
      "flos": 19461034103040.0,
      "grad_norm": 5.840333547120501,
      "language_loss": 0.74265909,
      "learning_rate": 3.9467133763882935e-06,
      "loss": 0.76468217,
      "num_input_tokens_seen": 17998145,
      "step": 844,
      "time_per_iteration": 2.7796790599823
    },
    {
      "auxiliary_loss_clip": 0.01237982,
      "auxiliary_loss_mlp": 0.01051872,
      "balance_loss_clip": 1.06918263,
      "balance_loss_mlp": 1.03982627,
      "epoch": 0.10160524259003187,
      "flos": 21104988791040.0,
      "grad_norm": 2.912928387307064,
      "language_loss": 0.86219418,
      "learning_rate": 3.9465346134793905e-06,
      "loss": 0.88509274,
      "num_input_tokens_seen": 18017955,
      "step": 845,
      "time_per_iteration": 2.7304959297180176
    },
    {
      "auxiliary_loss_clip": 0.01260227,
      "auxiliary_loss_mlp": 0.01038819,
      "balance_loss_clip": 1.07466269,
      "balance_loss_mlp": 1.02803659,
      "epoch": 0.10172548548067095,
      "flos": 17712687513600.0,
      "grad_norm": 3.4949538721923434,
      "language_loss": 0.79559487,
      "learning_rate": 3.9463555552827335e-06,
      "loss": 0.8185854,
      "num_input_tokens_seen": 18035125,
      "step": 846,
      "time_per_iteration": 2.6707262992858887
    },
    {
      "auxiliary_loss_clip": 0.01183495,
      "auxiliary_loss_mlp": 0.01040025,
      "balance_loss_clip": 1.06261528,
      "balance_loss_mlp": 1.02879524,
      "epoch": 0.10184572837131005,
      "flos": 21104845136640.0,
      "grad_norm": 29.124003259795245,
      "language_loss": 0.86140513,
      "learning_rate": 3.946176201825487e-06,
      "loss": 0.88364029,
      "num_input_tokens_seen": 18053160,
      "step": 847,
      "time_per_iteration": 2.8187673091888428
    },
    {
      "auxiliary_loss_clip": 0.01250267,
      "auxiliary_loss_mlp": 0.01047507,
      "balance_loss_clip": 1.07444906,
      "balance_loss_mlp": 1.0344243,
      "epoch": 0.10196597126194913,
      "flos": 26067591918720.0,
      "grad_norm": 2.2143581812484023,
      "language_loss": 0.83379412,
      "learning_rate": 3.9459965531348575e-06,
      "loss": 0.85677183,
      "num_input_tokens_seen": 18072815,
      "step": 848,
      "time_per_iteration": 3.6600611209869385
    },
    {
      "auxiliary_loss_clip": 0.0121409,
      "auxiliary_loss_mlp": 0.00765664,
      "balance_loss_clip": 1.06339705,
      "balance_loss_mlp": 1.00093651,
      "epoch": 0.10208621415258823,
      "flos": 29314634595840.0,
      "grad_norm": 3.0016117167714116,
      "language_loss": 0.85279894,
      "learning_rate": 3.945816609238098e-06,
      "loss": 0.8725965,
      "num_input_tokens_seen": 18092225,
      "step": 849,
      "time_per_iteration": 4.236155033111572
    },
    {
      "auxiliary_loss_clip": 0.0117576,
      "auxiliary_loss_mlp": 0.01051144,
      "balance_loss_clip": 1.06032562,
      "balance_loss_mlp": 1.03911614,
      "epoch": 0.10220645704322733,
      "flos": 23805794367360.0,
      "grad_norm": 2.1142153139116724,
      "language_loss": 0.85051203,
      "learning_rate": 3.945636370162507e-06,
      "loss": 0.87278116,
      "num_input_tokens_seen": 18112335,
      "step": 850,
      "time_per_iteration": 2.8842599391937256
    },
    {
      "auxiliary_loss_clip": 0.01226001,
      "auxiliary_loss_mlp": 0.01047843,
      "balance_loss_clip": 1.06864309,
      "balance_loss_mlp": 1.036304,
      "epoch": 0.10232669993386641,
      "flos": 23218546913280.0,
      "grad_norm": 1.9506054010901464,
      "language_loss": 0.79008913,
      "learning_rate": 3.945455835935425e-06,
      "loss": 0.81282759,
      "num_input_tokens_seen": 18131520,
      "step": 851,
      "time_per_iteration": 2.7098145484924316
    },
    {
      "auxiliary_loss_clip": 0.01242898,
      "auxiliary_loss_mlp": 0.01053007,
      "balance_loss_clip": 1.07066822,
      "balance_loss_mlp": 1.04058576,
      "epoch": 0.1024469428245055,
      "flos": 22922929981440.0,
      "grad_norm": 4.454104236642058,
      "language_loss": 0.75220364,
      "learning_rate": 3.94527500658424e-06,
      "loss": 0.7751627,
      "num_input_tokens_seen": 18149185,
      "step": 852,
      "time_per_iteration": 3.6755588054656982
    },
    {
      "auxiliary_loss_clip": 0.01225528,
      "auxiliary_loss_mlp": 0.01044451,
      "balance_loss_clip": 1.07176566,
      "balance_loss_mlp": 1.03230953,
      "epoch": 0.10256718571514459,
      "flos": 31359495957120.0,
      "grad_norm": 2.187876290408179,
      "language_loss": 0.811149,
      "learning_rate": 3.945093882136382e-06,
      "loss": 0.83384877,
      "num_input_tokens_seen": 18172960,
      "step": 853,
      "time_per_iteration": 3.9813437461853027
    },
    {
      "auxiliary_loss_clip": 0.01221681,
      "auxiliary_loss_mlp": 0.01047789,
      "balance_loss_clip": 1.06854355,
      "balance_loss_mlp": 1.03584373,
      "epoch": 0.10268742860578368,
      "flos": 23474877344640.0,
      "grad_norm": 1.9332529944482044,
      "language_loss": 0.85030806,
      "learning_rate": 3.944912462619329e-06,
      "loss": 0.87300277,
      "num_input_tokens_seen": 18191925,
      "step": 854,
      "time_per_iteration": 2.7541000843048096
    },
    {
      "auxiliary_loss_clip": 0.0125121,
      "auxiliary_loss_mlp": 0.01042435,
      "balance_loss_clip": 1.07651901,
      "balance_loss_mlp": 1.02942955,
      "epoch": 0.10280767149642277,
      "flos": 25520313323520.0,
      "grad_norm": 1.96155161262425,
      "language_loss": 0.80841839,
      "learning_rate": 3.9447307480606025e-06,
      "loss": 0.83135486,
      "num_input_tokens_seen": 18212010,
      "step": 855,
      "time_per_iteration": 2.725005626678467
    },
    {
      "auxiliary_loss_clip": 0.01246237,
      "auxiliary_loss_mlp": 0.01047565,
      "balance_loss_clip": 1.06852245,
      "balance_loss_mlp": 1.03567374,
      "epoch": 0.10292791438706186,
      "flos": 17347691462400.0,
      "grad_norm": 2.09866025391388,
      "language_loss": 0.89898372,
      "learning_rate": 3.944548738487767e-06,
      "loss": 0.92192173,
      "num_input_tokens_seen": 18229525,
      "step": 856,
      "time_per_iteration": 2.722170352935791
    },
    {
      "auxiliary_loss_clip": 0.01217094,
      "auxiliary_loss_mlp": 0.01049902,
      "balance_loss_clip": 1.06628835,
      "balance_loss_mlp": 1.03770101,
      "epoch": 0.10304815727770096,
      "flos": 27052693390080.0,
      "grad_norm": 2.0879436342206965,
      "language_loss": 0.9105674,
      "learning_rate": 3.944366433928434e-06,
      "loss": 0.93323743,
      "num_input_tokens_seen": 18249505,
      "step": 857,
      "time_per_iteration": 2.797571897506714
    },
    {
      "auxiliary_loss_clip": 0.01224861,
      "auxiliary_loss_mlp": 0.01048792,
      "balance_loss_clip": 1.06922412,
      "balance_loss_mlp": 1.03794396,
      "epoch": 0.10316840016834004,
      "flos": 22782591544320.0,
      "grad_norm": 1.5735214610757602,
      "language_loss": 0.83598447,
      "learning_rate": 3.9441838344102594e-06,
      "loss": 0.85872108,
      "num_input_tokens_seen": 18269230,
      "step": 858,
      "time_per_iteration": 2.725511074066162
    },
    {
      "auxiliary_loss_clip": 0.01143707,
      "auxiliary_loss_mlp": 0.01056035,
      "balance_loss_clip": 1.05385387,
      "balance_loss_mlp": 1.04354191,
      "epoch": 0.10328864305897914,
      "flos": 20704584908160.0,
      "grad_norm": 3.4890578649118695,
      "language_loss": 0.67372888,
      "learning_rate": 3.944000939960943e-06,
      "loss": 0.69572628,
      "num_input_tokens_seen": 18287955,
      "step": 859,
      "time_per_iteration": 2.89247989654541
    },
    {
      "auxiliary_loss_clip": 0.01223143,
      "auxiliary_loss_mlp": 0.01038676,
      "balance_loss_clip": 1.06537127,
      "balance_loss_mlp": 1.02746987,
      "epoch": 0.10340888594961822,
      "flos": 28478814048000.0,
      "grad_norm": 1.624819972779203,
      "language_loss": 0.80081308,
      "learning_rate": 3.943817750608229e-06,
      "loss": 0.82343125,
      "num_input_tokens_seen": 18310505,
      "step": 860,
      "time_per_iteration": 2.795809268951416
    },
    {
      "auxiliary_loss_clip": 0.01245255,
      "auxiliary_loss_mlp": 0.01047901,
      "balance_loss_clip": 1.07367027,
      "balance_loss_mlp": 1.03629017,
      "epoch": 0.10352912884025732,
      "flos": 13370333460480.0,
      "grad_norm": 2.172326155965229,
      "language_loss": 0.81789625,
      "learning_rate": 3.943634266379908e-06,
      "loss": 0.84082782,
      "num_input_tokens_seen": 18327400,
      "step": 861,
      "time_per_iteration": 2.7070083618164062
    },
    {
      "auxiliary_loss_clip": 0.0125788,
      "auxiliary_loss_mlp": 0.01050681,
      "balance_loss_clip": 1.07410693,
      "balance_loss_mlp": 1.03802109,
      "epoch": 0.10364937173089642,
      "flos": 25558558329600.0,
      "grad_norm": 1.7956973859353607,
      "language_loss": 0.84702528,
      "learning_rate": 3.943450487303815e-06,
      "loss": 0.87011087,
      "num_input_tokens_seen": 18347895,
      "step": 862,
      "time_per_iteration": 2.7260282039642334
    },
    {
      "auxiliary_loss_clip": 0.01220571,
      "auxiliary_loss_mlp": 0.00765415,
      "balance_loss_clip": 1.06415856,
      "balance_loss_mlp": 1.00061953,
      "epoch": 0.1037696146215355,
      "flos": 21215486004480.0,
      "grad_norm": 2.446308578625894,
      "language_loss": 0.85250175,
      "learning_rate": 3.943266413407827e-06,
      "loss": 0.87236154,
      "num_input_tokens_seen": 18367170,
      "step": 863,
      "time_per_iteration": 2.697739839553833
    },
    {
      "auxiliary_loss_clip": 0.01230838,
      "auxiliary_loss_mlp": 0.00765608,
      "balance_loss_clip": 1.06899261,
      "balance_loss_mlp": 1.00075722,
      "epoch": 0.1038898575121746,
      "flos": 25807382818560.0,
      "grad_norm": 2.0873362911701716,
      "language_loss": 0.84990215,
      "learning_rate": 3.94308204471987e-06,
      "loss": 0.86986655,
      "num_input_tokens_seen": 18386185,
      "step": 864,
      "time_per_iteration": 2.7868831157684326
    },
    {
      "auxiliary_loss_clip": 0.01148288,
      "auxiliary_loss_mlp": 0.01037902,
      "balance_loss_clip": 1.05796742,
      "balance_loss_mlp": 1.02781045,
      "epoch": 0.10401010040281368,
      "flos": 19062425900160.0,
      "grad_norm": 7.774643149104241,
      "language_loss": 0.74097669,
      "learning_rate": 3.942897381267912e-06,
      "loss": 0.7628386,
      "num_input_tokens_seen": 18402550,
      "step": 865,
      "time_per_iteration": 2.8122334480285645
    },
    {
      "auxiliary_loss_clip": 0.01233048,
      "auxiliary_loss_mlp": 0.0104391,
      "balance_loss_clip": 1.06869555,
      "balance_loss_mlp": 1.03066015,
      "epoch": 0.10413034329345278,
      "flos": 16355119962240.0,
      "grad_norm": 2.8015805193873287,
      "language_loss": 0.66149348,
      "learning_rate": 3.942712423079965e-06,
      "loss": 0.68426305,
      "num_input_tokens_seen": 18418940,
      "step": 866,
      "time_per_iteration": 2.786125659942627
    },
    {
      "auxiliary_loss_clip": 0.01240066,
      "auxiliary_loss_mlp": 0.01044226,
      "balance_loss_clip": 1.06884265,
      "balance_loss_mlp": 1.03296041,
      "epoch": 0.10425058618409186,
      "flos": 17236511890560.0,
      "grad_norm": 2.590200459888096,
      "language_loss": 0.9044047,
      "learning_rate": 3.942527170184088e-06,
      "loss": 0.92724764,
      "num_input_tokens_seen": 18435560,
      "step": 867,
      "time_per_iteration": 2.662936210632324
    },
    {
      "auxiliary_loss_clip": 0.01225096,
      "auxiliary_loss_mlp": 0.01042371,
      "balance_loss_clip": 1.07222927,
      "balance_loss_mlp": 1.03117776,
      "epoch": 0.10437082907473096,
      "flos": 17967365919360.0,
      "grad_norm": 2.5771522540311667,
      "language_loss": 0.77760422,
      "learning_rate": 3.942341622608385e-06,
      "loss": 0.8002789,
      "num_input_tokens_seen": 18452590,
      "step": 868,
      "time_per_iteration": 2.741323471069336
    },
    {
      "auxiliary_loss_clip": 0.01230421,
      "auxiliary_loss_mlp": 0.01042619,
      "balance_loss_clip": 1.07396555,
      "balance_loss_mlp": 1.03037,
      "epoch": 0.10449107196537005,
      "flos": 36283315374720.0,
      "grad_norm": 1.5664195126972278,
      "language_loss": 0.77840608,
      "learning_rate": 3.942155780381001e-06,
      "loss": 0.80113649,
      "num_input_tokens_seen": 18476325,
      "step": 869,
      "time_per_iteration": 2.832160472869873
    },
    {
      "auxiliary_loss_clip": 0.01215975,
      "auxiliary_loss_mlp": 0.01049277,
      "balance_loss_clip": 1.06293535,
      "balance_loss_mlp": 1.03696847,
      "epoch": 0.10461131485600914,
      "flos": 23802095266560.0,
      "grad_norm": 1.980219129283383,
      "language_loss": 0.75820267,
      "learning_rate": 3.94196964353013e-06,
      "loss": 0.78085524,
      "num_input_tokens_seen": 18495775,
      "step": 870,
      "time_per_iteration": 2.7036705017089844
    },
    {
      "auxiliary_loss_clip": 0.01193203,
      "auxiliary_loss_mlp": 0.01046793,
      "balance_loss_clip": 1.06796026,
      "balance_loss_mlp": 1.03510499,
      "epoch": 0.10473155774664823,
      "flos": 18405476104320.0,
      "grad_norm": 4.205722713511854,
      "language_loss": 0.80559772,
      "learning_rate": 3.941783212084008e-06,
      "loss": 0.82799768,
      "num_input_tokens_seen": 18513530,
      "step": 871,
      "time_per_iteration": 2.774487018585205
    },
    {
      "auxiliary_loss_clip": 0.01212546,
      "auxiliary_loss_mlp": 0.01046103,
      "balance_loss_clip": 1.06938505,
      "balance_loss_mlp": 1.03414655,
      "epoch": 0.10485180063728732,
      "flos": 25592637358080.0,
      "grad_norm": 2.8966814700196597,
      "language_loss": 0.79033911,
      "learning_rate": 3.941596486070916e-06,
      "loss": 0.81292564,
      "num_input_tokens_seen": 18531575,
      "step": 872,
      "time_per_iteration": 2.756108045578003
    },
    {
      "auxiliary_loss_clip": 0.01198882,
      "auxiliary_loss_mlp": 0.01040699,
      "balance_loss_clip": 1.06165218,
      "balance_loss_mlp": 1.02820575,
      "epoch": 0.10497204352792641,
      "flos": 27088747666560.0,
      "grad_norm": 2.1353659933403337,
      "language_loss": 0.58778763,
      "learning_rate": 3.941409465519182e-06,
      "loss": 0.61018348,
      "num_input_tokens_seen": 18552100,
      "step": 873,
      "time_per_iteration": 2.77939772605896
    },
    {
      "auxiliary_loss_clip": 0.01208755,
      "auxiliary_loss_mlp": 0.01041727,
      "balance_loss_clip": 1.07091141,
      "balance_loss_mlp": 1.02996063,
      "epoch": 0.10509228641856551,
      "flos": 32858479353600.0,
      "grad_norm": 1.6973363125182745,
      "language_loss": 0.85408235,
      "learning_rate": 3.941222150457176e-06,
      "loss": 0.87658721,
      "num_input_tokens_seen": 18575355,
      "step": 874,
      "time_per_iteration": 3.799955368041992
    },
    {
      "auxiliary_loss_clip": 0.01213355,
      "auxiliary_loss_mlp": 0.01048677,
      "balance_loss_clip": 1.06531823,
      "balance_loss_mlp": 1.03678584,
      "epoch": 0.10521252930920459,
      "flos": 14319165173760.0,
      "grad_norm": 3.243177244725585,
      "language_loss": 0.72177267,
      "learning_rate": 3.941034540913311e-06,
      "loss": 0.74439299,
      "num_input_tokens_seen": 18592885,
      "step": 875,
      "time_per_iteration": 3.5889127254486084
    },
    {
      "auxiliary_loss_clip": 0.01239927,
      "auxiliary_loss_mlp": 0.01044609,
      "balance_loss_clip": 1.07052708,
      "balance_loss_mlp": 1.03214025,
      "epoch": 0.10533277219984369,
      "flos": 21687028773120.0,
      "grad_norm": 1.7547741823959135,
      "language_loss": 0.82303417,
      "learning_rate": 3.940846636916051e-06,
      "loss": 0.84587955,
      "num_input_tokens_seen": 18612920,
      "step": 876,
      "time_per_iteration": 2.688748836517334
    },
    {
      "auxiliary_loss_clip": 0.0121825,
      "auxiliary_loss_mlp": 0.01040561,
      "balance_loss_clip": 1.06766868,
      "balance_loss_mlp": 1.02805042,
      "epoch": 0.10545301509048277,
      "flos": 22269787027200.0,
      "grad_norm": 3.2113967902687617,
      "language_loss": 0.86637908,
      "learning_rate": 3.940658438493899e-06,
      "loss": 0.88896716,
      "num_input_tokens_seen": 18630765,
      "step": 877,
      "time_per_iteration": 2.6976208686828613
    },
    {
      "auxiliary_loss_clip": 0.01177188,
      "auxiliary_loss_mlp": 0.00765956,
      "balance_loss_clip": 1.05699432,
      "balance_loss_mlp": 1.00052929,
      "epoch": 0.10557325798112187,
      "flos": 22199725549440.0,
      "grad_norm": 2.4311414984417334,
      "language_loss": 0.76123184,
      "learning_rate": 3.940469945675405e-06,
      "loss": 0.78066331,
      "num_input_tokens_seen": 18649150,
      "step": 878,
      "time_per_iteration": 3.8000311851501465
    },
    {
      "auxiliary_loss_clip": 0.01200803,
      "auxiliary_loss_mlp": 0.01048015,
      "balance_loss_clip": 1.06233609,
      "balance_loss_mlp": 1.0362668,
      "epoch": 0.10569350087176095,
      "flos": 25775889569280.0,
      "grad_norm": 1.9613903065472886,
      "language_loss": 0.91339862,
      "learning_rate": 3.940281158489163e-06,
      "loss": 0.93588686,
      "num_input_tokens_seen": 18668380,
      "step": 879,
      "time_per_iteration": 3.7066617012023926
    },
    {
      "auxiliary_loss_clip": 0.01185876,
      "auxiliary_loss_mlp": 0.01047813,
      "balance_loss_clip": 1.06135464,
      "balance_loss_mlp": 1.03607118,
      "epoch": 0.10581374376240005,
      "flos": 17311385790720.0,
      "grad_norm": 2.2549265650290446,
      "language_loss": 0.82938284,
      "learning_rate": 3.940092076963812e-06,
      "loss": 0.85171968,
      "num_input_tokens_seen": 18685875,
      "step": 880,
      "time_per_iteration": 2.7050318717956543
    },
    {
      "auxiliary_loss_clip": 0.01211797,
      "auxiliary_loss_mlp": 0.01045184,
      "balance_loss_clip": 1.07101321,
      "balance_loss_mlp": 1.03348374,
      "epoch": 0.10593398665303914,
      "flos": 34349454017280.0,
      "grad_norm": 2.1680455189707213,
      "language_loss": 0.79074049,
      "learning_rate": 3.9399027011280355e-06,
      "loss": 0.81331033,
      "num_input_tokens_seen": 18707970,
      "step": 881,
      "time_per_iteration": 2.8841278553009033
    },
    {
      "auxiliary_loss_clip": 0.01218324,
      "auxiliary_loss_mlp": 0.01056399,
      "balance_loss_clip": 1.0703721,
      "balance_loss_mlp": 1.04169464,
      "epoch": 0.10605422954367823,
      "flos": 23257977068160.0,
      "grad_norm": 2.491882299878421,
      "language_loss": 0.77403337,
      "learning_rate": 3.939713031010561e-06,
      "loss": 0.79678065,
      "num_input_tokens_seen": 18726335,
      "step": 882,
      "time_per_iteration": 2.95074200630188
    },
    {
      "auxiliary_loss_clip": 0.01244054,
      "auxiliary_loss_mlp": 0.01047427,
      "balance_loss_clip": 1.07333827,
      "balance_loss_mlp": 1.03458786,
      "epoch": 0.10617447243431732,
      "flos": 22820118278400.0,
      "grad_norm": 2.1203017078522644,
      "language_loss": 0.77464867,
      "learning_rate": 3.939523066640163e-06,
      "loss": 0.79756355,
      "num_input_tokens_seen": 18745230,
      "step": 883,
      "time_per_iteration": 2.645681619644165
    },
    {
      "auxiliary_loss_clip": 0.01197002,
      "auxiliary_loss_mlp": 0.01050865,
      "balance_loss_clip": 1.06129885,
      "balance_loss_mlp": 1.03867614,
      "epoch": 0.10629471532495641,
      "flos": 24386577373440.0,
      "grad_norm": 1.8022864163434267,
      "language_loss": 0.81020081,
      "learning_rate": 3.939332808045657e-06,
      "loss": 0.83267951,
      "num_input_tokens_seen": 18764880,
      "step": 884,
      "time_per_iteration": 2.774406671524048
    },
    {
      "auxiliary_loss_clip": 0.01195764,
      "auxiliary_loss_mlp": 0.01038465,
      "balance_loss_clip": 1.06456685,
      "balance_loss_mlp": 1.02722931,
      "epoch": 0.1064149582155955,
      "flos": 21105491581440.0,
      "grad_norm": 1.6146053357558408,
      "language_loss": 0.84520656,
      "learning_rate": 3.939142255255906e-06,
      "loss": 0.86754882,
      "num_input_tokens_seen": 18785765,
      "step": 885,
      "time_per_iteration": 2.780047655105591
    },
    {
      "auxiliary_loss_clip": 0.01221505,
      "auxiliary_loss_mlp": 0.01053507,
      "balance_loss_clip": 1.06530988,
      "balance_loss_mlp": 1.04169369,
      "epoch": 0.1065352011062346,
      "flos": 20702035042560.0,
      "grad_norm": 1.946375466685272,
      "language_loss": 0.86810756,
      "learning_rate": 3.938951408299817e-06,
      "loss": 0.8908577,
      "num_input_tokens_seen": 18804605,
      "step": 886,
      "time_per_iteration": 2.715280771255493
    },
    {
      "auxiliary_loss_clip": 0.01153964,
      "auxiliary_loss_mlp": 0.01008045,
      "balance_loss_clip": 1.0469327,
      "balance_loss_mlp": 1.00446916,
      "epoch": 0.10665544399687368,
      "flos": 62659632689280.0,
      "grad_norm": 0.8004446693263972,
      "language_loss": 0.54414856,
      "learning_rate": 3.938760267206342e-06,
      "loss": 0.5657686,
      "num_input_tokens_seen": 18866425,
      "step": 887,
      "time_per_iteration": 3.1733853816986084
    },
    {
      "auxiliary_loss_clip": 0.01172127,
      "auxiliary_loss_mlp": 0.00765595,
      "balance_loss_clip": 1.06062341,
      "balance_loss_mlp": 1.00053799,
      "epoch": 0.10677568688751278,
      "flos": 26140382830080.0,
      "grad_norm": 4.386867218378815,
      "language_loss": 0.78376812,
      "learning_rate": 3.938568832004475e-06,
      "loss": 0.80314535,
      "num_input_tokens_seen": 18885130,
      "step": 888,
      "time_per_iteration": 2.852372884750366
    },
    {
      "auxiliary_loss_clip": 0.01221933,
      "auxiliary_loss_mlp": 0.01041287,
      "balance_loss_clip": 1.06930852,
      "balance_loss_mlp": 1.02894294,
      "epoch": 0.10689592977815186,
      "flos": 12786533712000.0,
      "grad_norm": 2.136954029608697,
      "language_loss": 0.75704992,
      "learning_rate": 3.938377102723257e-06,
      "loss": 0.77968204,
      "num_input_tokens_seen": 18902265,
      "step": 889,
      "time_per_iteration": 2.7155089378356934
    },
    {
      "auxiliary_loss_clip": 0.01145689,
      "auxiliary_loss_mlp": 0.0076587,
      "balance_loss_clip": 1.05125284,
      "balance_loss_mlp": 1.00048959,
      "epoch": 0.10701617266879096,
      "flos": 22126683242880.0,
      "grad_norm": 2.1002473246470825,
      "language_loss": 0.83605719,
      "learning_rate": 3.938185079391774e-06,
      "loss": 0.85517281,
      "num_input_tokens_seen": 18919310,
      "step": 890,
      "time_per_iteration": 2.965235471725464
    },
    {
      "auxiliary_loss_clip": 0.01170852,
      "auxiliary_loss_mlp": 0.01045406,
      "balance_loss_clip": 1.05985188,
      "balance_loss_mlp": 1.03368759,
      "epoch": 0.10713641555943004,
      "flos": 19745625559680.0,
      "grad_norm": 2.795157466886573,
      "language_loss": 1.05622292,
      "learning_rate": 3.937992762039157e-06,
      "loss": 1.07838559,
      "num_input_tokens_seen": 18932635,
      "step": 891,
      "time_per_iteration": 2.8045248985290527
    },
    {
      "auxiliary_loss_clip": 0.01239262,
      "auxiliary_loss_mlp": 0.01045149,
      "balance_loss_clip": 1.07188451,
      "balance_loss_mlp": 1.03283417,
      "epoch": 0.10725665845006914,
      "flos": 23952992302080.0,
      "grad_norm": 1.7565377373226563,
      "language_loss": 0.80383217,
      "learning_rate": 3.937800150694577e-06,
      "loss": 0.82667631,
      "num_input_tokens_seen": 18953810,
      "step": 892,
      "time_per_iteration": 2.7386362552642822
    },
    {
      "auxiliary_loss_clip": 0.01185214,
      "auxiliary_loss_mlp": 0.01041778,
      "balance_loss_clip": 1.05846262,
      "balance_loss_mlp": 1.03003025,
      "epoch": 0.10737690134070824,
      "flos": 18551704371840.0,
      "grad_norm": 2.1901416952340633,
      "language_loss": 0.75955498,
      "learning_rate": 3.937607245387255e-06,
      "loss": 0.78182489,
      "num_input_tokens_seen": 18973175,
      "step": 893,
      "time_per_iteration": 2.8051400184631348
    },
    {
      "auxiliary_loss_clip": 0.01226592,
      "auxiliary_loss_mlp": 0.01045963,
      "balance_loss_clip": 1.07037771,
      "balance_loss_mlp": 1.03451335,
      "epoch": 0.10749714423134732,
      "flos": 22707609903360.0,
      "grad_norm": 2.0137335238928653,
      "language_loss": 0.72304648,
      "learning_rate": 3.937414046146455e-06,
      "loss": 0.74577206,
      "num_input_tokens_seen": 18991130,
      "step": 894,
      "time_per_iteration": 2.7720541954040527
    },
    {
      "auxiliary_loss_clip": 0.01211204,
      "auxiliary_loss_mlp": 0.01048329,
      "balance_loss_clip": 1.06544447,
      "balance_loss_mlp": 1.03607988,
      "epoch": 0.10761738712198642,
      "flos": 21106066199040.0,
      "grad_norm": 3.657428379893261,
      "language_loss": 0.75211942,
      "learning_rate": 3.9372205530014845e-06,
      "loss": 0.77471477,
      "num_input_tokens_seen": 19009610,
      "step": 895,
      "time_per_iteration": 2.743738889694214
    },
    {
      "auxiliary_loss_clip": 0.01238589,
      "auxiliary_loss_mlp": 0.0105192,
      "balance_loss_clip": 1.0681237,
      "balance_loss_mlp": 1.03931952,
      "epoch": 0.1077376300126255,
      "flos": 23766723348480.0,
      "grad_norm": 3.410804821703451,
      "language_loss": 0.71377927,
      "learning_rate": 3.937026765981696e-06,
      "loss": 0.73668438,
      "num_input_tokens_seen": 19029680,
      "step": 896,
      "time_per_iteration": 2.666814088821411
    },
    {
      "auxiliary_loss_clip": 0.01233717,
      "auxiliary_loss_mlp": 0.01058758,
      "balance_loss_clip": 1.06918716,
      "balance_loss_mlp": 1.04658628,
      "epoch": 0.1078578729032646,
      "flos": 20919581763840.0,
      "grad_norm": 2.779516683861264,
      "language_loss": 0.7961899,
      "learning_rate": 3.936832685116488e-06,
      "loss": 0.81911469,
      "num_input_tokens_seen": 19047775,
      "step": 897,
      "time_per_iteration": 2.710240602493286
    },
    {
      "auxiliary_loss_clip": 0.01205306,
      "auxiliary_loss_mlp": 0.01050046,
      "balance_loss_clip": 1.06353998,
      "balance_loss_mlp": 1.03755331,
      "epoch": 0.10797811579390369,
      "flos": 14829886702080.0,
      "grad_norm": 2.3778587705399636,
      "language_loss": 0.9008944,
      "learning_rate": 3.936638310435301e-06,
      "loss": 0.92344797,
      "num_input_tokens_seen": 19065640,
      "step": 898,
      "time_per_iteration": 2.682734727859497
    },
    {
      "auxiliary_loss_clip": 0.01222452,
      "auxiliary_loss_mlp": 0.01049512,
      "balance_loss_clip": 1.06863737,
      "balance_loss_mlp": 1.03793716,
      "epoch": 0.10809835868454278,
      "flos": 19536985411200.0,
      "grad_norm": 2.0689662441423184,
      "language_loss": 0.81583524,
      "learning_rate": 3.936443641967623e-06,
      "loss": 0.83855486,
      "num_input_tokens_seen": 19084470,
      "step": 899,
      "time_per_iteration": 2.6997082233428955
    },
    {
      "auxiliary_loss_clip": 0.01204474,
      "auxiliary_loss_mlp": 0.01037493,
      "balance_loss_clip": 1.06530786,
      "balance_loss_mlp": 1.02530396,
      "epoch": 0.10821860157518187,
      "flos": 18442320480000.0,
      "grad_norm": 1.9696155302057166,
      "language_loss": 0.83004439,
      "learning_rate": 3.936248679742983e-06,
      "loss": 0.85246408,
      "num_input_tokens_seen": 19102965,
      "step": 900,
      "time_per_iteration": 3.66463565826416
    },
    {
      "auxiliary_loss_clip": 0.01118937,
      "auxiliary_loss_mlp": 0.01003121,
      "balance_loss_clip": 1.04024267,
      "balance_loss_mlp": 0.99928278,
      "epoch": 0.10833884446582095,
      "flos": 49359468447360.0,
      "grad_norm": 1.0521757518143833,
      "language_loss": 0.70134997,
      "learning_rate": 3.936053423790959e-06,
      "loss": 0.7225706,
      "num_input_tokens_seen": 19151285,
      "step": 901,
      "time_per_iteration": 4.071150541305542
    },
    {
      "auxiliary_loss_clip": 0.0125428,
      "auxiliary_loss_mlp": 0.01047296,
      "balance_loss_clip": 1.07384813,
      "balance_loss_mlp": 1.03494048,
      "epoch": 0.10845908735646005,
      "flos": 20411912891520.0,
      "grad_norm": 1.8940418875941623,
      "language_loss": 0.77346891,
      "learning_rate": 3.935857874141168e-06,
      "loss": 0.79648471,
      "num_input_tokens_seen": 19170120,
      "step": 902,
      "time_per_iteration": 2.6661529541015625
    },
    {
      "auxiliary_loss_clip": 0.0116052,
      "auxiliary_loss_mlp": 0.01047958,
      "balance_loss_clip": 1.05935311,
      "balance_loss_mlp": 1.03530931,
      "epoch": 0.10857933024709913,
      "flos": 14027750133120.0,
      "grad_norm": 2.3546783023210294,
      "language_loss": 0.83238184,
      "learning_rate": 3.935662030823279e-06,
      "loss": 0.85446656,
      "num_input_tokens_seen": 19186305,
      "step": 903,
      "time_per_iteration": 3.7549631595611572
    },
    {
      "auxiliary_loss_clip": 0.01239183,
      "auxiliary_loss_mlp": 0.01056633,
      "balance_loss_clip": 1.07029748,
      "balance_loss_mlp": 1.04456306,
      "epoch": 0.10869957313773823,
      "flos": 13369004657280.0,
      "grad_norm": 3.002867943874866,
      "language_loss": 0.72352022,
      "learning_rate": 3.935465893866998e-06,
      "loss": 0.74647838,
      "num_input_tokens_seen": 19204530,
      "step": 904,
      "time_per_iteration": 2.746872901916504
    },
    {
      "auxiliary_loss_clip": 0.01244683,
      "auxiliary_loss_mlp": 0.01051997,
      "balance_loss_clip": 1.07266843,
      "balance_loss_mlp": 1.04004073,
      "epoch": 0.10881981602837733,
      "flos": 25807095509760.0,
      "grad_norm": 6.511467503176962,
      "language_loss": 0.80427599,
      "learning_rate": 3.935269463302079e-06,
      "loss": 0.82724279,
      "num_input_tokens_seen": 19222735,
      "step": 905,
      "time_per_iteration": 3.6980438232421875
    },
    {
      "auxiliary_loss_clip": 0.01143786,
      "auxiliary_loss_mlp": 0.00766304,
      "balance_loss_clip": 1.05227637,
      "balance_loss_mlp": 1.00034869,
      "epoch": 0.10894005891901641,
      "flos": 20777555387520.0,
      "grad_norm": 1.6606421806446594,
      "language_loss": 0.76303172,
      "learning_rate": 3.935072739158322e-06,
      "loss": 0.78213263,
      "num_input_tokens_seen": 19242445,
      "step": 906,
      "time_per_iteration": 2.9058079719543457
    },
    {
      "auxiliary_loss_clip": 0.012273,
      "auxiliary_loss_mlp": 0.01049546,
      "balance_loss_clip": 1.07339692,
      "balance_loss_mlp": 1.03780437,
      "epoch": 0.10906030180965551,
      "flos": 26649883296000.0,
      "grad_norm": 1.6488727697683407,
      "language_loss": 0.79651308,
      "learning_rate": 3.934875721465569e-06,
      "loss": 0.81928158,
      "num_input_tokens_seen": 19262865,
      "step": 907,
      "time_per_iteration": 2.7432498931884766
    },
    {
      "auxiliary_loss_clip": 0.01226813,
      "auxiliary_loss_mlp": 0.01051934,
      "balance_loss_clip": 1.06999397,
      "balance_loss_mlp": 1.03852344,
      "epoch": 0.10918054470029459,
      "flos": 36534402420480.0,
      "grad_norm": 2.394385987399249,
      "language_loss": 0.71600878,
      "learning_rate": 3.9346784102537076e-06,
      "loss": 0.73879623,
      "num_input_tokens_seen": 19285000,
      "step": 908,
      "time_per_iteration": 2.8891232013702393
    },
    {
      "auxiliary_loss_clip": 0.01255825,
      "auxiliary_loss_mlp": 0.01043592,
      "balance_loss_clip": 1.07310009,
      "balance_loss_mlp": 1.03174305,
      "epoch": 0.10930078759093369,
      "flos": 21762549118080.0,
      "grad_norm": 6.139199839834398,
      "language_loss": 0.78187883,
      "learning_rate": 3.934480805552669e-06,
      "loss": 0.80487305,
      "num_input_tokens_seen": 19306010,
      "step": 909,
      "time_per_iteration": 2.630566358566284
    },
    {
      "auxiliary_loss_clip": 0.01220996,
      "auxiliary_loss_mlp": 0.01049322,
      "balance_loss_clip": 1.06804514,
      "balance_loss_mlp": 1.03645372,
      "epoch": 0.10942103048157277,
      "flos": 22601781457920.0,
      "grad_norm": 1.88863874902926,
      "language_loss": 0.88176477,
      "learning_rate": 3.93428290739243e-06,
      "loss": 0.90446794,
      "num_input_tokens_seen": 19325380,
      "step": 910,
      "time_per_iteration": 2.717771053314209
    },
    {
      "auxiliary_loss_clip": 0.01238015,
      "auxiliary_loss_mlp": 0.0104273,
      "balance_loss_clip": 1.06983447,
      "balance_loss_mlp": 1.02931285,
      "epoch": 0.10954127337221187,
      "flos": 15045781397760.0,
      "grad_norm": 2.3306990357216812,
      "language_loss": 0.7996859,
      "learning_rate": 3.9340847158030125e-06,
      "loss": 0.82249331,
      "num_input_tokens_seen": 19338960,
      "step": 911,
      "time_per_iteration": 2.6318094730377197
    },
    {
      "auxiliary_loss_clip": 0.01188243,
      "auxiliary_loss_mlp": 0.01054139,
      "balance_loss_clip": 1.06275403,
      "balance_loss_mlp": 1.04148531,
      "epoch": 0.10966151626285096,
      "flos": 21650974496640.0,
      "grad_norm": 1.8462808782069946,
      "language_loss": 0.75496262,
      "learning_rate": 3.9338862308144814e-06,
      "loss": 0.77738643,
      "num_input_tokens_seen": 19357780,
      "step": 912,
      "time_per_iteration": 2.7844619750976562
    },
    {
      "auxiliary_loss_clip": 0.01243439,
      "auxiliary_loss_mlp": 0.0104088,
      "balance_loss_clip": 1.07077479,
      "balance_loss_mlp": 1.02914405,
      "epoch": 0.10978175915349005,
      "flos": 20121359777280.0,
      "grad_norm": 1.7374409868976453,
      "language_loss": 0.84395826,
      "learning_rate": 3.933687452456946e-06,
      "loss": 0.8668015,
      "num_input_tokens_seen": 19377680,
      "step": 913,
      "time_per_iteration": 2.6723854541778564
    },
    {
      "auxiliary_loss_clip": 0.0124424,
      "auxiliary_loss_mlp": 0.01042504,
      "balance_loss_clip": 1.06862009,
      "balance_loss_mlp": 1.02899194,
      "epoch": 0.10990200204412914,
      "flos": 20412667077120.0,
      "grad_norm": 2.3754829473085413,
      "language_loss": 0.86144179,
      "learning_rate": 3.933488380760562e-06,
      "loss": 0.88430923,
      "num_input_tokens_seen": 19397040,
      "step": 914,
      "time_per_iteration": 2.6687798500061035
    },
    {
      "auxiliary_loss_clip": 0.01224579,
      "auxiliary_loss_mlp": 0.00766027,
      "balance_loss_clip": 1.06557488,
      "balance_loss_mlp": 1.00042224,
      "epoch": 0.11002224493476823,
      "flos": 17530117660800.0,
      "grad_norm": 2.046685363704294,
      "language_loss": 0.87693053,
      "learning_rate": 3.9332890157555286e-06,
      "loss": 0.89683664,
      "num_input_tokens_seen": 19413975,
      "step": 915,
      "time_per_iteration": 2.711285352706909
    },
    {
      "auxiliary_loss_clip": 0.01203576,
      "auxiliary_loss_mlp": 0.01050514,
      "balance_loss_clip": 1.06346035,
      "balance_loss_mlp": 1.03809285,
      "epoch": 0.11014248782540732,
      "flos": 12203093099520.0,
      "grad_norm": 3.920998767638207,
      "language_loss": 0.76522577,
      "learning_rate": 3.933089357472088e-06,
      "loss": 0.7877667,
      "num_input_tokens_seen": 19432005,
      "step": 916,
      "time_per_iteration": 2.757885456085205
    },
    {
      "auxiliary_loss_clip": 0.01205266,
      "auxiliary_loss_mlp": 0.01043573,
      "balance_loss_clip": 1.06572628,
      "balance_loss_mlp": 1.03115129,
      "epoch": 0.11026273071604642,
      "flos": 22382977760640.0,
      "grad_norm": 1.8011933653247063,
      "language_loss": 0.86145914,
      "learning_rate": 3.932889405940529e-06,
      "loss": 0.88394749,
      "num_input_tokens_seen": 19450100,
      "step": 917,
      "time_per_iteration": 2.730933666229248
    },
    {
      "auxiliary_loss_clip": 0.01227938,
      "auxiliary_loss_mlp": 0.01058609,
      "balance_loss_clip": 1.0723474,
      "balance_loss_mlp": 1.0462414,
      "epoch": 0.1103829736066855,
      "flos": 19829046896640.0,
      "grad_norm": 2.172179116020284,
      "language_loss": 0.79929709,
      "learning_rate": 3.932689161191184e-06,
      "loss": 0.82216257,
      "num_input_tokens_seen": 19467805,
      "step": 918,
      "time_per_iteration": 2.72609543800354
    },
    {
      "auxiliary_loss_clip": 0.01175324,
      "auxiliary_loss_mlp": 0.01045197,
      "balance_loss_clip": 1.06105947,
      "balance_loss_mlp": 1.03163064,
      "epoch": 0.1105032164973246,
      "flos": 22669616292480.0,
      "grad_norm": 8.433804080948628,
      "language_loss": 0.88046771,
      "learning_rate": 3.93248862325443e-06,
      "loss": 0.90267289,
      "num_input_tokens_seen": 19486710,
      "step": 919,
      "time_per_iteration": 2.8022191524505615
    },
    {
      "auxiliary_loss_clip": 0.01122911,
      "auxiliary_loss_mlp": 0.01004512,
      "balance_loss_clip": 1.04069793,
      "balance_loss_mlp": 1.00091219,
      "epoch": 0.11062345938796368,
      "flos": 66483507876480.0,
      "grad_norm": 0.9367635853820407,
      "language_loss": 0.64484131,
      "learning_rate": 3.932287792160688e-06,
      "loss": 0.66611552,
      "num_input_tokens_seen": 19545170,
      "step": 920,
      "time_per_iteration": 3.2067248821258545
    },
    {
      "auxiliary_loss_clip": 0.01194449,
      "auxiliary_loss_mlp": 0.01046493,
      "balance_loss_clip": 1.06009722,
      "balance_loss_mlp": 1.03376734,
      "epoch": 0.11074370227860278,
      "flos": 21907771804800.0,
      "grad_norm": 2.3757620964077786,
      "language_loss": 0.80513364,
      "learning_rate": 3.932086667940424e-06,
      "loss": 0.82754302,
      "num_input_tokens_seen": 19561875,
      "step": 921,
      "time_per_iteration": 2.755520820617676
    },
    {
      "auxiliary_loss_clip": 0.01218565,
      "auxiliary_loss_mlp": 0.00765328,
      "balance_loss_clip": 1.06664264,
      "balance_loss_mlp": 1.00045967,
      "epoch": 0.11086394516924186,
      "flos": 28658115763200.0,
      "grad_norm": 2.4059742779854068,
      "language_loss": 0.8221159,
      "learning_rate": 3.93188525062415e-06,
      "loss": 0.84195483,
      "num_input_tokens_seen": 19582340,
      "step": 922,
      "time_per_iteration": 2.7427122592926025
    },
    {
      "auxiliary_loss_clip": 0.01242106,
      "auxiliary_loss_mlp": 0.01047618,
      "balance_loss_clip": 1.07081604,
      "balance_loss_mlp": 1.03403974,
      "epoch": 0.11098418805988096,
      "flos": 24535247765760.0,
      "grad_norm": 2.1617329698730168,
      "language_loss": 0.86134815,
      "learning_rate": 3.931683540242418e-06,
      "loss": 0.88424534,
      "num_input_tokens_seen": 19603405,
      "step": 923,
      "time_per_iteration": 2.7115285396575928
    },
    {
      "auxiliary_loss_clip": 0.01242339,
      "auxiliary_loss_mlp": 0.01051466,
      "balance_loss_clip": 1.06979072,
      "balance_loss_mlp": 1.03965223,
      "epoch": 0.11110443095052006,
      "flos": 22960384888320.0,
      "grad_norm": 2.793820558559009,
      "language_loss": 0.9104079,
      "learning_rate": 3.9314815368258295e-06,
      "loss": 0.93334591,
      "num_input_tokens_seen": 19619885,
      "step": 924,
      "time_per_iteration": 2.742687463760376
    },
    {
      "auxiliary_loss_clip": 0.01262719,
      "auxiliary_loss_mlp": 0.01045999,
      "balance_loss_clip": 1.07887709,
      "balance_loss_mlp": 1.03496647,
      "epoch": 0.11122467384115914,
      "flos": 18950025265920.0,
      "grad_norm": 2.274341501043423,
      "language_loss": 0.79011446,
      "learning_rate": 3.9312792404050275e-06,
      "loss": 0.81320167,
      "num_input_tokens_seen": 19637940,
      "step": 925,
      "time_per_iteration": 2.645670175552368
    },
    {
      "auxiliary_loss_clip": 0.01198874,
      "auxiliary_loss_mlp": 0.01039915,
      "balance_loss_clip": 1.06065273,
      "balance_loss_mlp": 1.02844143,
      "epoch": 0.11134491673179824,
      "flos": 25082957324160.0,
      "grad_norm": 1.898885691241595,
      "language_loss": 0.77415854,
      "learning_rate": 3.9310766510107e-06,
      "loss": 0.7965464,
      "num_input_tokens_seen": 19657115,
      "step": 926,
      "time_per_iteration": 3.6831414699554443
    },
    {
      "auxiliary_loss_clip": 0.01244796,
      "auxiliary_loss_mlp": 0.01048523,
      "balance_loss_clip": 1.07027173,
      "balance_loss_mlp": 1.03502285,
      "epoch": 0.11146515962243732,
      "flos": 24499121662080.0,
      "grad_norm": 2.0678920099026827,
      "language_loss": 0.92399329,
      "learning_rate": 3.9308737686735806e-06,
      "loss": 0.94692647,
      "num_input_tokens_seen": 19677075,
      "step": 927,
      "time_per_iteration": 3.6951520442962646
    },
    {
      "auxiliary_loss_clip": 0.01198815,
      "auxiliary_loss_mlp": 0.01052819,
      "balance_loss_clip": 1.0692327,
      "balance_loss_mlp": 1.04017675,
      "epoch": 0.11158540251307641,
      "flos": 22343763087360.0,
      "grad_norm": 2.1792659284439844,
      "language_loss": 0.82690579,
      "learning_rate": 3.9306705934244455e-06,
      "loss": 0.8494221,
      "num_input_tokens_seen": 19697155,
      "step": 928,
      "time_per_iteration": 2.7737433910369873
    },
    {
      "auxiliary_loss_clip": 0.01221759,
      "auxiliary_loss_mlp": 0.01049753,
      "balance_loss_clip": 1.07210016,
      "balance_loss_mlp": 1.03858304,
      "epoch": 0.11170564540371551,
      "flos": 19902304684800.0,
      "grad_norm": 1.6715249325264696,
      "language_loss": 0.88136333,
      "learning_rate": 3.930467125294116e-06,
      "loss": 0.90407848,
      "num_input_tokens_seen": 19716705,
      "step": 929,
      "time_per_iteration": 3.6250522136688232
    },
    {
      "auxiliary_loss_clip": 0.01170518,
      "auxiliary_loss_mlp": 0.01004065,
      "balance_loss_clip": 1.05310941,
      "balance_loss_mlp": 1.0005846,
      "epoch": 0.1118258882943546,
      "flos": 64586239499520.0,
      "grad_norm": 0.917464861603979,
      "language_loss": 0.60388494,
      "learning_rate": 3.930263364313458e-06,
      "loss": 0.62563074,
      "num_input_tokens_seen": 19767275,
      "step": 930,
      "time_per_iteration": 3.047334909439087
    },
    {
      "auxiliary_loss_clip": 0.01183766,
      "auxiliary_loss_mlp": 0.01057621,
      "balance_loss_clip": 1.06465161,
      "balance_loss_mlp": 1.04472852,
      "epoch": 0.11194613118499369,
      "flos": 17201965985280.0,
      "grad_norm": 1.8827048515468505,
      "language_loss": 0.83113968,
      "learning_rate": 3.930059310513384e-06,
      "loss": 0.85355353,
      "num_input_tokens_seen": 19786315,
      "step": 931,
      "time_per_iteration": 3.7271523475646973
    },
    {
      "auxiliary_loss_clip": 0.01200734,
      "auxiliary_loss_mlp": 0.01051223,
      "balance_loss_clip": 1.06340289,
      "balance_loss_mlp": 1.03897417,
      "epoch": 0.11206637407563277,
      "flos": 31863465728640.0,
      "grad_norm": 1.8876887915642886,
      "language_loss": 0.84085441,
      "learning_rate": 3.929854963924846e-06,
      "loss": 0.86337399,
      "num_input_tokens_seen": 19806580,
      "step": 932,
      "time_per_iteration": 2.8406472206115723
    },
    {
      "auxiliary_loss_clip": 0.01195372,
      "auxiliary_loss_mlp": 0.01044028,
      "balance_loss_clip": 1.0634613,
      "balance_loss_mlp": 1.03234589,
      "epoch": 0.11218661696627187,
      "flos": 21945621761280.0,
      "grad_norm": 2.0924399797655573,
      "language_loss": 0.77502251,
      "learning_rate": 3.929650324578845e-06,
      "loss": 0.79741645,
      "num_input_tokens_seen": 19826045,
      "step": 933,
      "time_per_iteration": 2.7853243350982666
    },
    {
      "auxiliary_loss_clip": 0.01166599,
      "auxiliary_loss_mlp": 0.01041692,
      "balance_loss_clip": 1.06078506,
      "balance_loss_mlp": 1.02804875,
      "epoch": 0.11230685985691095,
      "flos": 25878198481920.0,
      "grad_norm": 4.5220886619518685,
      "language_loss": 0.8259536,
      "learning_rate": 3.929445392506423e-06,
      "loss": 0.84803641,
      "num_input_tokens_seen": 19843985,
      "step": 934,
      "time_per_iteration": 2.834606409072876
    },
    {
      "auxiliary_loss_clip": 0.01204851,
      "auxiliary_loss_mlp": 0.01056196,
      "balance_loss_clip": 1.06595588,
      "balance_loss_mlp": 1.043262,
      "epoch": 0.11242710274755005,
      "flos": 22231506107520.0,
      "grad_norm": 2.68533609647213,
      "language_loss": 0.75926787,
      "learning_rate": 3.92924016773867e-06,
      "loss": 0.78187835,
      "num_input_tokens_seen": 19860480,
      "step": 935,
      "time_per_iteration": 2.7732467651367188
    },
    {
      "auxiliary_loss_clip": 0.01173122,
      "auxiliary_loss_mlp": 0.01040283,
      "balance_loss_clip": 1.0582931,
      "balance_loss_mlp": 1.02891624,
      "epoch": 0.11254734563818915,
      "flos": 17712184723200.0,
      "grad_norm": 2.3522874557697797,
      "language_loss": 0.73228586,
      "learning_rate": 3.9290346503067175e-06,
      "loss": 0.75441992,
      "num_input_tokens_seen": 19877145,
      "step": 936,
      "time_per_iteration": 2.760021924972534
    },
    {
      "auxiliary_loss_clip": 0.01199464,
      "auxiliary_loss_mlp": 0.00766102,
      "balance_loss_clip": 1.05821919,
      "balance_loss_mlp": 1.00036931,
      "epoch": 0.11266758852882823,
      "flos": 54930397334400.0,
      "grad_norm": 1.610368856692866,
      "language_loss": 0.78982794,
      "learning_rate": 3.9288288402417415e-06,
      "loss": 0.80948365,
      "num_input_tokens_seen": 19903405,
      "step": 937,
      "time_per_iteration": 3.041670560836792
    },
    {
      "auxiliary_loss_clip": 0.01245155,
      "auxiliary_loss_mlp": 0.01037273,
      "balance_loss_clip": 1.07025838,
      "balance_loss_mlp": 1.02393341,
      "epoch": 0.11278783141946733,
      "flos": 18878132194560.0,
      "grad_norm": 2.7919178357474084,
      "language_loss": 0.70718652,
      "learning_rate": 3.928622737574964e-06,
      "loss": 0.73001081,
      "num_input_tokens_seen": 19918740,
      "step": 938,
      "time_per_iteration": 2.6439192295074463
    },
    {
      "auxiliary_loss_clip": 0.01200075,
      "auxiliary_loss_mlp": 0.01040831,
      "balance_loss_clip": 1.06503344,
      "balance_loss_mlp": 1.02820039,
      "epoch": 0.11290807431010641,
      "flos": 26469252777600.0,
      "grad_norm": 1.8323784796626277,
      "language_loss": 0.90883875,
      "learning_rate": 3.928416342337652e-06,
      "loss": 0.93124783,
      "num_input_tokens_seen": 19938475,
      "step": 939,
      "time_per_iteration": 2.8130569458007812
    },
    {
      "auxiliary_loss_clip": 0.01224643,
      "auxiliary_loss_mlp": 0.0104549,
      "balance_loss_clip": 1.06812537,
      "balance_loss_mlp": 1.03324747,
      "epoch": 0.1130283172007455,
      "flos": 22710590732160.0,
      "grad_norm": 1.684844637654413,
      "language_loss": 0.8299998,
      "learning_rate": 3.928209654561113e-06,
      "loss": 0.85270113,
      "num_input_tokens_seen": 19959310,
      "step": 940,
      "time_per_iteration": 2.7221667766571045
    },
    {
      "auxiliary_loss_clip": 0.01250994,
      "auxiliary_loss_mlp": 0.01043119,
      "balance_loss_clip": 1.07315016,
      "balance_loss_mlp": 1.03242564,
      "epoch": 0.1131485600913846,
      "flos": 23219911630080.0,
      "grad_norm": 2.8402086385605902,
      "language_loss": 0.81522501,
      "learning_rate": 3.928002674276703e-06,
      "loss": 0.83816612,
      "num_input_tokens_seen": 19978700,
      "step": 941,
      "time_per_iteration": 2.6861095428466797
    },
    {
      "auxiliary_loss_clip": 0.01211785,
      "auxiliary_loss_mlp": 0.01049043,
      "balance_loss_clip": 1.062482,
      "balance_loss_mlp": 1.03659773,
      "epoch": 0.11326880298202369,
      "flos": 14064271286400.0,
      "grad_norm": 2.043320964838832,
      "language_loss": 0.75862873,
      "learning_rate": 3.92779540151582e-06,
      "loss": 0.78123701,
      "num_input_tokens_seen": 19995785,
      "step": 942,
      "time_per_iteration": 2.6924564838409424
    },
    {
      "auxiliary_loss_clip": 0.01238282,
      "auxiliary_loss_mlp": 0.01040953,
      "balance_loss_clip": 1.06839204,
      "balance_loss_mlp": 1.02946186,
      "epoch": 0.11338904587266278,
      "flos": 16325386479360.0,
      "grad_norm": 1.7658831609889354,
      "language_loss": 0.85581911,
      "learning_rate": 3.927587836309907e-06,
      "loss": 0.87861145,
      "num_input_tokens_seen": 20013615,
      "step": 943,
      "time_per_iteration": 2.6721203327178955
    },
    {
      "auxiliary_loss_clip": 0.01253085,
      "auxiliary_loss_mlp": 0.01042273,
      "balance_loss_clip": 1.07202697,
      "balance_loss_mlp": 1.03009582,
      "epoch": 0.11350928876330187,
      "flos": 24426258923520.0,
      "grad_norm": 8.400774543969513,
      "language_loss": 0.78273714,
      "learning_rate": 3.927379978690452e-06,
      "loss": 0.80569065,
      "num_input_tokens_seen": 20032880,
      "step": 944,
      "time_per_iteration": 2.6439177989959717
    },
    {
      "auxiliary_loss_clip": 0.01216052,
      "auxiliary_loss_mlp": 0.01036205,
      "balance_loss_clip": 1.06809127,
      "balance_loss_mlp": 1.02563131,
      "epoch": 0.11362953165394096,
      "flos": 24497074586880.0,
      "grad_norm": 2.4443332970194276,
      "language_loss": 0.87582052,
      "learning_rate": 3.927171828688987e-06,
      "loss": 0.89834309,
      "num_input_tokens_seen": 20052405,
      "step": 945,
      "time_per_iteration": 2.811194658279419
    },
    {
      "auxiliary_loss_clip": 0.01211544,
      "auxiliary_loss_mlp": 0.0104323,
      "balance_loss_clip": 1.06723642,
      "balance_loss_mlp": 1.03134465,
      "epoch": 0.11374977454458005,
      "flos": 24060831909120.0,
      "grad_norm": 4.8142998271374315,
      "language_loss": 0.82487983,
      "learning_rate": 3.926963386337088e-06,
      "loss": 0.84742761,
      "num_input_tokens_seen": 20070635,
      "step": 946,
      "time_per_iteration": 2.772714138031006
    },
    {
      "auxiliary_loss_clip": 0.01245081,
      "auxiliary_loss_mlp": 0.01044607,
      "balance_loss_clip": 1.07193804,
      "balance_loss_mlp": 1.03085649,
      "epoch": 0.11387001743521914,
      "flos": 39457638967680.0,
      "grad_norm": 2.3370750142573558,
      "language_loss": 0.70106405,
      "learning_rate": 3.926754651666375e-06,
      "loss": 0.72396094,
      "num_input_tokens_seen": 20091195,
      "step": 947,
      "time_per_iteration": 2.7628235816955566
    },
    {
      "auxiliary_loss_clip": 0.01224829,
      "auxiliary_loss_mlp": 0.0105366,
      "balance_loss_clip": 1.06683993,
      "balance_loss_mlp": 1.04187047,
      "epoch": 0.11399026032585824,
      "flos": 25082454533760.0,
      "grad_norm": 2.739463628217387,
      "language_loss": 0.78038323,
      "learning_rate": 3.926545624708513e-06,
      "loss": 0.80316818,
      "num_input_tokens_seen": 20110435,
      "step": 948,
      "time_per_iteration": 2.776820421218872
    },
    {
      "auxiliary_loss_clip": 0.01215722,
      "auxiliary_loss_mlp": 0.01038613,
      "balance_loss_clip": 1.06393981,
      "balance_loss_mlp": 1.02732992,
      "epoch": 0.11411050321649732,
      "flos": 17961835224960.0,
      "grad_norm": 2.0711633832529777,
      "language_loss": 0.85893261,
      "learning_rate": 3.926336305495213e-06,
      "loss": 0.88147599,
      "num_input_tokens_seen": 20128995,
      "step": 949,
      "time_per_iteration": 2.66129207611084
    },
    {
      "auxiliary_loss_clip": 0.01234076,
      "auxiliary_loss_mlp": 0.01040117,
      "balance_loss_clip": 1.07098293,
      "balance_loss_mlp": 1.0282383,
      "epoch": 0.11423074610713642,
      "flos": 22455409536000.0,
      "grad_norm": 4.680188965769591,
      "language_loss": 0.8864367,
      "learning_rate": 3.926126694058226e-06,
      "loss": 0.90917861,
      "num_input_tokens_seen": 20148145,
      "step": 950,
      "time_per_iteration": 2.6664254665374756
    },
    {
      "auxiliary_loss_clip": 0.01230676,
      "auxiliary_loss_mlp": 0.00765188,
      "balance_loss_clip": 1.06872928,
      "balance_loss_mlp": 1.00047874,
      "epoch": 0.1143509889977755,
      "flos": 19717687756800.0,
      "grad_norm": 1.597102028996383,
      "language_loss": 0.81939375,
      "learning_rate": 3.92591679042935e-06,
      "loss": 0.83935237,
      "num_input_tokens_seen": 20168035,
      "step": 951,
      "time_per_iteration": 2.7025365829467773
    },
    {
      "auxiliary_loss_clip": 0.01195717,
      "auxiliary_loss_mlp": 0.01049778,
      "balance_loss_clip": 1.06808889,
      "balance_loss_mlp": 1.03830373,
      "epoch": 0.1144712318884146,
      "flos": 19822869757440.0,
      "grad_norm": 1.662087786800047,
      "language_loss": 0.82111192,
      "learning_rate": 3.92570659464043e-06,
      "loss": 0.84356689,
      "num_input_tokens_seen": 20186095,
      "step": 952,
      "time_per_iteration": 3.6108574867248535
    },
    {
      "auxiliary_loss_clip": 0.01229121,
      "auxiliary_loss_mlp": 0.01044552,
      "balance_loss_clip": 1.067909,
      "balance_loss_mlp": 1.03316188,
      "epoch": 0.1145914747790537,
      "flos": 14939198766720.0,
      "grad_norm": 16.396108708830475,
      "language_loss": 0.79960024,
      "learning_rate": 3.925496106723349e-06,
      "loss": 0.82233697,
      "num_input_tokens_seen": 20203535,
      "step": 953,
      "time_per_iteration": 3.6345202922821045
    },
    {
      "auxiliary_loss_clip": 0.01196968,
      "auxiliary_loss_mlp": 0.0103917,
      "balance_loss_clip": 1.06583023,
      "balance_loss_mlp": 1.02750564,
      "epoch": 0.11471171766969278,
      "flos": 19865029345920.0,
      "grad_norm": 1.8772559142621197,
      "language_loss": 0.83734012,
      "learning_rate": 3.9252853267100405e-06,
      "loss": 0.85970151,
      "num_input_tokens_seen": 20222780,
      "step": 954,
      "time_per_iteration": 2.7093794345855713
    },
    {
      "auxiliary_loss_clip": 0.01191955,
      "auxiliary_loss_mlp": 0.00766079,
      "balance_loss_clip": 1.0637598,
      "balance_loss_mlp": 1.00054145,
      "epoch": 0.11483196056033187,
      "flos": 22526476594560.0,
      "grad_norm": 1.8016139109313718,
      "language_loss": 0.83943152,
      "learning_rate": 3.9250742546324786e-06,
      "loss": 0.85901189,
      "num_input_tokens_seen": 20243015,
      "step": 955,
      "time_per_iteration": 2.8244965076446533
    },
    {
      "auxiliary_loss_clip": 0.01206536,
      "auxiliary_loss_mlp": 0.00765266,
      "balance_loss_clip": 1.06251204,
      "balance_loss_mlp": 1.0003686,
      "epoch": 0.11495220345097096,
      "flos": 28220292887040.0,
      "grad_norm": 1.7633667413668894,
      "language_loss": 0.86558795,
      "learning_rate": 3.924862890522683e-06,
      "loss": 0.88530594,
      "num_input_tokens_seen": 20263025,
      "step": 956,
      "time_per_iteration": 3.7624151706695557
    },
    {
      "auxiliary_loss_clip": 0.01253702,
      "auxiliary_loss_mlp": 0.01044626,
      "balance_loss_clip": 1.07082152,
      "balance_loss_mlp": 1.03291345,
      "epoch": 0.11507244634161005,
      "flos": 17492267704320.0,
      "grad_norm": 3.348027436912016,
      "language_loss": 0.85888773,
      "learning_rate": 3.9246512344127174e-06,
      "loss": 0.88187099,
      "num_input_tokens_seen": 20280685,
      "step": 957,
      "time_per_iteration": 3.5719265937805176
    },
    {
      "auxiliary_loss_clip": 0.01188629,
      "auxiliary_loss_mlp": 0.01042729,
      "balance_loss_clip": 1.05891252,
      "balance_loss_mlp": 1.03065884,
      "epoch": 0.11519268923224914,
      "flos": 22564937082240.0,
      "grad_norm": 1.9303628297337103,
      "language_loss": 0.81825793,
      "learning_rate": 3.9244392863346895e-06,
      "loss": 0.84057152,
      "num_input_tokens_seen": 20300090,
      "step": 958,
      "time_per_iteration": 2.7377617359161377
    },
    {
      "auxiliary_loss_clip": 0.01211024,
      "auxiliary_loss_mlp": 0.01043572,
      "balance_loss_clip": 1.06642222,
      "balance_loss_mlp": 1.03053081,
      "epoch": 0.11531293212288823,
      "flos": 16982839065600.0,
      "grad_norm": 1.9741197951348979,
      "language_loss": 0.92367715,
      "learning_rate": 3.9242270463207524e-06,
      "loss": 0.94622308,
      "num_input_tokens_seen": 20318480,
      "step": 959,
      "time_per_iteration": 2.657332181930542
    },
    {
      "auxiliary_loss_clip": 0.01210473,
      "auxiliary_loss_mlp": 0.01044913,
      "balance_loss_clip": 1.06589508,
      "balance_loss_mlp": 1.03280115,
      "epoch": 0.11543317501352733,
      "flos": 12422004537600.0,
      "grad_norm": 2.7055605920966372,
      "language_loss": 0.85078192,
      "learning_rate": 3.924014514403102e-06,
      "loss": 0.87333584,
      "num_input_tokens_seen": 20334635,
      "step": 960,
      "time_per_iteration": 2.7375566959381104
    },
    {
      "auxiliary_loss_clip": 0.01205242,
      "auxiliary_loss_mlp": 0.01038478,
      "balance_loss_clip": 1.06426263,
      "balance_loss_mlp": 1.02544832,
      "epoch": 0.11555341790416641,
      "flos": 19821648695040.0,
      "grad_norm": 2.014071126710349,
      "language_loss": 0.91418099,
      "learning_rate": 3.92380169061398e-06,
      "loss": 0.93661821,
      "num_input_tokens_seen": 20352415,
      "step": 961,
      "time_per_iteration": 2.7268917560577393
    },
    {
      "auxiliary_loss_clip": 0.01200275,
      "auxiliary_loss_mlp": 0.01042386,
      "balance_loss_clip": 1.06094205,
      "balance_loss_mlp": 1.03041172,
      "epoch": 0.11567366079480551,
      "flos": 25738865625600.0,
      "grad_norm": 1.9646774051967288,
      "language_loss": 0.84166944,
      "learning_rate": 3.9235885749856705e-06,
      "loss": 0.86409605,
      "num_input_tokens_seen": 20371095,
      "step": 962,
      "time_per_iteration": 2.759117364883423
    },
    {
      "auxiliary_loss_clip": 0.01210995,
      "auxiliary_loss_mlp": 0.01040902,
      "balance_loss_clip": 1.06653929,
      "balance_loss_mlp": 1.02821207,
      "epoch": 0.1157939036854446,
      "flos": 18223301301120.0,
      "grad_norm": 2.1405939228867092,
      "language_loss": 0.8242119,
      "learning_rate": 3.9233751675505035e-06,
      "loss": 0.84673089,
      "num_input_tokens_seen": 20389805,
      "step": 963,
      "time_per_iteration": 2.8083367347717285
    },
    {
      "auxiliary_loss_clip": 0.01244569,
      "auxiliary_loss_mlp": 0.00765698,
      "balance_loss_clip": 1.07481647,
      "balance_loss_mlp": 1.00044453,
      "epoch": 0.11591414657608369,
      "flos": 23073755189760.0,
      "grad_norm": 2.120599432999408,
      "language_loss": 0.84846169,
      "learning_rate": 3.923161468340853e-06,
      "loss": 0.86856437,
      "num_input_tokens_seen": 20409640,
      "step": 964,
      "time_per_iteration": 2.687742233276367
    },
    {
      "auxiliary_loss_clip": 0.01182211,
      "auxiliary_loss_mlp": 0.01040506,
      "balance_loss_clip": 1.05924773,
      "balance_loss_mlp": 1.02783954,
      "epoch": 0.11603438946672277,
      "flos": 19461716461440.0,
      "grad_norm": 1.844540117880501,
      "language_loss": 0.81521225,
      "learning_rate": 3.9229474773891374e-06,
      "loss": 0.83743942,
      "num_input_tokens_seen": 20428180,
      "step": 965,
      "time_per_iteration": 2.799740791320801
    },
    {
      "auxiliary_loss_clip": 0.01200093,
      "auxiliary_loss_mlp": 0.01046319,
      "balance_loss_clip": 1.05667424,
      "balance_loss_mlp": 1.03399277,
      "epoch": 0.11615463235736187,
      "flos": 26831986272000.0,
      "grad_norm": 2.3055882035767676,
      "language_loss": 0.83772433,
      "learning_rate": 3.922733194727818e-06,
      "loss": 0.86018848,
      "num_input_tokens_seen": 20447975,
      "step": 966,
      "time_per_iteration": 2.788776159286499
    },
    {
      "auxiliary_loss_clip": 0.01221987,
      "auxiliary_loss_mlp": 0.01039411,
      "balance_loss_clip": 1.06732178,
      "balance_loss_mlp": 1.02701354,
      "epoch": 0.11627487524800097,
      "flos": 18580324533120.0,
      "grad_norm": 2.26684537669028,
      "language_loss": 0.87464356,
      "learning_rate": 3.922518620389402e-06,
      "loss": 0.89725757,
      "num_input_tokens_seen": 20464840,
      "step": 967,
      "time_per_iteration": 2.6827316284179688
    },
    {
      "auxiliary_loss_clip": 0.01214344,
      "auxiliary_loss_mlp": 0.01046892,
      "balance_loss_clip": 1.06573081,
      "balance_loss_mlp": 1.03662777,
      "epoch": 0.11639511813864005,
      "flos": 18150474476160.0,
      "grad_norm": 2.085270310402683,
      "language_loss": 0.89410806,
      "learning_rate": 3.922303754406439e-06,
      "loss": 0.91672051,
      "num_input_tokens_seen": 20482680,
      "step": 968,
      "time_per_iteration": 2.645942211151123
    },
    {
      "auxiliary_loss_clip": 0.01187575,
      "auxiliary_loss_mlp": 0.01046821,
      "balance_loss_clip": 1.0616976,
      "balance_loss_mlp": 1.03414869,
      "epoch": 0.11651536102927915,
      "flos": 20922023888640.0,
      "grad_norm": 1.823930607354902,
      "language_loss": 0.78769076,
      "learning_rate": 3.922088596811526e-06,
      "loss": 0.81003475,
      "num_input_tokens_seen": 20501810,
      "step": 969,
      "time_per_iteration": 2.8245155811309814
    },
    {
      "auxiliary_loss_clip": 0.0124344,
      "auxiliary_loss_mlp": 0.01045346,
      "balance_loss_clip": 1.07397246,
      "balance_loss_mlp": 1.03333616,
      "epoch": 0.11663560391991823,
      "flos": 16508602776960.0,
      "grad_norm": 2.313908150108955,
      "language_loss": 0.86715096,
      "learning_rate": 3.9218731476373e-06,
      "loss": 0.89003885,
      "num_input_tokens_seen": 20517995,
      "step": 970,
      "time_per_iteration": 2.6210505962371826
    },
    {
      "auxiliary_loss_clip": 0.01243956,
      "auxiliary_loss_mlp": 0.01047767,
      "balance_loss_clip": 1.07297051,
      "balance_loss_mlp": 1.03443933,
      "epoch": 0.11675584681055733,
      "flos": 19865029345920.0,
      "grad_norm": 2.2307116300935865,
      "language_loss": 0.84546924,
      "learning_rate": 3.9216574069164455e-06,
      "loss": 0.86838651,
      "num_input_tokens_seen": 20536970,
      "step": 971,
      "time_per_iteration": 2.6479201316833496
    },
    {
      "auxiliary_loss_clip": 0.01230196,
      "auxiliary_loss_mlp": 0.01039444,
      "balance_loss_clip": 1.0643177,
      "balance_loss_mlp": 1.02832794,
      "epoch": 0.11687608970119642,
      "flos": 21944364785280.0,
      "grad_norm": 1.5821627894381647,
      "language_loss": 0.80168879,
      "learning_rate": 3.921441374681691e-06,
      "loss": 0.82438523,
      "num_input_tokens_seen": 20557030,
      "step": 972,
      "time_per_iteration": 2.620694875717163
    },
    {
      "auxiliary_loss_clip": 0.01185689,
      "auxiliary_loss_mlp": 0.01043394,
      "balance_loss_clip": 1.05579221,
      "balance_loss_mlp": 1.03144944,
      "epoch": 0.1169963325918355,
      "flos": 24061155131520.0,
      "grad_norm": 2.044867764627898,
      "language_loss": 0.64918113,
      "learning_rate": 3.921225050965808e-06,
      "loss": 0.67147195,
      "num_input_tokens_seen": 20576915,
      "step": 973,
      "time_per_iteration": 2.731482744216919
    },
    {
      "auxiliary_loss_clip": 0.01172095,
      "auxiliary_loss_mlp": 0.01045199,
      "balance_loss_clip": 1.06189442,
      "balance_loss_mlp": 1.03342676,
      "epoch": 0.1171165754824746,
      "flos": 23368151059200.0,
      "grad_norm": 2.4676422153633237,
      "language_loss": 0.74889398,
      "learning_rate": 3.921008435801612e-06,
      "loss": 0.7710669,
      "num_input_tokens_seen": 20596000,
      "step": 974,
      "time_per_iteration": 2.826157808303833
    },
    {
      "auxiliary_loss_clip": 0.01189192,
      "auxiliary_loss_mlp": 0.01047645,
      "balance_loss_clip": 1.05726409,
      "balance_loss_mlp": 1.03481817,
      "epoch": 0.11723681837311369,
      "flos": 18552243075840.0,
      "grad_norm": 2.3572720060860526,
      "language_loss": 0.75769401,
      "learning_rate": 3.920791529221963e-06,
      "loss": 0.78006244,
      "num_input_tokens_seen": 20614675,
      "step": 975,
      "time_per_iteration": 2.75502347946167
    },
    {
      "auxiliary_loss_clip": 0.01213199,
      "auxiliary_loss_mlp": 0.0076592,
      "balance_loss_clip": 1.06605983,
      "balance_loss_mlp": 1.00053859,
      "epoch": 0.11735706126375278,
      "flos": 23550541344000.0,
      "grad_norm": 2.005088958510715,
      "language_loss": 0.76556671,
      "learning_rate": 3.920574331259768e-06,
      "loss": 0.78535789,
      "num_input_tokens_seen": 20635875,
      "step": 976,
      "time_per_iteration": 2.777291774749756
    },
    {
      "auxiliary_loss_clip": 0.01187591,
      "auxiliary_loss_mlp": 0.01049305,
      "balance_loss_clip": 1.06618786,
      "balance_loss_mlp": 1.03694296,
      "epoch": 0.11747730415439187,
      "flos": 22381541216640.0,
      "grad_norm": 2.5737822047840857,
      "language_loss": 0.79631811,
      "learning_rate": 3.9203568419479716e-06,
      "loss": 0.81868708,
      "num_input_tokens_seen": 20656430,
      "step": 977,
      "time_per_iteration": 3.6550371646881104
    },
    {
      "auxiliary_loss_clip": 0.01222427,
      "auxiliary_loss_mlp": 0.0104439,
      "balance_loss_clip": 1.06253505,
      "balance_loss_mlp": 1.03267717,
      "epoch": 0.11759754704503096,
      "flos": 22200731130240.0,
      "grad_norm": 1.9367113154022784,
      "language_loss": 0.75281221,
      "learning_rate": 3.92013906131957e-06,
      "loss": 0.77548033,
      "num_input_tokens_seen": 20675360,
      "step": 978,
      "time_per_iteration": 2.8685085773468018
    },
    {
      "auxiliary_loss_clip": 0.01205436,
      "auxiliary_loss_mlp": 0.01045043,
      "balance_loss_clip": 1.06528878,
      "balance_loss_mlp": 1.03445137,
      "epoch": 0.11771778993567006,
      "flos": 22309755886080.0,
      "grad_norm": 1.7964236580360156,
      "language_loss": 0.82760811,
      "learning_rate": 3.9199209894076e-06,
      "loss": 0.8501128,
      "num_input_tokens_seen": 20695675,
      "step": 979,
      "time_per_iteration": 3.6236791610717773
    },
    {
      "auxiliary_loss_clip": 0.01166576,
      "auxiliary_loss_mlp": 0.01038288,
      "balance_loss_clip": 1.05406618,
      "balance_loss_mlp": 1.02566993,
      "epoch": 0.11783803282630914,
      "flos": 21288169175040.0,
      "grad_norm": 2.026129602234755,
      "language_loss": 0.90098143,
      "learning_rate": 3.919702626245142e-06,
      "loss": 0.92303008,
      "num_input_tokens_seen": 20715330,
      "step": 980,
      "time_per_iteration": 3.0341501235961914
    },
    {
      "auxiliary_loss_clip": 0.01194374,
      "auxiliary_loss_mlp": 0.00765847,
      "balance_loss_clip": 1.06209159,
      "balance_loss_mlp": 1.00044751,
      "epoch": 0.11795827571694824,
      "flos": 25371535190400.0,
      "grad_norm": 2.1603019791029756,
      "language_loss": 0.66531038,
      "learning_rate": 3.919483971865322e-06,
      "loss": 0.68491256,
      "num_input_tokens_seen": 20735325,
      "step": 981,
      "time_per_iteration": 2.81779146194458
    },
    {
      "auxiliary_loss_clip": 0.01210028,
      "auxiliary_loss_mlp": 0.00764579,
      "balance_loss_clip": 1.06665933,
      "balance_loss_mlp": 1.00050163,
      "epoch": 0.11807851860758732,
      "flos": 23622218933760.0,
      "grad_norm": 2.0652642096091673,
      "language_loss": 0.87803507,
      "learning_rate": 3.91926502630131e-06,
      "loss": 0.89778113,
      "num_input_tokens_seen": 20755940,
      "step": 982,
      "time_per_iteration": 4.610082149505615
    },
    {
      "auxiliary_loss_clip": 0.01238357,
      "auxiliary_loss_mlp": 0.0105148,
      "balance_loss_clip": 1.07360327,
      "balance_loss_mlp": 1.04076326,
      "epoch": 0.11819876149822642,
      "flos": 24972496024320.0,
      "grad_norm": 2.0228084660532244,
      "language_loss": 0.72431338,
      "learning_rate": 3.91904578958632e-06,
      "loss": 0.74721169,
      "num_input_tokens_seen": 20775355,
      "step": 983,
      "time_per_iteration": 2.659411668777466
    },
    {
      "auxiliary_loss_clip": 0.01202593,
      "auxiliary_loss_mlp": 0.01043106,
      "balance_loss_clip": 1.06262422,
      "balance_loss_mlp": 1.03153694,
      "epoch": 0.11831900438886551,
      "flos": 23003226835200.0,
      "grad_norm": 2.974846275370122,
      "language_loss": 0.84352899,
      "learning_rate": 3.918826261753608e-06,
      "loss": 0.86598593,
      "num_input_tokens_seen": 20794935,
      "step": 984,
      "time_per_iteration": 2.696531057357788
    },
    {
      "auxiliary_loss_clip": 0.01226137,
      "auxiliary_loss_mlp": 0.01040277,
      "balance_loss_clip": 1.06551623,
      "balance_loss_mlp": 1.02915442,
      "epoch": 0.1184392472795046,
      "flos": 27965147604480.0,
      "grad_norm": 2.882368531977893,
      "language_loss": 0.71602374,
      "learning_rate": 3.918606442836478e-06,
      "loss": 0.73868787,
      "num_input_tokens_seen": 20817155,
      "step": 985,
      "time_per_iteration": 2.6971445083618164
    },
    {
      "auxiliary_loss_clip": 0.01214973,
      "auxiliary_loss_mlp": 0.01043185,
      "balance_loss_clip": 1.06967199,
      "balance_loss_mlp": 1.03311801,
      "epoch": 0.1185594901701437,
      "flos": 19898497843200.0,
      "grad_norm": 1.7582896806795871,
      "language_loss": 0.77747095,
      "learning_rate": 3.918386332868277e-06,
      "loss": 0.80005252,
      "num_input_tokens_seen": 20835125,
      "step": 986,
      "time_per_iteration": 2.7326135635375977
    },
    {
      "auxiliary_loss_clip": 0.01206438,
      "auxiliary_loss_mlp": 0.01050202,
      "balance_loss_clip": 1.06396163,
      "balance_loss_mlp": 1.03879344,
      "epoch": 0.11867973306078278,
      "flos": 18912354877440.0,
      "grad_norm": 1.852823466792518,
      "language_loss": 0.94369888,
      "learning_rate": 3.918165931882394e-06,
      "loss": 0.9662652,
      "num_input_tokens_seen": 20853525,
      "step": 987,
      "time_per_iteration": 2.722661018371582
    },
    {
      "auxiliary_loss_clip": 0.01215535,
      "auxiliary_loss_mlp": 0.01038396,
      "balance_loss_clip": 1.06791329,
      "balance_loss_mlp": 1.02743506,
      "epoch": 0.11879997595142187,
      "flos": 16982803152000.0,
      "grad_norm": 3.381426710461559,
      "language_loss": 0.75517178,
      "learning_rate": 3.917945239912264e-06,
      "loss": 0.77771109,
      "num_input_tokens_seen": 20871000,
      "step": 988,
      "time_per_iteration": 2.6587677001953125
    },
    {
      "auxiliary_loss_clip": 0.01127406,
      "auxiliary_loss_mlp": 0.0103794,
      "balance_loss_clip": 1.05475211,
      "balance_loss_mlp": 1.0271579,
      "epoch": 0.11892021884206096,
      "flos": 17530369056000.0,
      "grad_norm": 2.0299550988572657,
      "language_loss": 0.75447178,
      "learning_rate": 3.917724256991367e-06,
      "loss": 0.77612519,
      "num_input_tokens_seen": 20889745,
      "step": 989,
      "time_per_iteration": 3.20344614982605
    },
    {
      "auxiliary_loss_clip": 0.01152274,
      "auxiliary_loss_mlp": 0.01040825,
      "balance_loss_clip": 1.05704355,
      "balance_loss_mlp": 1.02943993,
      "epoch": 0.11904046173270005,
      "flos": 30955895763840.0,
      "grad_norm": 2.6592440378522606,
      "language_loss": 0.81708282,
      "learning_rate": 3.9175029831532245e-06,
      "loss": 0.83901381,
      "num_input_tokens_seen": 20909260,
      "step": 990,
      "time_per_iteration": 3.3328588008880615
    },
    {
      "auxiliary_loss_clip": 0.01231456,
      "auxiliary_loss_mlp": 0.01049488,
      "balance_loss_clip": 1.06876254,
      "balance_loss_mlp": 1.03827047,
      "epoch": 0.11916070462333915,
      "flos": 20157234485760.0,
      "grad_norm": 1.9532158853968153,
      "language_loss": 0.88442969,
      "learning_rate": 3.917281418431404e-06,
      "loss": 0.90723914,
      "num_input_tokens_seen": 20928305,
      "step": 991,
      "time_per_iteration": 2.748448133468628
    },
    {
      "auxiliary_loss_clip": 0.01221101,
      "auxiliary_loss_mlp": 0.01044804,
      "balance_loss_clip": 1.07090771,
      "balance_loss_mlp": 1.03312778,
      "epoch": 0.11928094751397823,
      "flos": 23551115961600.0,
      "grad_norm": 4.219503183664438,
      "language_loss": 0.77057594,
      "learning_rate": 3.917059562859516e-06,
      "loss": 0.79323494,
      "num_input_tokens_seen": 20947630,
      "step": 992,
      "time_per_iteration": 2.692129373550415
    },
    {
      "auxiliary_loss_clip": 0.01233545,
      "auxiliary_loss_mlp": 0.01047385,
      "balance_loss_clip": 1.07074213,
      "balance_loss_mlp": 1.0366087,
      "epoch": 0.11940119040461733,
      "flos": 23908426502400.0,
      "grad_norm": 2.250733740880552,
      "language_loss": 0.88731664,
      "learning_rate": 3.916837416471218e-06,
      "loss": 0.91012597,
      "num_input_tokens_seen": 20964250,
      "step": 993,
      "time_per_iteration": 2.717998743057251
    },
    {
      "auxiliary_loss_clip": 0.01167658,
      "auxiliary_loss_mlp": 0.01035355,
      "balance_loss_clip": 1.05762768,
      "balance_loss_mlp": 1.02425098,
      "epoch": 0.11952143329525641,
      "flos": 13844533835520.0,
      "grad_norm": 2.6610899289784995,
      "language_loss": 0.72413588,
      "learning_rate": 3.916614979300207e-06,
      "loss": 0.74616599,
      "num_input_tokens_seen": 20979095,
      "step": 994,
      "time_per_iteration": 2.859813690185547
    },
    {
      "auxiliary_loss_clip": 0.01233291,
      "auxiliary_loss_mlp": 0.00765372,
      "balance_loss_clip": 1.07185626,
      "balance_loss_mlp": 1.00049615,
      "epoch": 0.11964167618589551,
      "flos": 27015525792000.0,
      "grad_norm": 1.5564687260701755,
      "language_loss": 0.78686655,
      "learning_rate": 3.9163922513802274e-06,
      "loss": 0.80685318,
      "num_input_tokens_seen": 21001430,
      "step": 995,
      "time_per_iteration": 2.7196919918060303
    },
    {
      "auxiliary_loss_clip": 0.01220627,
      "auxiliary_loss_mlp": 0.01040508,
      "balance_loss_clip": 1.06512201,
      "balance_loss_mlp": 1.02911186,
      "epoch": 0.1197619190765346,
      "flos": 12567622273920.0,
      "grad_norm": 2.515966936719116,
      "language_loss": 0.82549274,
      "learning_rate": 3.916169232745067e-06,
      "loss": 0.84810412,
      "num_input_tokens_seen": 21019105,
      "step": 996,
      "time_per_iteration": 2.716442584991455
    },
    {
      "auxiliary_loss_clip": 0.0120772,
      "auxiliary_loss_mlp": 0.01044861,
      "balance_loss_clip": 1.06543803,
      "balance_loss_mlp": 1.03205812,
      "epoch": 0.11988216196717369,
      "flos": 16909437623040.0,
      "grad_norm": 3.031108478476725,
      "language_loss": 0.91700816,
      "learning_rate": 3.915945923428559e-06,
      "loss": 0.93953401,
      "num_input_tokens_seen": 21035630,
      "step": 997,
      "time_per_iteration": 2.7407562732696533
    },
    {
      "auxiliary_loss_clip": 0.01217755,
      "auxiliary_loss_mlp": 0.01049787,
      "balance_loss_clip": 1.0632962,
      "balance_loss_mlp": 1.03837872,
      "epoch": 0.12000240485781279,
      "flos": 16216577205120.0,
      "grad_norm": 2.3338593040189646,
      "language_loss": 0.83554822,
      "learning_rate": 3.915722323464577e-06,
      "loss": 0.85822362,
      "num_input_tokens_seen": 21054235,
      "step": 998,
      "time_per_iteration": 2.6456892490386963
    },
    {
      "auxiliary_loss_clip": 0.01209907,
      "auxiliary_loss_mlp": 0.01041004,
      "balance_loss_clip": 1.0659287,
      "balance_loss_mlp": 1.02858853,
      "epoch": 0.12012264774845187,
      "flos": 49344887525760.0,
      "grad_norm": 2.3413457748589948,
      "language_loss": 0.70557976,
      "learning_rate": 3.91549843288704e-06,
      "loss": 0.7280888,
      "num_input_tokens_seen": 21077915,
      "step": 999,
      "time_per_iteration": 2.8755686283111572
    },
    {
      "auxiliary_loss_clip": 0.01138601,
      "auxiliary_loss_mlp": 0.01045973,
      "balance_loss_clip": 1.05660963,
      "balance_loss_mlp": 1.03466642,
      "epoch": 0.12024289063909097,
      "flos": 26979435601920.0,
      "grad_norm": 1.9123328985667187,
      "language_loss": 0.78746563,
      "learning_rate": 3.915274251729916e-06,
      "loss": 0.80931139,
      "num_input_tokens_seen": 21099205,
      "step": 1000,
      "time_per_iteration": 2.8947675228118896
    },
    {
      "auxiliary_loss_clip": 0.01218193,
      "auxiliary_loss_mlp": 0.01041917,
      "balance_loss_clip": 1.06611645,
      "balance_loss_mlp": 1.03056216,
      "epoch": 0.12036313352973005,
      "flos": 19537308633600.0,
      "grad_norm": 2.5086687401289693,
      "language_loss": 0.9038282,
      "learning_rate": 3.91504978002721e-06,
      "loss": 0.92642933,
      "num_input_tokens_seen": 21118260,
      "step": 1001,
      "time_per_iteration": 2.685427665710449
    },
    {
      "auxiliary_loss_clip": 0.01213316,
      "auxiliary_loss_mlp": 0.01042867,
      "balance_loss_clip": 1.06494188,
      "balance_loss_mlp": 1.0307436,
      "epoch": 0.12048337642036915,
      "flos": 17268256535040.0,
      "grad_norm": 2.959104787556186,
      "language_loss": 0.76296252,
      "learning_rate": 3.914825017812974e-06,
      "loss": 0.78552437,
      "num_input_tokens_seen": 21134910,
      "step": 1002,
      "time_per_iteration": 3.8697071075439453
    },
    {
      "auxiliary_loss_clip": 0.01237836,
      "auxiliary_loss_mlp": 0.01042297,
      "balance_loss_clip": 1.06881392,
      "balance_loss_mlp": 1.03031635,
      "epoch": 0.12060361931100824,
      "flos": 22856962654080.0,
      "grad_norm": 2.311759242376226,
      "language_loss": 0.72344011,
      "learning_rate": 3.9145999651213065e-06,
      "loss": 0.74624145,
      "num_input_tokens_seen": 21154150,
      "step": 1003,
      "time_per_iteration": 2.71954083442688
    },
    {
      "auxiliary_loss_clip": 0.01184884,
      "auxiliary_loss_mlp": 0.01042077,
      "balance_loss_clip": 1.06037533,
      "balance_loss_mlp": 1.03096652,
      "epoch": 0.12072386220164733,
      "flos": 16726795943040.0,
      "grad_norm": 2.660142536043153,
      "language_loss": 0.88285244,
      "learning_rate": 3.9143746219863465e-06,
      "loss": 0.90512204,
      "num_input_tokens_seen": 21171255,
      "step": 1004,
      "time_per_iteration": 2.7434239387512207
    },
    {
      "auxiliary_loss_clip": 0.01086143,
      "auxiliary_loss_mlp": 0.01008405,
      "balance_loss_clip": 1.04182792,
      "balance_loss_mlp": 1.00513864,
      "epoch": 0.12084410509228642,
      "flos": 55144176105600.0,
      "grad_norm": 0.9636415869163153,
      "language_loss": 0.64792013,
      "learning_rate": 3.914148988442278e-06,
      "loss": 0.66886568,
      "num_input_tokens_seen": 21227045,
      "step": 1005,
      "time_per_iteration": 4.715017795562744
    },
    {
      "auxiliary_loss_clip": 0.01220734,
      "auxiliary_loss_mlp": 0.01047375,
      "balance_loss_clip": 1.06774855,
      "balance_loss_mlp": 1.03538263,
      "epoch": 0.1209643479829255,
      "flos": 26760236855040.0,
      "grad_norm": 2.573510931909728,
      "language_loss": 0.95198435,
      "learning_rate": 3.91392306452333e-06,
      "loss": 0.9746654,
      "num_input_tokens_seen": 21244120,
      "step": 1006,
      "time_per_iteration": 3.0064895153045654
    },
    {
      "auxiliary_loss_clip": 0.01238533,
      "auxiliary_loss_mlp": 0.01049161,
      "balance_loss_clip": 1.06692243,
      "balance_loss_mlp": 1.03698325,
      "epoch": 0.1210845908735646,
      "flos": 11035026725760.0,
      "grad_norm": 2.9399135403903944,
      "language_loss": 0.66269457,
      "learning_rate": 3.913696850263774e-06,
      "loss": 0.68557155,
      "num_input_tokens_seen": 21258485,
      "step": 1007,
      "time_per_iteration": 2.600024938583374
    },
    {
      "auxiliary_loss_clip": 0.01176498,
      "auxiliary_loss_mlp": 0.01048998,
      "balance_loss_clip": 1.0607971,
      "balance_loss_mlp": 1.03683865,
      "epoch": 0.1212048337642037,
      "flos": 20484631975680.0,
      "grad_norm": 2.1974049580718154,
      "language_loss": 0.79368842,
      "learning_rate": 3.913470345697929e-06,
      "loss": 0.81594336,
      "num_input_tokens_seen": 21277115,
      "step": 1008,
      "time_per_iteration": 3.9063613414764404
    },
    {
      "auxiliary_loss_clip": 0.01232997,
      "auxiliary_loss_mlp": 0.00765467,
      "balance_loss_clip": 1.07060111,
      "balance_loss_mlp": 1.00059342,
      "epoch": 0.12132507665484278,
      "flos": 22346061557760.0,
      "grad_norm": 2.207134343519735,
      "language_loss": 0.85303944,
      "learning_rate": 3.913243550860153e-06,
      "loss": 0.87302411,
      "num_input_tokens_seen": 21294880,
      "step": 1009,
      "time_per_iteration": 3.9482977390289307
    },
    {
      "auxiliary_loss_clip": 0.01211466,
      "auxiliary_loss_mlp": 0.01046914,
      "balance_loss_clip": 1.06864548,
      "balance_loss_mlp": 1.03473115,
      "epoch": 0.12144531954548188,
      "flos": 29314957818240.0,
      "grad_norm": 3.301654126253166,
      "language_loss": 0.7595793,
      "learning_rate": 3.913016465784852e-06,
      "loss": 0.78216308,
      "num_input_tokens_seen": 21315555,
      "step": 1010,
      "time_per_iteration": 2.8686635494232178
    },
    {
      "auxiliary_loss_clip": 0.0122605,
      "auxiliary_loss_mlp": 0.01045882,
      "balance_loss_clip": 1.06724024,
      "balance_loss_mlp": 1.03385401,
      "epoch": 0.12156556243612096,
      "flos": 20485242506880.0,
      "grad_norm": 2.4697312691302975,
      "language_loss": 0.72224236,
      "learning_rate": 3.912789090506474e-06,
      "loss": 0.74496162,
      "num_input_tokens_seen": 21334815,
      "step": 1011,
      "time_per_iteration": 2.6620256900787354
    },
    {
      "auxiliary_loss_clip": 0.01251586,
      "auxiliary_loss_mlp": 0.01038957,
      "balance_loss_clip": 1.07193136,
      "balance_loss_mlp": 1.02760196,
      "epoch": 0.12168580532676006,
      "flos": 16472009796480.0,
      "grad_norm": 2.1715923183723547,
      "language_loss": 0.72081399,
      "learning_rate": 3.9125614250595114e-06,
      "loss": 0.74371946,
      "num_input_tokens_seen": 21351025,
      "step": 1012,
      "time_per_iteration": 2.688713312149048
    },
    {
      "auxiliary_loss_clip": 0.01249388,
      "auxiliary_loss_mlp": 0.01041903,
      "balance_loss_clip": 1.06855583,
      "balance_loss_mlp": 1.03067362,
      "epoch": 0.12180604821739914,
      "flos": 15341290588800.0,
      "grad_norm": 2.7723179721330227,
      "language_loss": 0.89364213,
      "learning_rate": 3.912333469478502e-06,
      "loss": 0.91655505,
      "num_input_tokens_seen": 21368990,
      "step": 1013,
      "time_per_iteration": 2.638922691345215
    },
    {
      "auxiliary_loss_clip": 0.01150488,
      "auxiliary_loss_mlp": 0.00764687,
      "balance_loss_clip": 1.05464184,
      "balance_loss_mlp": 1.00053835,
      "epoch": 0.12192629110803824,
      "flos": 19318038059520.0,
      "grad_norm": 2.215263593605911,
      "language_loss": 0.78227079,
      "learning_rate": 3.912105223798025e-06,
      "loss": 0.8014226,
      "num_input_tokens_seen": 21388410,
      "step": 1014,
      "time_per_iteration": 2.804736375808716
    },
    {
      "auxiliary_loss_clip": 0.01115682,
      "auxiliary_loss_mlp": 0.01014393,
      "balance_loss_clip": 1.04678822,
      "balance_loss_mlp": 1.01136494,
      "epoch": 0.12204653399867733,
      "flos": 47725354085760.0,
      "grad_norm": 1.02703960369017,
      "language_loss": 0.67631483,
      "learning_rate": 3.9118766880527065e-06,
      "loss": 0.6976155,
      "num_input_tokens_seen": 21442845,
      "step": 1015,
      "time_per_iteration": 3.2173290252685547
    },
    {
      "auxiliary_loss_clip": 0.01216806,
      "auxiliary_loss_mlp": 0.01034651,
      "balance_loss_clip": 1.06563807,
      "balance_loss_mlp": 1.02326608,
      "epoch": 0.12216677688931642,
      "flos": 18221936584320.0,
      "grad_norm": 1.7242352556179892,
      "language_loss": 0.73820388,
      "learning_rate": 3.9116478622772145e-06,
      "loss": 0.76071852,
      "num_input_tokens_seen": 21461420,
      "step": 1016,
      "time_per_iteration": 2.7188985347747803
    },
    {
      "auxiliary_loss_clip": 0.01195598,
      "auxiliary_loss_mlp": 0.01041411,
      "balance_loss_clip": 1.06097078,
      "balance_loss_mlp": 1.02935886,
      "epoch": 0.12228701977995551,
      "flos": 27525636789120.0,
      "grad_norm": 1.8673017436170336,
      "language_loss": 0.87975103,
      "learning_rate": 3.911418746506261e-06,
      "loss": 0.90212119,
      "num_input_tokens_seen": 21481550,
      "step": 1017,
      "time_per_iteration": 2.787818670272827
    },
    {
      "auxiliary_loss_clip": 0.01206449,
      "auxiliary_loss_mlp": 0.01046726,
      "balance_loss_clip": 1.06728697,
      "balance_loss_mlp": 1.03499556,
      "epoch": 0.1224072626705946,
      "flos": 21798136517760.0,
      "grad_norm": 2.310639317568059,
      "language_loss": 0.78288382,
      "learning_rate": 3.911189340774604e-06,
      "loss": 0.80541557,
      "num_input_tokens_seen": 21501680,
      "step": 1018,
      "time_per_iteration": 2.7360663414001465
    },
    {
      "auxiliary_loss_clip": 0.01207624,
      "auxiliary_loss_mlp": 0.01047208,
      "balance_loss_clip": 1.06310105,
      "balance_loss_mlp": 1.03593111,
      "epoch": 0.1225275055612337,
      "flos": 20703758895360.0,
      "grad_norm": 2.3614440413692077,
      "language_loss": 0.79363263,
      "learning_rate": 3.910959645117043e-06,
      "loss": 0.81618094,
      "num_input_tokens_seen": 21521015,
      "step": 1019,
      "time_per_iteration": 2.721245288848877
    },
    {
      "auxiliary_loss_clip": 0.01110064,
      "auxiliary_loss_mlp": 0.01006246,
      "balance_loss_clip": 1.04319227,
      "balance_loss_mlp": 1.00309908,
      "epoch": 0.12264774845187278,
      "flos": 57745294462080.0,
      "grad_norm": 0.8160543281966237,
      "language_loss": 0.5665834,
      "learning_rate": 3.910729659568423e-06,
      "loss": 0.5877465,
      "num_input_tokens_seen": 21578200,
      "step": 1020,
      "time_per_iteration": 3.2669734954833984
    },
    {
      "auxiliary_loss_clip": 0.01233044,
      "auxiliary_loss_mlp": 0.01045366,
      "balance_loss_clip": 1.06815565,
      "balance_loss_mlp": 1.03512049,
      "epoch": 0.12276799134251187,
      "flos": 26396282298240.0,
      "grad_norm": 1.9437085439236008,
      "language_loss": 0.82264644,
      "learning_rate": 3.9104993841636344e-06,
      "loss": 0.84543055,
      "num_input_tokens_seen": 21598770,
      "step": 1021,
      "time_per_iteration": 2.747901678085327
    },
    {
      "auxiliary_loss_clip": 0.01204934,
      "auxiliary_loss_mlp": 0.01043247,
      "balance_loss_clip": 1.06190729,
      "balance_loss_mlp": 1.03288782,
      "epoch": 0.12288823423315097,
      "flos": 21064193919360.0,
      "grad_norm": 1.7733158616961706,
      "language_loss": 0.80682391,
      "learning_rate": 3.910268818937608e-06,
      "loss": 0.82930565,
      "num_input_tokens_seen": 21616925,
      "step": 1022,
      "time_per_iteration": 2.732736825942993
    },
    {
      "auxiliary_loss_clip": 0.01216891,
      "auxiliary_loss_mlp": 0.01041182,
      "balance_loss_clip": 1.06522846,
      "balance_loss_mlp": 1.03033972,
      "epoch": 0.12300847712379005,
      "flos": 12312441077760.0,
      "grad_norm": 3.1060688482591168,
      "language_loss": 0.87165171,
      "learning_rate": 3.9100379639253196e-06,
      "loss": 0.89423245,
      "num_input_tokens_seen": 21633645,
      "step": 1023,
      "time_per_iteration": 2.6963424682617188
    },
    {
      "auxiliary_loss_clip": 0.01231624,
      "auxiliary_loss_mlp": 0.0104127,
      "balance_loss_clip": 1.06749487,
      "balance_loss_mlp": 1.03063619,
      "epoch": 0.12312872001442915,
      "flos": 16762239688320.0,
      "grad_norm": 2.7796457182101553,
      "language_loss": 0.8648122,
      "learning_rate": 3.909806819161791e-06,
      "loss": 0.88754117,
      "num_input_tokens_seen": 21649120,
      "step": 1024,
      "time_per_iteration": 2.669355630874634
    },
    {
      "auxiliary_loss_clip": 0.01225944,
      "auxiliary_loss_mlp": 0.01046689,
      "balance_loss_clip": 1.06706643,
      "balance_loss_mlp": 1.03411198,
      "epoch": 0.12324896290506823,
      "flos": 18404937400320.0,
      "grad_norm": 1.895387558186852,
      "language_loss": 0.8580277,
      "learning_rate": 3.909575384682086e-06,
      "loss": 0.88075399,
      "num_input_tokens_seen": 21668000,
      "step": 1025,
      "time_per_iteration": 2.7095723152160645
    },
    {
      "auxiliary_loss_clip": 0.01174778,
      "auxiliary_loss_mlp": 0.0104469,
      "balance_loss_clip": 1.05976856,
      "balance_loss_mlp": 1.03382444,
      "epoch": 0.12336920579570733,
      "flos": 18915407533440.0,
      "grad_norm": 3.526186663914366,
      "language_loss": 0.69223964,
      "learning_rate": 3.9093436605213144e-06,
      "loss": 0.71443433,
      "num_input_tokens_seen": 21688500,
      "step": 1026,
      "time_per_iteration": 2.797375202178955
    },
    {
      "auxiliary_loss_clip": 0.01177968,
      "auxiliary_loss_mlp": 0.01044788,
      "balance_loss_clip": 1.05466294,
      "balance_loss_mlp": 1.03339124,
      "epoch": 0.12348944868634643,
      "flos": 23878369797120.0,
      "grad_norm": 1.8410566895681209,
      "language_loss": 0.79690421,
      "learning_rate": 3.909111646714627e-06,
      "loss": 0.81913173,
      "num_input_tokens_seen": 21709345,
      "step": 1027,
      "time_per_iteration": 2.776338577270508
    },
    {
      "auxiliary_loss_clip": 0.01213123,
      "auxiliary_loss_mlp": 0.01052538,
      "balance_loss_clip": 1.06800532,
      "balance_loss_mlp": 1.04161298,
      "epoch": 0.12360969157698551,
      "flos": 19026084314880.0,
      "grad_norm": 2.0868555151813877,
      "language_loss": 0.72578168,
      "learning_rate": 3.9088793432972206e-06,
      "loss": 0.74843836,
      "num_input_tokens_seen": 21728165,
      "step": 1028,
      "time_per_iteration": 2.7151052951812744
    },
    {
      "auxiliary_loss_clip": 0.01188356,
      "auxiliary_loss_mlp": 0.01046048,
      "balance_loss_clip": 1.06458378,
      "balance_loss_mlp": 1.03571296,
      "epoch": 0.1237299344676246,
      "flos": 13224607983360.0,
      "grad_norm": 2.0271405809235485,
      "language_loss": 0.82539845,
      "learning_rate": 3.908646750304336e-06,
      "loss": 0.8477425,
      "num_input_tokens_seen": 21745850,
      "step": 1029,
      "time_per_iteration": 3.6844773292541504
    },
    {
      "auxiliary_loss_clip": 0.01221574,
      "auxiliary_loss_mlp": 0.01052728,
      "balance_loss_clip": 1.07042539,
      "balance_loss_mlp": 1.04064059,
      "epoch": 0.12385017735826369,
      "flos": 20485673470080.0,
      "grad_norm": 1.7480195764045736,
      "language_loss": 0.87117732,
      "learning_rate": 3.908413867771257e-06,
      "loss": 0.89392042,
      "num_input_tokens_seen": 21764760,
      "step": 1030,
      "time_per_iteration": 2.782160520553589
    },
    {
      "auxiliary_loss_clip": 0.01219171,
      "auxiliary_loss_mlp": 0.01041617,
      "balance_loss_clip": 1.06604314,
      "balance_loss_mlp": 1.029899,
      "epoch": 0.12397042024890279,
      "flos": 17347835116800.0,
      "grad_norm": 3.8442099826545633,
      "language_loss": 0.80784583,
      "learning_rate": 3.908180695733311e-06,
      "loss": 0.83045369,
      "num_input_tokens_seen": 21784250,
      "step": 1031,
      "time_per_iteration": 3.6405856609344482
    },
    {
      "auxiliary_loss_clip": 0.01212426,
      "auxiliary_loss_mlp": 0.01037367,
      "balance_loss_clip": 1.06294322,
      "balance_loss_mlp": 1.02666855,
      "epoch": 0.12409066313954187,
      "flos": 20412343854720.0,
      "grad_norm": 2.0349190906650514,
      "language_loss": 0.82738328,
      "learning_rate": 3.907947234225871e-06,
      "loss": 0.84988117,
      "num_input_tokens_seen": 21803260,
      "step": 1032,
      "time_per_iteration": 2.76497220993042
    },
    {
      "auxiliary_loss_clip": 0.01212773,
      "auxiliary_loss_mlp": 0.01043986,
      "balance_loss_clip": 1.06763375,
      "balance_loss_mlp": 1.0331974,
      "epoch": 0.12421090603018096,
      "flos": 20736688688640.0,
      "grad_norm": 2.0510951698782507,
      "language_loss": 0.8718577,
      "learning_rate": 3.907713483284352e-06,
      "loss": 0.89442527,
      "num_input_tokens_seen": 21822735,
      "step": 1033,
      "time_per_iteration": 2.734189510345459
    },
    {
      "auxiliary_loss_clip": 0.01216713,
      "auxiliary_loss_mlp": 0.01038539,
      "balance_loss_clip": 1.06556845,
      "balance_loss_mlp": 1.0275476,
      "epoch": 0.12433114892082006,
      "flos": 24498834353280.0,
      "grad_norm": 2.520306947259053,
      "language_loss": 0.97520828,
      "learning_rate": 3.907479442944216e-06,
      "loss": 0.99776083,
      "num_input_tokens_seen": 21841140,
      "step": 1034,
      "time_per_iteration": 3.6374473571777344
    },
    {
      "auxiliary_loss_clip": 0.01216014,
      "auxiliary_loss_mlp": 0.01037562,
      "balance_loss_clip": 1.06524396,
      "balance_loss_mlp": 1.02651107,
      "epoch": 0.12445139181145914,
      "flos": 19682315838720.0,
      "grad_norm": 3.9088782662336246,
      "language_loss": 0.92828977,
      "learning_rate": 3.907245113240963e-06,
      "loss": 0.95082551,
      "num_input_tokens_seen": 21859260,
      "step": 1035,
      "time_per_iteration": 3.943352460861206
    },
    {
      "auxiliary_loss_clip": 0.01219305,
      "auxiliary_loss_mlp": 0.01035899,
      "balance_loss_clip": 1.06591034,
      "balance_loss_mlp": 1.02496219,
      "epoch": 0.12457163470209824,
      "flos": 46423087522560.0,
      "grad_norm": 2.144412748696375,
      "language_loss": 0.73778337,
      "learning_rate": 3.907010494210144e-06,
      "loss": 0.76033545,
      "num_input_tokens_seen": 21881920,
      "step": 1036,
      "time_per_iteration": 2.94035267829895
    },
    {
      "auxiliary_loss_clip": 0.01208313,
      "auxiliary_loss_mlp": 0.00765422,
      "balance_loss_clip": 1.0622462,
      "balance_loss_mlp": 1.00085473,
      "epoch": 0.12469187759273732,
      "flos": 20376289578240.0,
      "grad_norm": 2.2161892697571894,
      "language_loss": 0.91672277,
      "learning_rate": 3.9067755858873495e-06,
      "loss": 0.93646014,
      "num_input_tokens_seen": 21898720,
      "step": 1037,
      "time_per_iteration": 2.7470946311950684
    },
    {
      "auxiliary_loss_clip": 0.01165694,
      "auxiliary_loss_mlp": 0.01010376,
      "balance_loss_clip": 1.0587728,
      "balance_loss_mlp": 1.0058459,
      "epoch": 0.12481212048337642,
      "flos": 69224641447680.0,
      "grad_norm": 0.8587265430636277,
      "language_loss": 0.62822413,
      "learning_rate": 3.906540388308214e-06,
      "loss": 0.64998484,
      "num_input_tokens_seen": 21958305,
      "step": 1038,
      "time_per_iteration": 3.258749485015869
    },
    {
      "auxiliary_loss_clip": 0.01194214,
      "auxiliary_loss_mlp": 0.01039336,
      "balance_loss_clip": 1.06158459,
      "balance_loss_mlp": 1.02909613,
      "epoch": 0.12493236337401552,
      "flos": 18223696350720.0,
      "grad_norm": 4.391935421618155,
      "language_loss": 0.81438804,
      "learning_rate": 3.906304901508417e-06,
      "loss": 0.83672357,
      "num_input_tokens_seen": 21977205,
      "step": 1039,
      "time_per_iteration": 2.7938008308410645
    },
    {
      "auxiliary_loss_clip": 0.01202636,
      "auxiliary_loss_mlp": 0.01042855,
      "balance_loss_clip": 1.06985831,
      "balance_loss_mlp": 1.0315423,
      "epoch": 0.12505260626465461,
      "flos": 30044375303040.0,
      "grad_norm": 2.755955177864249,
      "language_loss": 0.75750506,
      "learning_rate": 3.9060691255236835e-06,
      "loss": 0.77995992,
      "num_input_tokens_seen": 21997770,
      "step": 1040,
      "time_per_iteration": 2.7805564403533936
    },
    {
      "auxiliary_loss_clip": 0.01230747,
      "auxiliary_loss_mlp": 0.01042648,
      "balance_loss_clip": 1.06711602,
      "balance_loss_mlp": 1.03130531,
      "epoch": 0.1251728491552937,
      "flos": 24433980347520.0,
      "grad_norm": 1.8913555173171557,
      "language_loss": 0.80411005,
      "learning_rate": 3.905833060389778e-06,
      "loss": 0.82684398,
      "num_input_tokens_seen": 22021890,
      "step": 1041,
      "time_per_iteration": 2.7446184158325195
    },
    {
      "auxiliary_loss_clip": 0.01215557,
      "auxiliary_loss_mlp": 0.01045085,
      "balance_loss_clip": 1.06909788,
      "balance_loss_mlp": 1.03400993,
      "epoch": 0.12529309204593278,
      "flos": 27119809952640.0,
      "grad_norm": 2.671066472445692,
      "language_loss": 0.78050745,
      "learning_rate": 3.905596706142513e-06,
      "loss": 0.80311382,
      "num_input_tokens_seen": 22043300,
      "step": 1042,
      "time_per_iteration": 2.812837600708008
    },
    {
      "auxiliary_loss_clip": 0.01210267,
      "auxiliary_loss_mlp": 0.01037922,
      "balance_loss_clip": 1.0652988,
      "balance_loss_mlp": 1.02747977,
      "epoch": 0.12541333493657186,
      "flos": 30774151923840.0,
      "grad_norm": 2.4539416446826787,
      "language_loss": 0.86040294,
      "learning_rate": 3.9053600628177435e-06,
      "loss": 0.88288486,
      "num_input_tokens_seen": 22062910,
      "step": 1043,
      "time_per_iteration": 2.7819652557373047
    },
    {
      "auxiliary_loss_clip": 0.01215059,
      "auxiliary_loss_mlp": 0.01038138,
      "balance_loss_clip": 1.06216753,
      "balance_loss_mlp": 1.02815413,
      "epoch": 0.12553357782721097,
      "flos": 23659566099840.0,
      "grad_norm": 2.1779614275639987,
      "language_loss": 0.84622508,
      "learning_rate": 3.905123130451367e-06,
      "loss": 0.86875713,
      "num_input_tokens_seen": 22084010,
      "step": 1044,
      "time_per_iteration": 2.706038475036621
    },
    {
      "auxiliary_loss_clip": 0.01207875,
      "auxiliary_loss_mlp": 0.01044603,
      "balance_loss_clip": 1.06624258,
      "balance_loss_mlp": 1.03352821,
      "epoch": 0.12565382071785006,
      "flos": 24863758577280.0,
      "grad_norm": 1.826680487198115,
      "language_loss": 0.79566413,
      "learning_rate": 3.904885909079326e-06,
      "loss": 0.81818891,
      "num_input_tokens_seen": 22102795,
      "step": 1045,
      "time_per_iteration": 2.817748785018921
    },
    {
      "auxiliary_loss_clip": 0.01167602,
      "auxiliary_loss_mlp": 0.01044581,
      "balance_loss_clip": 1.05431068,
      "balance_loss_mlp": 1.03322005,
      "epoch": 0.12577406360848914,
      "flos": 21360780518400.0,
      "grad_norm": 3.73367499819671,
      "language_loss": 0.78425479,
      "learning_rate": 3.904648398737607e-06,
      "loss": 0.80637658,
      "num_input_tokens_seen": 22121360,
      "step": 1046,
      "time_per_iteration": 2.7096657752990723
    },
    {
      "auxiliary_loss_clip": 0.01209717,
      "auxiliary_loss_mlp": 0.00764196,
      "balance_loss_clip": 1.06306648,
      "balance_loss_mlp": 1.00117087,
      "epoch": 0.12589430649912825,
      "flos": 36138056774400.0,
      "grad_norm": 1.8878782641514296,
      "language_loss": 0.77969158,
      "learning_rate": 3.9044105994622406e-06,
      "loss": 0.79943073,
      "num_input_tokens_seen": 22142505,
      "step": 1047,
      "time_per_iteration": 2.862276792526245
    },
    {
      "auxiliary_loss_clip": 0.01219414,
      "auxiliary_loss_mlp": 0.01038037,
      "balance_loss_clip": 1.06496811,
      "balance_loss_mlp": 1.02648592,
      "epoch": 0.12601454938976733,
      "flos": 25337671643520.0,
      "grad_norm": 2.566255617852633,
      "language_loss": 0.81854618,
      "learning_rate": 3.9041725112893005e-06,
      "loss": 0.84112066,
      "num_input_tokens_seen": 22163730,
      "step": 1048,
      "time_per_iteration": 2.747633934020996
    },
    {
      "auxiliary_loss_clip": 0.01216319,
      "auxiliary_loss_mlp": 0.01035482,
      "balance_loss_clip": 1.06396163,
      "balance_loss_mlp": 1.02490211,
      "epoch": 0.12613479228040642,
      "flos": 15560094286080.0,
      "grad_norm": 2.218943021428406,
      "language_loss": 0.75081438,
      "learning_rate": 3.903934134254904e-06,
      "loss": 0.77333236,
      "num_input_tokens_seen": 22181520,
      "step": 1049,
      "time_per_iteration": 2.659261465072632
    },
    {
      "auxiliary_loss_clip": 0.01201814,
      "auxiliary_loss_mlp": 0.01038346,
      "balance_loss_clip": 1.06530976,
      "balance_loss_mlp": 1.02711034,
      "epoch": 0.1262550351710455,
      "flos": 21470595373440.0,
      "grad_norm": 2.396353500580679,
      "language_loss": 0.84524465,
      "learning_rate": 3.903695468395213e-06,
      "loss": 0.86764622,
      "num_input_tokens_seen": 22199390,
      "step": 1050,
      "time_per_iteration": 2.749138355255127
    },
    {
      "auxiliary_loss_clip": 0.01199945,
      "auxiliary_loss_mlp": 0.01036054,
      "balance_loss_clip": 1.06325185,
      "balance_loss_mlp": 1.02548027,
      "epoch": 0.1263752780616846,
      "flos": 31576719456000.0,
      "grad_norm": 2.157183140461772,
      "language_loss": 0.56199753,
      "learning_rate": 3.903456513746434e-06,
      "loss": 0.5843575,
      "num_input_tokens_seen": 22220365,
      "step": 1051,
      "time_per_iteration": 2.767632246017456
    },
    {
      "auxiliary_loss_clip": 0.01196324,
      "auxiliary_loss_mlp": 0.01030802,
      "balance_loss_clip": 1.06274807,
      "balance_loss_mlp": 1.02046704,
      "epoch": 0.1264955209523237,
      "flos": 28768217927040.0,
      "grad_norm": 3.2294911812200975,
      "language_loss": 0.87588239,
      "learning_rate": 3.903217270344815e-06,
      "loss": 0.89815366,
      "num_input_tokens_seen": 22240615,
      "step": 1052,
      "time_per_iteration": 2.8084933757781982
    },
    {
      "auxiliary_loss_clip": 0.01169506,
      "auxiliary_loss_mlp": 0.01040739,
      "balance_loss_clip": 1.06115925,
      "balance_loss_mlp": 1.03017759,
      "epoch": 0.12661576384296278,
      "flos": 29241125412480.0,
      "grad_norm": 1.7669997410457317,
      "language_loss": 0.82370257,
      "learning_rate": 3.902977738226648e-06,
      "loss": 0.84580505,
      "num_input_tokens_seen": 22261350,
      "step": 1053,
      "time_per_iteration": 2.8843138217926025
    },
    {
      "auxiliary_loss_clip": 0.0117314,
      "auxiliary_loss_mlp": 0.01043076,
      "balance_loss_clip": 1.05481911,
      "balance_loss_mlp": 1.03080344,
      "epoch": 0.12673600673360189,
      "flos": 20850346298880.0,
      "grad_norm": 2.1235572003646275,
      "language_loss": 0.91320527,
      "learning_rate": 3.902737917428273e-06,
      "loss": 0.93536735,
      "num_input_tokens_seen": 22279515,
      "step": 1054,
      "time_per_iteration": 3.7094836235046387
    },
    {
      "auxiliary_loss_clip": 0.01233572,
      "auxiliary_loss_mlp": 0.00764811,
      "balance_loss_clip": 1.06522346,
      "balance_loss_mlp": 1.00132263,
      "epoch": 0.12685624962424097,
      "flos": 25263695583360.0,
      "grad_norm": 1.7234922958986298,
      "language_loss": 0.84237593,
      "learning_rate": 3.902497807986068e-06,
      "loss": 0.86235982,
      "num_input_tokens_seen": 22299535,
      "step": 1055,
      "time_per_iteration": 2.741575241088867
    },
    {
      "auxiliary_loss_clip": 0.01184709,
      "auxiliary_loss_mlp": 0.01040636,
      "balance_loss_clip": 1.0563488,
      "balance_loss_mlp": 1.02970457,
      "epoch": 0.12697649251488005,
      "flos": 27527109246720.0,
      "grad_norm": 3.0403544166164225,
      "language_loss": 0.84031087,
      "learning_rate": 3.902257409936458e-06,
      "loss": 0.86256433,
      "num_input_tokens_seen": 22320300,
      "step": 1056,
      "time_per_iteration": 3.7457423210144043
    },
    {
      "auxiliary_loss_clip": 0.012244,
      "auxiliary_loss_mlp": 0.01038609,
      "balance_loss_clip": 1.06630087,
      "balance_loss_mlp": 1.02798712,
      "epoch": 0.12709673540551916,
      "flos": 21251863503360.0,
      "grad_norm": 2.188862769232487,
      "language_loss": 0.84149319,
      "learning_rate": 3.902016723315912e-06,
      "loss": 0.86412328,
      "num_input_tokens_seen": 22338240,
      "step": 1057,
      "time_per_iteration": 2.645205497741699
    },
    {
      "auxiliary_loss_clip": 0.0122904,
      "auxiliary_loss_mlp": 0.01038674,
      "balance_loss_clip": 1.06808209,
      "balance_loss_mlp": 1.02792144,
      "epoch": 0.12721697829615825,
      "flos": 25337707557120.0,
      "grad_norm": 2.3519395158797356,
      "language_loss": 0.69493932,
      "learning_rate": 3.901775748160941e-06,
      "loss": 0.71761644,
      "num_input_tokens_seen": 22357420,
      "step": 1058,
      "time_per_iteration": 2.7220680713653564
    },
    {
      "auxiliary_loss_clip": 0.01215624,
      "auxiliary_loss_mlp": 0.00759965,
      "balance_loss_clip": 1.08347178,
      "balance_loss_mlp": 1.00046885,
      "epoch": 0.12733722118679733,
      "flos": 61943287754880.0,
      "grad_norm": 0.8000429653541585,
      "language_loss": 0.60924339,
      "learning_rate": 3.901534484508101e-06,
      "loss": 0.62899935,
      "num_input_tokens_seen": 22420095,
      "step": 1059,
      "time_per_iteration": 3.2608399391174316
    },
    {
      "auxiliary_loss_clip": 0.01225317,
      "auxiliary_loss_mlp": 0.01034101,
      "balance_loss_clip": 1.06364751,
      "balance_loss_mlp": 1.0231756,
      "epoch": 0.1274574640774364,
      "flos": 26976742081920.0,
      "grad_norm": 3.2840661726313036,
      "language_loss": 0.74781734,
      "learning_rate": 3.901292932393991e-06,
      "loss": 0.77041155,
      "num_input_tokens_seen": 22438975,
      "step": 1060,
      "time_per_iteration": 3.6434147357940674
    },
    {
      "auxiliary_loss_clip": 0.01213691,
      "auxiliary_loss_mlp": 0.00763989,
      "balance_loss_clip": 1.06467211,
      "balance_loss_mlp": 1.00121284,
      "epoch": 0.12757770696807552,
      "flos": 22236318529920.0,
      "grad_norm": 2.2718398933655743,
      "language_loss": 0.85439479,
      "learning_rate": 3.9010510918552555e-06,
      "loss": 0.87417161,
      "num_input_tokens_seen": 22458050,
      "step": 1061,
      "time_per_iteration": 3.650049924850464
    },
    {
      "auxiliary_loss_clip": 0.01229539,
      "auxiliary_loss_mlp": 0.01042027,
      "balance_loss_clip": 1.06813359,
      "balance_loss_mlp": 1.03069007,
      "epoch": 0.1276979498587146,
      "flos": 28547905858560.0,
      "grad_norm": 2.950578417507809,
      "language_loss": 0.74903834,
      "learning_rate": 3.900808962928581e-06,
      "loss": 0.77175403,
      "num_input_tokens_seen": 22475665,
      "step": 1062,
      "time_per_iteration": 2.663411855697632
    },
    {
      "auxiliary_loss_clip": 0.01209053,
      "auxiliary_loss_mlp": 0.01043801,
      "balance_loss_clip": 1.06750512,
      "balance_loss_mlp": 1.03388858,
      "epoch": 0.1278181927493537,
      "flos": 17420338719360.0,
      "grad_norm": 2.0611797595088204,
      "language_loss": 0.89088184,
      "learning_rate": 3.900566545650698e-06,
      "loss": 0.91341043,
      "num_input_tokens_seen": 22493335,
      "step": 1063,
      "time_per_iteration": 2.7773821353912354
    },
    {
      "auxiliary_loss_clip": 0.01205492,
      "auxiliary_loss_mlp": 0.01047754,
      "balance_loss_clip": 1.06447077,
      "balance_loss_mlp": 1.03636336,
      "epoch": 0.1279384356399928,
      "flos": 21138636856320.0,
      "grad_norm": 2.050591303079612,
      "language_loss": 0.82175255,
      "learning_rate": 3.900323840058381e-06,
      "loss": 0.84428501,
      "num_input_tokens_seen": 22511045,
      "step": 1064,
      "time_per_iteration": 2.682894229888916
    },
    {
      "auxiliary_loss_clip": 0.0120946,
      "auxiliary_loss_mlp": 0.01039582,
      "balance_loss_clip": 1.05995679,
      "balance_loss_mlp": 1.02886534,
      "epoch": 0.12805867853063188,
      "flos": 26576733248640.0,
      "grad_norm": 2.1166918500187224,
      "language_loss": 0.81709874,
      "learning_rate": 3.900080846188449e-06,
      "loss": 0.83958912,
      "num_input_tokens_seen": 22529635,
      "step": 1065,
      "time_per_iteration": 2.7488439083099365
    },
    {
      "auxiliary_loss_clip": 0.01213616,
      "auxiliary_loss_mlp": 0.00764297,
      "balance_loss_clip": 1.06350541,
      "balance_loss_mlp": 1.001284,
      "epoch": 0.12817892142127096,
      "flos": 16436206915200.0,
      "grad_norm": 1.7788568205897803,
      "language_loss": 0.81403136,
      "learning_rate": 3.8998375640777625e-06,
      "loss": 0.83381045,
      "num_input_tokens_seen": 22547505,
      "step": 1066,
      "time_per_iteration": 2.7024431228637695
    },
    {
      "auxiliary_loss_clip": 0.01122284,
      "auxiliary_loss_mlp": 0.01035883,
      "balance_loss_clip": 1.05002141,
      "balance_loss_mlp": 1.03301036,
      "epoch": 0.12829916431191005,
      "flos": 60757049099520.0,
      "grad_norm": 0.7115117128734016,
      "language_loss": 0.52644563,
      "learning_rate": 3.899593993763229e-06,
      "loss": 0.54802728,
      "num_input_tokens_seen": 22608465,
      "step": 1067,
      "time_per_iteration": 3.179032325744629
    },
    {
      "auxiliary_loss_clip": 0.01202925,
      "auxiliary_loss_mlp": 0.01039969,
      "balance_loss_clip": 1.06423688,
      "balance_loss_mlp": 1.02836418,
      "epoch": 0.12841940720254916,
      "flos": 29786895636480.0,
      "grad_norm": 2.229597509050876,
      "language_loss": 0.81429636,
      "learning_rate": 3.899350135281796e-06,
      "loss": 0.83672535,
      "num_input_tokens_seen": 22629465,
      "step": 1068,
      "time_per_iteration": 2.7142460346221924
    },
    {
      "auxiliary_loss_clip": 0.01193488,
      "auxiliary_loss_mlp": 0.0105184,
      "balance_loss_clip": 1.06368542,
      "balance_loss_mlp": 1.04073536,
      "epoch": 0.12853965009318824,
      "flos": 25951851319680.0,
      "grad_norm": 2.0601605973504826,
      "language_loss": 0.79775107,
      "learning_rate": 3.8991059886704585e-06,
      "loss": 0.82020438,
      "num_input_tokens_seen": 22648970,
      "step": 1069,
      "time_per_iteration": 2.824763298034668
    },
    {
      "auxiliary_loss_clip": 0.01182414,
      "auxiliary_loss_mlp": 0.01048822,
      "balance_loss_clip": 1.05755854,
      "balance_loss_mlp": 1.03737223,
      "epoch": 0.12865989298382732,
      "flos": 30846871008000.0,
      "grad_norm": 1.8874279857998464,
      "language_loss": 0.8263284,
      "learning_rate": 3.898861553966252e-06,
      "loss": 0.8486408,
      "num_input_tokens_seen": 22668620,
      "step": 1070,
      "time_per_iteration": 2.798670768737793
    },
    {
      "auxiliary_loss_clip": 0.01216728,
      "auxiliary_loss_mlp": 0.01042115,
      "balance_loss_clip": 1.06692016,
      "balance_loss_mlp": 1.03195262,
      "epoch": 0.12878013587446643,
      "flos": 25885776251520.0,
      "grad_norm": 1.638936410415994,
      "language_loss": 0.87989962,
      "learning_rate": 3.898616831206257e-06,
      "loss": 0.90248805,
      "num_input_tokens_seen": 22689045,
      "step": 1071,
      "time_per_iteration": 2.7345592975616455
    },
    {
      "auxiliary_loss_clip": 0.01235072,
      "auxiliary_loss_mlp": 0.01033602,
      "balance_loss_clip": 1.06729388,
      "balance_loss_mlp": 1.02273643,
      "epoch": 0.12890037876510552,
      "flos": 23333138277120.0,
      "grad_norm": 2.0047640775665947,
      "language_loss": 0.77048773,
      "learning_rate": 3.8983718204276e-06,
      "loss": 0.79317451,
      "num_input_tokens_seen": 22711265,
      "step": 1072,
      "time_per_iteration": 2.749471426010132
    },
    {
      "auxiliary_loss_clip": 0.01209297,
      "auxiliary_loss_mlp": 0.01043713,
      "balance_loss_clip": 1.06604719,
      "balance_loss_mlp": 1.03268647,
      "epoch": 0.1290206216557446,
      "flos": 23587242065280.0,
      "grad_norm": 2.367096337878051,
      "language_loss": 0.82676733,
      "learning_rate": 3.898126521667446e-06,
      "loss": 0.8492974,
      "num_input_tokens_seen": 22731420,
      "step": 1073,
      "time_per_iteration": 2.66985821723938
    },
    {
      "auxiliary_loss_clip": 0.01213789,
      "auxiliary_loss_mlp": 0.01045392,
      "balance_loss_clip": 1.06758034,
      "balance_loss_mlp": 1.03423357,
      "epoch": 0.12914086454638368,
      "flos": 24170610850560.0,
      "grad_norm": 1.797981162704054,
      "language_loss": 0.83404386,
      "learning_rate": 3.897880934963007e-06,
      "loss": 0.85663563,
      "num_input_tokens_seen": 22750970,
      "step": 1074,
      "time_per_iteration": 2.826704502105713
    },
    {
      "auxiliary_loss_clip": 0.01200985,
      "auxiliary_loss_mlp": 0.01046899,
      "balance_loss_clip": 1.06004119,
      "balance_loss_mlp": 1.0354315,
      "epoch": 0.1292611074370228,
      "flos": 20267157081600.0,
      "grad_norm": 2.082385206174595,
      "language_loss": 0.7830807,
      "learning_rate": 3.89763506035154e-06,
      "loss": 0.80555964,
      "num_input_tokens_seen": 22768820,
      "step": 1075,
      "time_per_iteration": 2.7079989910125732
    },
    {
      "auxiliary_loss_clip": 0.01106093,
      "auxiliary_loss_mlp": 0.01045226,
      "balance_loss_clip": 1.04997945,
      "balance_loss_mlp": 1.03382349,
      "epoch": 0.12938135032766188,
      "flos": 27377684668800.0,
      "grad_norm": 1.8301532075712423,
      "language_loss": 0.81533229,
      "learning_rate": 3.897388897870343e-06,
      "loss": 0.8368454,
      "num_input_tokens_seen": 22789460,
      "step": 1076,
      "time_per_iteration": 3.1551146507263184
    },
    {
      "auxiliary_loss_clip": 0.01172873,
      "auxiliary_loss_mlp": 0.01049675,
      "balance_loss_clip": 1.06063414,
      "balance_loss_mlp": 1.03662157,
      "epoch": 0.12950159321830096,
      "flos": 29277107861760.0,
      "grad_norm": 1.7810453924621932,
      "language_loss": 0.74975377,
      "learning_rate": 3.89714244755676e-06,
      "loss": 0.77197927,
      "num_input_tokens_seen": 22810820,
      "step": 1077,
      "time_per_iteration": 3.1931498050689697
    },
    {
      "auxiliary_loss_clip": 0.01196634,
      "auxiliary_loss_mlp": 0.01040296,
      "balance_loss_clip": 1.06181252,
      "balance_loss_mlp": 1.02908492,
      "epoch": 0.12962183610894007,
      "flos": 24534888629760.0,
      "grad_norm": 2.7715741224004975,
      "language_loss": 0.8596856,
      "learning_rate": 3.896895709448175e-06,
      "loss": 0.88205492,
      "num_input_tokens_seen": 22830570,
      "step": 1078,
      "time_per_iteration": 2.784313678741455
    },
    {
      "auxiliary_loss_clip": 0.01210646,
      "auxiliary_loss_mlp": 0.01037293,
      "balance_loss_clip": 1.06460357,
      "balance_loss_mlp": 1.02587271,
      "epoch": 0.12974207899957915,
      "flos": 11215944552960.0,
      "grad_norm": 2.439663274326565,
      "language_loss": 0.76923907,
      "learning_rate": 3.896648683582019e-06,
      "loss": 0.79171842,
      "num_input_tokens_seen": 22845905,
      "step": 1079,
      "time_per_iteration": 2.738452434539795
    },
    {
      "auxiliary_loss_clip": 0.01217043,
      "auxiliary_loss_mlp": 0.01035381,
      "balance_loss_clip": 1.06281471,
      "balance_loss_mlp": 1.02346587,
      "epoch": 0.12986232189021824,
      "flos": 24717889445760.0,
      "grad_norm": 2.4892986732815605,
      "language_loss": 0.8080039,
      "learning_rate": 3.896401369995766e-06,
      "loss": 0.83052814,
      "num_input_tokens_seen": 22865710,
      "step": 1080,
      "time_per_iteration": 4.097114086151123
    },
    {
      "auxiliary_loss_clip": 0.01242098,
      "auxiliary_loss_mlp": 0.01046275,
      "balance_loss_clip": 1.06841302,
      "balance_loss_mlp": 1.03588605,
      "epoch": 0.12998256478085732,
      "flos": 23915357827200.0,
      "grad_norm": 1.982882673792692,
      "language_loss": 0.7968964,
      "learning_rate": 3.896153768726932e-06,
      "loss": 0.81978017,
      "num_input_tokens_seen": 22886020,
      "step": 1081,
      "time_per_iteration": 2.7276573181152344
    },
    {
      "auxiliary_loss_clip": 0.01181719,
      "auxiliary_loss_mlp": 0.0076512,
      "balance_loss_clip": 1.06125605,
      "balance_loss_mlp": 1.00142741,
      "epoch": 0.13010280767149643,
      "flos": 18624207974400.0,
      "grad_norm": 3.850428324698909,
      "language_loss": 0.87867284,
      "learning_rate": 3.8959058798130806e-06,
      "loss": 0.89814126,
      "num_input_tokens_seen": 22903995,
      "step": 1082,
      "time_per_iteration": 3.709508180618286
    },
    {
      "auxiliary_loss_clip": 0.01227476,
      "auxiliary_loss_mlp": 0.01041651,
      "balance_loss_clip": 1.06658459,
      "balance_loss_mlp": 1.02909184,
      "epoch": 0.1302230505621355,
      "flos": 22783992174720.0,
      "grad_norm": 1.745663377076772,
      "language_loss": 0.75019109,
      "learning_rate": 3.895657703291814e-06,
      "loss": 0.77288234,
      "num_input_tokens_seen": 22924100,
      "step": 1083,
      "time_per_iteration": 2.7456064224243164
    },
    {
      "auxiliary_loss_clip": 0.01195366,
      "auxiliary_loss_mlp": 0.01044656,
      "balance_loss_clip": 1.05855632,
      "balance_loss_mlp": 1.03276479,
      "epoch": 0.1303432934527746,
      "flos": 21323612920320.0,
      "grad_norm": 3.2389424954536445,
      "language_loss": 0.80418909,
      "learning_rate": 3.895409239200781e-06,
      "loss": 0.82658935,
      "num_input_tokens_seen": 22939985,
      "step": 1084,
      "time_per_iteration": 2.7918453216552734
    },
    {
      "auxiliary_loss_clip": 0.01120316,
      "auxiliary_loss_mlp": 0.00765675,
      "balance_loss_clip": 1.05517769,
      "balance_loss_mlp": 1.00144255,
      "epoch": 0.1304635363434137,
      "flos": 20922490765440.0,
      "grad_norm": 3.1438861222539622,
      "language_loss": 0.9156183,
      "learning_rate": 3.895160487577673e-06,
      "loss": 0.93447828,
      "num_input_tokens_seen": 22957555,
      "step": 1085,
      "time_per_iteration": 4.208585262298584
    },
    {
      "auxiliary_loss_clip": 0.01175153,
      "auxiliary_loss_mlp": 0.01014602,
      "balance_loss_clip": 1.05177593,
      "balance_loss_mlp": 1.0085938,
      "epoch": 0.1305837792340528,
      "flos": 63245659080960.0,
      "grad_norm": 0.7864949960514005,
      "language_loss": 0.60874444,
      "learning_rate": 3.894911448460226e-06,
      "loss": 0.630642,
      "num_input_tokens_seen": 23016870,
      "step": 1086,
      "time_per_iteration": 3.6237521171569824
    },
    {
      "auxiliary_loss_clip": 0.01196422,
      "auxiliary_loss_mlp": 0.0104307,
      "balance_loss_clip": 1.05961239,
      "balance_loss_mlp": 1.03164935,
      "epoch": 0.13070402212469187,
      "flos": 26428852955520.0,
      "grad_norm": 2.0599550295312294,
      "language_loss": 0.72849804,
      "learning_rate": 3.8946621218862195e-06,
      "loss": 0.750893,
      "num_input_tokens_seen": 23037870,
      "step": 1087,
      "time_per_iteration": 3.8341715335845947
    },
    {
      "auxiliary_loss_clip": 0.01209612,
      "auxiliary_loss_mlp": 0.01031859,
      "balance_loss_clip": 1.06213641,
      "balance_loss_mlp": 1.02051675,
      "epoch": 0.13082426501533098,
      "flos": 27673409341440.0,
      "grad_norm": 2.011605013752465,
      "language_loss": 0.89367568,
      "learning_rate": 3.894412507893475e-06,
      "loss": 0.91609043,
      "num_input_tokens_seen": 23058150,
      "step": 1088,
      "time_per_iteration": 2.8267650604248047
    },
    {
      "auxiliary_loss_clip": 0.0121312,
      "auxiliary_loss_mlp": 0.01047056,
      "balance_loss_clip": 1.06210709,
      "balance_loss_mlp": 1.03614795,
      "epoch": 0.13094450790597006,
      "flos": 24826770547200.0,
      "grad_norm": 2.0480692277433823,
      "language_loss": 0.71996295,
      "learning_rate": 3.894162606519859e-06,
      "loss": 0.74256468,
      "num_input_tokens_seen": 23077100,
      "step": 1089,
      "time_per_iteration": 2.69512939453125
    },
    {
      "auxiliary_loss_clip": 0.01210006,
      "auxiliary_loss_mlp": 0.0103716,
      "balance_loss_clip": 1.06648183,
      "balance_loss_mlp": 1.0263536,
      "epoch": 0.13106475079660915,
      "flos": 19062605468160.0,
      "grad_norm": 2.0378949240764967,
      "language_loss": 0.76948357,
      "learning_rate": 3.893912417803282e-06,
      "loss": 0.79195523,
      "num_input_tokens_seen": 23096815,
      "step": 1090,
      "time_per_iteration": 2.707449436187744
    },
    {
      "auxiliary_loss_clip": 0.01180655,
      "auxiliary_loss_mlp": 0.01038793,
      "balance_loss_clip": 1.05638731,
      "balance_loss_mlp": 1.02706289,
      "epoch": 0.13118499368724823,
      "flos": 28913189218560.0,
      "grad_norm": 1.8752739559838694,
      "language_loss": 0.77553236,
      "learning_rate": 3.8936619417816975e-06,
      "loss": 0.79772681,
      "num_input_tokens_seen": 23117145,
      "step": 1091,
      "time_per_iteration": 2.783369779586792
    },
    {
      "auxiliary_loss_clip": 0.01177034,
      "auxiliary_loss_mlp": 0.01049184,
      "balance_loss_clip": 1.0586549,
      "balance_loss_mlp": 1.03794241,
      "epoch": 0.13130523657788734,
      "flos": 14283398206080.0,
      "grad_norm": 1.9172417518436582,
      "language_loss": 0.71615559,
      "learning_rate": 3.8934111784931015e-06,
      "loss": 0.73841774,
      "num_input_tokens_seen": 23134595,
      "step": 1092,
      "time_per_iteration": 2.769176721572876
    },
    {
      "auxiliary_loss_clip": 0.01115194,
      "auxiliary_loss_mlp": 0.01009198,
      "balance_loss_clip": 1.0389564,
      "balance_loss_mlp": 1.00366712,
      "epoch": 0.13142547946852642,
      "flos": 70174155519360.0,
      "grad_norm": 0.9205625545877718,
      "language_loss": 0.59003544,
      "learning_rate": 3.893160127975535e-06,
      "loss": 0.61127937,
      "num_input_tokens_seen": 23195285,
      "step": 1093,
      "time_per_iteration": 3.3783438205718994
    },
    {
      "auxiliary_loss_clip": 0.01199824,
      "auxiliary_loss_mlp": 0.01045615,
      "balance_loss_clip": 1.0645082,
      "balance_loss_mlp": 1.0346303,
      "epoch": 0.1315457223591655,
      "flos": 45805998844800.0,
      "grad_norm": 7.3385711219624215,
      "language_loss": 0.81350768,
      "learning_rate": 3.8929087902670826e-06,
      "loss": 0.83596206,
      "num_input_tokens_seen": 23216915,
      "step": 1094,
      "time_per_iteration": 2.9556479454040527
    },
    {
      "auxiliary_loss_clip": 0.01161547,
      "auxiliary_loss_mlp": 0.01009557,
      "balance_loss_clip": 1.04268765,
      "balance_loss_mlp": 1.00426412,
      "epoch": 0.13166596524980462,
      "flos": 62881165820160.0,
      "grad_norm": 0.9223459100520732,
      "language_loss": 0.60711116,
      "learning_rate": 3.8926571654058715e-06,
      "loss": 0.62882221,
      "num_input_tokens_seen": 23273560,
      "step": 1095,
      "time_per_iteration": 3.1885018348693848
    },
    {
      "auxiliary_loss_clip": 0.01225796,
      "auxiliary_loss_mlp": 0.01038086,
      "balance_loss_clip": 1.06401277,
      "balance_loss_mlp": 1.02696395,
      "epoch": 0.1317862081404437,
      "flos": 23586523793280.0,
      "grad_norm": 2.3390871203201318,
      "language_loss": 0.77370763,
      "learning_rate": 3.892405253430074e-06,
      "loss": 0.79634643,
      "num_input_tokens_seen": 23291080,
      "step": 1096,
      "time_per_iteration": 2.721402883529663
    },
    {
      "auxiliary_loss_clip": 0.01205273,
      "auxiliary_loss_mlp": 0.01049352,
      "balance_loss_clip": 1.06304145,
      "balance_loss_mlp": 1.03782403,
      "epoch": 0.13190645103108278,
      "flos": 20260764460800.0,
      "grad_norm": 3.5399093331941494,
      "language_loss": 0.82534313,
      "learning_rate": 3.892153054377904e-06,
      "loss": 0.84788942,
      "num_input_tokens_seen": 23308485,
      "step": 1097,
      "time_per_iteration": 2.776935577392578
    },
    {
      "auxiliary_loss_clip": 0.01113872,
      "auxiliary_loss_mlp": 0.01005927,
      "balance_loss_clip": 1.03419065,
      "balance_loss_mlp": 1.00123012,
      "epoch": 0.13202669392172187,
      "flos": 53455440136320.0,
      "grad_norm": 0.932623731096068,
      "language_loss": 0.59425765,
      "learning_rate": 3.891900568287619e-06,
      "loss": 0.61545569,
      "num_input_tokens_seen": 23360870,
      "step": 1098,
      "time_per_iteration": 3.1374783515930176
    },
    {
      "auxiliary_loss_clip": 0.01220697,
      "auxiliary_loss_mlp": 0.0076572,
      "balance_loss_clip": 1.06520367,
      "balance_loss_mlp": 1.00114369,
      "epoch": 0.13214693681236098,
      "flos": 15851293845120.0,
      "grad_norm": 2.6606541440848033,
      "language_loss": 0.72225463,
      "learning_rate": 3.891647795197523e-06,
      "loss": 0.74211884,
      "num_input_tokens_seen": 23376910,
      "step": 1099,
      "time_per_iteration": 2.6745107173919678
    },
    {
      "auxiliary_loss_clip": 0.01223684,
      "auxiliary_loss_mlp": 0.01045046,
      "balance_loss_clip": 1.06508982,
      "balance_loss_mlp": 1.0328151,
      "epoch": 0.13226717970300006,
      "flos": 19353840940800.0,
      "grad_norm": 2.373901878756309,
      "language_loss": 0.6880607,
      "learning_rate": 3.8913947351459605e-06,
      "loss": 0.71074796,
      "num_input_tokens_seen": 23394450,
      "step": 1100,
      "time_per_iteration": 2.7319705486297607
    },
    {
      "auxiliary_loss_clip": 0.01223903,
      "auxiliary_loss_mlp": 0.01046371,
      "balance_loss_clip": 1.06361616,
      "balance_loss_mlp": 1.03528404,
      "epoch": 0.13238742259363914,
      "flos": 20698084546560.0,
      "grad_norm": 1.867104295054864,
      "language_loss": 0.67572409,
      "learning_rate": 3.89114138817132e-06,
      "loss": 0.69842678,
      "num_input_tokens_seen": 23411115,
      "step": 1101,
      "time_per_iteration": 2.6461668014526367
    },
    {
      "auxiliary_loss_clip": 0.01185639,
      "auxiliary_loss_mlp": 0.00764932,
      "balance_loss_clip": 1.05842483,
      "balance_loss_mlp": 1.00123429,
      "epoch": 0.13250766548427825,
      "flos": 21032449274880.0,
      "grad_norm": 2.1112349241291466,
      "language_loss": 0.84419298,
      "learning_rate": 3.890887754312035e-06,
      "loss": 0.86369872,
      "num_input_tokens_seen": 23429360,
      "step": 1102,
      "time_per_iteration": 2.783324956893921
    },
    {
      "auxiliary_loss_clip": 0.01163343,
      "auxiliary_loss_mlp": 0.01045566,
      "balance_loss_clip": 1.06205606,
      "balance_loss_mlp": 1.03474808,
      "epoch": 0.13262790837491734,
      "flos": 22637871648000.0,
      "grad_norm": 1.9871143073263176,
      "language_loss": 0.87586898,
      "learning_rate": 3.890633833606581e-06,
      "loss": 0.8979581,
      "num_input_tokens_seen": 23449050,
      "step": 1103,
      "time_per_iteration": 2.811296224594116
    },
    {
      "auxiliary_loss_clip": 0.01170461,
      "auxiliary_loss_mlp": 0.01040786,
      "balance_loss_clip": 1.06138253,
      "balance_loss_mlp": 1.0297296,
      "epoch": 0.13274815126555642,
      "flos": 19683141851520.0,
      "grad_norm": 1.7761003854766266,
      "language_loss": 0.69239008,
      "learning_rate": 3.890379626093477e-06,
      "loss": 0.71450257,
      "num_input_tokens_seen": 23468800,
      "step": 1104,
      "time_per_iteration": 2.7751588821411133
    },
    {
      "auxiliary_loss_clip": 0.01180811,
      "auxiliary_loss_mlp": 0.01040409,
      "balance_loss_clip": 1.06242681,
      "balance_loss_mlp": 1.02840471,
      "epoch": 0.1328683941561955,
      "flos": 21317687176320.0,
      "grad_norm": 2.185997457475326,
      "language_loss": 0.92428094,
      "learning_rate": 3.890125131811287e-06,
      "loss": 0.94649315,
      "num_input_tokens_seen": 23486850,
      "step": 1105,
      "time_per_iteration": 2.772221803665161
    },
    {
      "auxiliary_loss_clip": 0.01188923,
      "auxiliary_loss_mlp": 0.01041675,
      "balance_loss_clip": 1.05923915,
      "balance_loss_mlp": 1.03063655,
      "epoch": 0.1329886370468346,
      "flos": 13699131580800.0,
      "grad_norm": 2.7728424849430056,
      "language_loss": 0.75388938,
      "learning_rate": 3.889870350798618e-06,
      "loss": 0.77619529,
      "num_input_tokens_seen": 23504195,
      "step": 1106,
      "time_per_iteration": 3.676539421081543
    },
    {
      "auxiliary_loss_clip": 0.01200215,
      "auxiliary_loss_mlp": 0.01048522,
      "balance_loss_clip": 1.06078374,
      "balance_loss_mlp": 1.0374589,
      "epoch": 0.1331088799374737,
      "flos": 21032413361280.0,
      "grad_norm": 1.7558647843531117,
      "language_loss": 0.78323424,
      "learning_rate": 3.889615283094119e-06,
      "loss": 0.80572164,
      "num_input_tokens_seen": 23523385,
      "step": 1107,
      "time_per_iteration": 2.7113358974456787
    },
    {
      "auxiliary_loss_clip": 0.01205028,
      "auxiliary_loss_mlp": 0.01043249,
      "balance_loss_clip": 1.06166172,
      "balance_loss_mlp": 1.03086877,
      "epoch": 0.13322912282811278,
      "flos": 18260432985600.0,
      "grad_norm": 4.1797351471617405,
      "language_loss": 0.84383333,
      "learning_rate": 3.889359928736485e-06,
      "loss": 0.86631608,
      "num_input_tokens_seen": 23541330,
      "step": 1108,
      "time_per_iteration": 3.7933428287506104
    },
    {
      "auxiliary_loss_clip": 0.0121286,
      "auxiliary_loss_mlp": 0.00765163,
      "balance_loss_clip": 1.06358862,
      "balance_loss_mlp": 1.00110817,
      "epoch": 0.1333493657187519,
      "flos": 24460876656000.0,
      "grad_norm": 2.107340536703627,
      "language_loss": 0.90960145,
      "learning_rate": 3.889104287764451e-06,
      "loss": 0.92938167,
      "num_input_tokens_seen": 23561705,
      "step": 1109,
      "time_per_iteration": 2.7933857440948486
    },
    {
      "auxiliary_loss_clip": 0.01202007,
      "auxiliary_loss_mlp": 0.01039151,
      "balance_loss_clip": 1.06250846,
      "balance_loss_mlp": 1.02748048,
      "epoch": 0.13346960860939097,
      "flos": 22158930677760.0,
      "grad_norm": 2.8900282985972003,
      "language_loss": 0.90784359,
      "learning_rate": 3.888848360216798e-06,
      "loss": 0.93025517,
      "num_input_tokens_seen": 23579350,
      "step": 1110,
      "time_per_iteration": 2.7681949138641357
    },
    {
      "auxiliary_loss_clip": 0.0111882,
      "auxiliary_loss_mlp": 0.01008887,
      "balance_loss_clip": 1.03250909,
      "balance_loss_mlp": 1.00500059,
      "epoch": 0.13358985150003005,
      "flos": 67931212608000.0,
      "grad_norm": 0.8006360094683375,
      "language_loss": 0.56616402,
      "learning_rate": 3.888592146132351e-06,
      "loss": 0.58744109,
      "num_input_tokens_seen": 23640620,
      "step": 1111,
      "time_per_iteration": 4.310925006866455
    },
    {
      "auxiliary_loss_clip": 0.01230739,
      "auxiliary_loss_mlp": 0.01041958,
      "balance_loss_clip": 1.06718934,
      "balance_loss_mlp": 1.03109169,
      "epoch": 0.13371009439066917,
      "flos": 26834284742400.0,
      "grad_norm": 7.64438779175482,
      "language_loss": 0.7823835,
      "learning_rate": 3.888335645549978e-06,
      "loss": 0.80511045,
      "num_input_tokens_seen": 23661040,
      "step": 1112,
      "time_per_iteration": 2.714881420135498
    },
    {
      "auxiliary_loss_clip": 0.01197679,
      "auxiliary_loss_mlp": 0.01037266,
      "balance_loss_clip": 1.05977976,
      "balance_loss_mlp": 1.02624488,
      "epoch": 0.13383033728130825,
      "flos": 26322844942080.0,
      "grad_norm": 2.0775521799430834,
      "language_loss": 0.81707776,
      "learning_rate": 3.888078858508588e-06,
      "loss": 0.83942729,
      "num_input_tokens_seen": 23680900,
      "step": 1113,
      "time_per_iteration": 3.6895885467529297
    },
    {
      "auxiliary_loss_clip": 0.01211612,
      "auxiliary_loss_mlp": 0.01034134,
      "balance_loss_clip": 1.06623864,
      "balance_loss_mlp": 1.02298152,
      "epoch": 0.13395058017194733,
      "flos": 22563931501440.0,
      "grad_norm": 3.5206316147741576,
      "language_loss": 0.84434539,
      "learning_rate": 3.8878217850471365e-06,
      "loss": 0.86680281,
      "num_input_tokens_seen": 23700815,
      "step": 1114,
      "time_per_iteration": 2.7035093307495117
    },
    {
      "auxiliary_loss_clip": 0.01208648,
      "auxiliary_loss_mlp": 0.01053086,
      "balance_loss_clip": 1.06685472,
      "balance_loss_mlp": 1.04102182,
      "epoch": 0.13407082306258641,
      "flos": 25810938264960.0,
      "grad_norm": 1.9629969207221107,
      "language_loss": 0.74131012,
      "learning_rate": 3.887564425204621e-06,
      "loss": 0.76392752,
      "num_input_tokens_seen": 23722500,
      "step": 1115,
      "time_per_iteration": 2.8134453296661377
    },
    {
      "auxiliary_loss_clip": 0.01098553,
      "auxiliary_loss_mlp": 0.01004559,
      "balance_loss_clip": 1.0313133,
      "balance_loss_mlp": 1.00114918,
      "epoch": 0.13419106595322552,
      "flos": 68338365269760.0,
      "grad_norm": 0.847679360680931,
      "language_loss": 0.54618484,
      "learning_rate": 3.887306779020083e-06,
      "loss": 0.56721592,
      "num_input_tokens_seen": 23777155,
      "step": 1116,
      "time_per_iteration": 3.19225811958313
    },
    {
      "auxiliary_loss_clip": 0.0115347,
      "auxiliary_loss_mlp": 0.01042534,
      "balance_loss_clip": 1.05728865,
      "balance_loss_mlp": 1.03184068,
      "epoch": 0.1343113088438646,
      "flos": 20449080489600.0,
      "grad_norm": 3.7951222200986474,
      "language_loss": 0.70403683,
      "learning_rate": 3.887048846532608e-06,
      "loss": 0.72599685,
      "num_input_tokens_seen": 23794130,
      "step": 1117,
      "time_per_iteration": 2.792240619659424
    },
    {
      "auxiliary_loss_clip": 0.01149111,
      "auxiliary_loss_mlp": 0.0100446,
      "balance_loss_clip": 1.03605199,
      "balance_loss_mlp": 1.00100279,
      "epoch": 0.1344315517345037,
      "flos": 67389784951680.0,
      "grad_norm": 0.7642339731495188,
      "language_loss": 0.58101034,
      "learning_rate": 3.8867906277813224e-06,
      "loss": 0.60254598,
      "num_input_tokens_seen": 23852285,
      "step": 1118,
      "time_per_iteration": 3.0471861362457275
    },
    {
      "auxiliary_loss_clip": 0.01243341,
      "auxiliary_loss_mlp": 0.01041752,
      "balance_loss_clip": 1.0697422,
      "balance_loss_mlp": 1.03025424,
      "epoch": 0.1345517946251428,
      "flos": 40734442788480.0,
      "grad_norm": 2.028838584607625,
      "language_loss": 0.73429513,
      "learning_rate": 3.886532122805399e-06,
      "loss": 0.757146,
      "num_input_tokens_seen": 23874765,
      "step": 1119,
      "time_per_iteration": 2.869540214538574
    },
    {
      "auxiliary_loss_clip": 0.01178897,
      "auxiliary_loss_mlp": 0.01049571,
      "balance_loss_clip": 1.06185615,
      "balance_loss_mlp": 1.03791881,
      "epoch": 0.13467203751578188,
      "flos": 22816850140800.0,
      "grad_norm": 2.166758141711714,
      "language_loss": 0.89982432,
      "learning_rate": 3.886273331644053e-06,
      "loss": 0.92210901,
      "num_input_tokens_seen": 23893635,
      "step": 1120,
      "time_per_iteration": 2.84761905670166
    },
    {
      "auxiliary_loss_clip": 0.01208801,
      "auxiliary_loss_mlp": 0.01041736,
      "balance_loss_clip": 1.05939722,
      "balance_loss_mlp": 1.03057241,
      "epoch": 0.13479228040642097,
      "flos": 17091576512640.0,
      "grad_norm": 1.9662963209139768,
      "language_loss": 0.8228749,
      "learning_rate": 3.886014254336542e-06,
      "loss": 0.84538031,
      "num_input_tokens_seen": 23910110,
      "step": 1121,
      "time_per_iteration": 2.6620750427246094
    },
    {
      "auxiliary_loss_clip": 0.01219202,
      "auxiliary_loss_mlp": 0.01041171,
      "balance_loss_clip": 1.06178665,
      "balance_loss_mlp": 1.02979219,
      "epoch": 0.13491252329706005,
      "flos": 23730525417600.0,
      "grad_norm": 1.7891352802161389,
      "language_loss": 0.92629158,
      "learning_rate": 3.885754890922168e-06,
      "loss": 0.94889534,
      "num_input_tokens_seen": 23930440,
      "step": 1122,
      "time_per_iteration": 2.7974765300750732
    },
    {
      "auxiliary_loss_clip": 0.01196557,
      "auxiliary_loss_mlp": 0.01046095,
      "balance_loss_clip": 1.06522512,
      "balance_loss_mlp": 1.03373921,
      "epoch": 0.13503276618769916,
      "flos": 34127058960000.0,
      "grad_norm": 2.091161359692645,
      "language_loss": 0.78821927,
      "learning_rate": 3.885495241440277e-06,
      "loss": 0.81064582,
      "num_input_tokens_seen": 23954535,
      "step": 1123,
      "time_per_iteration": 2.8508238792419434
    },
    {
      "auxiliary_loss_clip": 0.01244194,
      "auxiliary_loss_mlp": 0.01038055,
      "balance_loss_clip": 1.07020116,
      "balance_loss_mlp": 1.0272541,
      "epoch": 0.13515300907833824,
      "flos": 17712328377600.0,
      "grad_norm": 1.7668494458010335,
      "language_loss": 0.74617088,
      "learning_rate": 3.885235305930257e-06,
      "loss": 0.76899338,
      "num_input_tokens_seen": 23972735,
      "step": 1124,
      "time_per_iteration": 2.6256320476531982
    },
    {
      "auxiliary_loss_clip": 0.01204639,
      "auxiliary_loss_mlp": 0.01051878,
      "balance_loss_clip": 1.06890094,
      "balance_loss_mlp": 1.0399158,
      "epoch": 0.13527325196897733,
      "flos": 20260872201600.0,
      "grad_norm": 1.8430415306790155,
      "language_loss": 0.85171545,
      "learning_rate": 3.884975084431539e-06,
      "loss": 0.87428069,
      "num_input_tokens_seen": 23987685,
      "step": 1125,
      "time_per_iteration": 2.7700865268707275
    },
    {
      "auxiliary_loss_clip": 0.01168808,
      "auxiliary_loss_mlp": 0.01040941,
      "balance_loss_clip": 1.05591667,
      "balance_loss_mlp": 1.02971792,
      "epoch": 0.13539349485961644,
      "flos": 18186492839040.0,
      "grad_norm": 7.526129752019831,
      "language_loss": 0.91690755,
      "learning_rate": 3.8847145769836e-06,
      "loss": 0.93900508,
      "num_input_tokens_seen": 24004105,
      "step": 1126,
      "time_per_iteration": 2.7940120697021484
    },
    {
      "auxiliary_loss_clip": 0.01236104,
      "auxiliary_loss_mlp": 0.01057045,
      "balance_loss_clip": 1.07119846,
      "balance_loss_mlp": 1.04511797,
      "epoch": 0.13551373775025552,
      "flos": 19317463441920.0,
      "grad_norm": 2.2241961771161325,
      "language_loss": 0.66526198,
      "learning_rate": 3.884453783625959e-06,
      "loss": 0.68819344,
      "num_input_tokens_seen": 24021715,
      "step": 1127,
      "time_per_iteration": 2.6973068714141846
    },
    {
      "auxiliary_loss_clip": 0.0120276,
      "auxiliary_loss_mlp": 0.01058275,
      "balance_loss_clip": 1.06328464,
      "balance_loss_mlp": 1.0464915,
      "epoch": 0.1356339806408946,
      "flos": 20850813175680.0,
      "grad_norm": 2.488227706801925,
      "language_loss": 0.84846163,
      "learning_rate": 3.884192704398176e-06,
      "loss": 0.87107193,
      "num_input_tokens_seen": 24038915,
      "step": 1128,
      "time_per_iteration": 2.7340943813323975
    },
    {
      "auxiliary_loss_clip": 0.01229372,
      "auxiliary_loss_mlp": 0.01038942,
      "balance_loss_clip": 1.06922913,
      "balance_loss_mlp": 1.02776599,
      "epoch": 0.13575422353153369,
      "flos": 50476037696640.0,
      "grad_norm": 1.8893939669019681,
      "language_loss": 0.74542022,
      "learning_rate": 3.883931339339858e-06,
      "loss": 0.76810336,
      "num_input_tokens_seen": 24063300,
      "step": 1129,
      "time_per_iteration": 2.949233055114746
    },
    {
      "auxiliary_loss_clip": 0.01236644,
      "auxiliary_loss_mlp": 0.00765472,
      "balance_loss_clip": 1.07010615,
      "balance_loss_mlp": 1.00089049,
      "epoch": 0.1358744664221728,
      "flos": 18150797698560.0,
      "grad_norm": 1.8594751347054401,
      "language_loss": 0.788836,
      "learning_rate": 3.883669688490654e-06,
      "loss": 0.8088572,
      "num_input_tokens_seen": 24081070,
      "step": 1130,
      "time_per_iteration": 2.6728477478027344
    },
    {
      "auxiliary_loss_clip": 0.01133224,
      "auxiliary_loss_mlp": 0.01048164,
      "balance_loss_clip": 1.04881847,
      "balance_loss_mlp": 1.03760195,
      "epoch": 0.13599470931281188,
      "flos": 18442966924800.0,
      "grad_norm": 4.398490714730146,
      "language_loss": 0.85566384,
      "learning_rate": 3.883407751890256e-06,
      "loss": 0.87747777,
      "num_input_tokens_seen": 24099675,
      "step": 1131,
      "time_per_iteration": 2.968536853790283
    },
    {
      "auxiliary_loss_clip": 0.01198365,
      "auxiliary_loss_mlp": 0.0105113,
      "balance_loss_clip": 1.06370068,
      "balance_loss_mlp": 1.03931665,
      "epoch": 0.13611495220345096,
      "flos": 26680766014080.0,
      "grad_norm": 2.2715169637358876,
      "language_loss": 0.85853726,
      "learning_rate": 3.8831455295783994e-06,
      "loss": 0.88103223,
      "num_input_tokens_seen": 24118925,
      "step": 1132,
      "time_per_iteration": 4.167387247085571
    },
    {
      "auxiliary_loss_clip": 0.01166343,
      "auxiliary_loss_mlp": 0.0104199,
      "balance_loss_clip": 1.06262505,
      "balance_loss_mlp": 1.03108883,
      "epoch": 0.13623519509409007,
      "flos": 21686238673920.0,
      "grad_norm": 2.176995642279193,
      "language_loss": 0.74184859,
      "learning_rate": 3.882883021594864e-06,
      "loss": 0.76393193,
      "num_input_tokens_seen": 24137065,
      "step": 1133,
      "time_per_iteration": 2.8417611122131348
    },
    {
      "auxiliary_loss_clip": 0.01215152,
      "auxiliary_loss_mlp": 0.01054828,
      "balance_loss_clip": 1.065835,
      "balance_loss_mlp": 1.04326451,
      "epoch": 0.13635543798472916,
      "flos": 14830389492480.0,
      "grad_norm": 2.24371208511187,
      "language_loss": 0.86927003,
      "learning_rate": 3.8826202279794705e-06,
      "loss": 0.8919698,
      "num_input_tokens_seen": 24154125,
      "step": 1134,
      "time_per_iteration": 3.678713083267212
    },
    {
      "auxiliary_loss_clip": 0.01170455,
      "auxiliary_loss_mlp": 0.00764727,
      "balance_loss_clip": 1.06052208,
      "balance_loss_mlp": 1.00080657,
      "epoch": 0.13647568087536824,
      "flos": 22890323410560.0,
      "grad_norm": 2.009144842036835,
      "language_loss": 0.70630002,
      "learning_rate": 3.882357148772085e-06,
      "loss": 0.72565186,
      "num_input_tokens_seen": 24171550,
      "step": 1135,
      "time_per_iteration": 2.8336493968963623
    },
    {
      "auxiliary_loss_clip": 0.01242161,
      "auxiliary_loss_mlp": 0.01041733,
      "balance_loss_clip": 1.07044351,
      "balance_loss_mlp": 1.03006887,
      "epoch": 0.13659592376600732,
      "flos": 19937927998080.0,
      "grad_norm": 2.808519219159115,
      "language_loss": 0.84430331,
      "learning_rate": 3.882093784012617e-06,
      "loss": 0.8671422,
      "num_input_tokens_seen": 24190190,
      "step": 1136,
      "time_per_iteration": 2.626735210418701
    },
    {
      "auxiliary_loss_clip": 0.01213515,
      "auxiliary_loss_mlp": 0.01048234,
      "balance_loss_clip": 1.06736004,
      "balance_loss_mlp": 1.03670096,
      "epoch": 0.13671616665664643,
      "flos": 21428579439360.0,
      "grad_norm": 2.130001189933125,
      "language_loss": 0.84163189,
      "learning_rate": 3.881830133741019e-06,
      "loss": 0.86424941,
      "num_input_tokens_seen": 24209055,
      "step": 1137,
      "time_per_iteration": 3.7023186683654785
    },
    {
      "auxiliary_loss_clip": 0.01196627,
      "auxiliary_loss_mlp": 0.01037659,
      "balance_loss_clip": 1.06670511,
      "balance_loss_mlp": 1.02610123,
      "epoch": 0.13683640954728551,
      "flos": 22778138257920.0,
      "grad_norm": 3.013918165615462,
      "language_loss": 0.76192248,
      "learning_rate": 3.881566197997285e-06,
      "loss": 0.78426528,
      "num_input_tokens_seen": 24225490,
      "step": 1138,
      "time_per_iteration": 2.710286855697632
    },
    {
      "auxiliary_loss_clip": 0.01194533,
      "auxiliary_loss_mlp": 0.01046426,
      "balance_loss_clip": 1.06202388,
      "balance_loss_mlp": 1.03607869,
      "epoch": 0.1369566524379246,
      "flos": 21725884310400.0,
      "grad_norm": 1.597118288140642,
      "language_loss": 0.74834788,
      "learning_rate": 3.881301976821456e-06,
      "loss": 0.77075744,
      "num_input_tokens_seen": 24245520,
      "step": 1139,
      "time_per_iteration": 3.6633102893829346
    },
    {
      "auxiliary_loss_clip": 0.01192037,
      "auxiliary_loss_mlp": 0.01038271,
      "balance_loss_clip": 1.06063199,
      "balance_loss_mlp": 1.02778029,
      "epoch": 0.1370768953285637,
      "flos": 18624459369600.0,
      "grad_norm": 2.2979870724536915,
      "language_loss": 0.90881705,
      "learning_rate": 3.881037470253612e-06,
      "loss": 0.93112016,
      "num_input_tokens_seen": 24265035,
      "step": 1140,
      "time_per_iteration": 2.713775157928467
    },
    {
      "auxiliary_loss_clip": 0.01209906,
      "auxiliary_loss_mlp": 0.01036288,
      "balance_loss_clip": 1.06353331,
      "balance_loss_mlp": 1.02560091,
      "epoch": 0.1371971382192028,
      "flos": 14939521989120.0,
      "grad_norm": 2.915147936175038,
      "language_loss": 0.79610717,
      "learning_rate": 3.88077267833388e-06,
      "loss": 0.81856906,
      "num_input_tokens_seen": 24281550,
      "step": 1141,
      "time_per_iteration": 2.6730215549468994
    },
    {
      "auxiliary_loss_clip": 0.01189782,
      "auxiliary_loss_mlp": 0.00764411,
      "balance_loss_clip": 1.06156468,
      "balance_loss_mlp": 1.00081265,
      "epoch": 0.13731738110984187,
      "flos": 19023785844480.0,
      "grad_norm": 2.3477601661415277,
      "language_loss": 0.84000373,
      "learning_rate": 3.880507601102427e-06,
      "loss": 0.85954565,
      "num_input_tokens_seen": 24299485,
      "step": 1142,
      "time_per_iteration": 2.768481969833374
    },
    {
      "auxiliary_loss_clip": 0.01208605,
      "auxiliary_loss_mlp": 0.01043672,
      "balance_loss_clip": 1.06681585,
      "balance_loss_mlp": 1.03293705,
      "epoch": 0.13743762400048098,
      "flos": 18187462506240.0,
      "grad_norm": 1.8405018606085009,
      "language_loss": 0.82342553,
      "learning_rate": 3.880242238599467e-06,
      "loss": 0.84594822,
      "num_input_tokens_seen": 24316010,
      "step": 1143,
      "time_per_iteration": 2.6885857582092285
    },
    {
      "auxiliary_loss_clip": 0.0117594,
      "auxiliary_loss_mlp": 0.01048297,
      "balance_loss_clip": 1.05820847,
      "balance_loss_mlp": 1.03706193,
      "epoch": 0.13755786689112007,
      "flos": 21031982398080.0,
      "grad_norm": 1.915707054207543,
      "language_loss": 0.83318341,
      "learning_rate": 3.879976590865254e-06,
      "loss": 0.85542572,
      "num_input_tokens_seen": 24335465,
      "step": 1144,
      "time_per_iteration": 2.810774087905884
    },
    {
      "auxiliary_loss_clip": 0.0121994,
      "auxiliary_loss_mlp": 0.01044947,
      "balance_loss_clip": 1.06436014,
      "balance_loss_mlp": 1.03408146,
      "epoch": 0.13767810978175915,
      "flos": 21360636864000.0,
      "grad_norm": 2.0254604443079747,
      "language_loss": 0.87754118,
      "learning_rate": 3.879710657940087e-06,
      "loss": 0.90019006,
      "num_input_tokens_seen": 24354415,
      "step": 1145,
      "time_per_iteration": 2.6654748916625977
    },
    {
      "auxiliary_loss_clip": 0.01156735,
      "auxiliary_loss_mlp": 0.01037632,
      "balance_loss_clip": 1.05533671,
      "balance_loss_mlp": 1.02569342,
      "epoch": 0.13779835267239823,
      "flos": 30592084861440.0,
      "grad_norm": 2.2032396230240763,
      "language_loss": 0.70163596,
      "learning_rate": 3.879444439864308e-06,
      "loss": 0.72357965,
      "num_input_tokens_seen": 24373990,
      "step": 1146,
      "time_per_iteration": 2.862295389175415
    },
    {
      "auxiliary_loss_clip": 0.01243921,
      "auxiliary_loss_mlp": 0.01044288,
      "balance_loss_clip": 1.07247925,
      "balance_loss_mlp": 1.03328514,
      "epoch": 0.13791859556303734,
      "flos": 22669867687680.0,
      "grad_norm": 1.8902736777028906,
      "language_loss": 0.85635459,
      "learning_rate": 3.879177936678301e-06,
      "loss": 0.87923664,
      "num_input_tokens_seen": 24392995,
      "step": 1147,
      "time_per_iteration": 2.7191474437713623
    },
    {
      "auxiliary_loss_clip": 0.01194978,
      "auxiliary_loss_mlp": 0.01038449,
      "balance_loss_clip": 1.06054521,
      "balance_loss_mlp": 1.02763712,
      "epoch": 0.13803883845367643,
      "flos": 35224166016000.0,
      "grad_norm": 2.027706335550698,
      "language_loss": 0.7713846,
      "learning_rate": 3.878911148422496e-06,
      "loss": 0.79371881,
      "num_input_tokens_seen": 24414470,
      "step": 1148,
      "time_per_iteration": 2.8767576217651367
    },
    {
      "auxiliary_loss_clip": 0.0124019,
      "auxiliary_loss_mlp": 0.01040121,
      "balance_loss_clip": 1.06781828,
      "balance_loss_mlp": 1.02920139,
      "epoch": 0.1381590813443155,
      "flos": 32014542332160.0,
      "grad_norm": 2.318298978788916,
      "language_loss": 0.70295769,
      "learning_rate": 3.878644075137364e-06,
      "loss": 0.72576082,
      "num_input_tokens_seen": 24435120,
      "step": 1149,
      "time_per_iteration": 2.7657811641693115
    },
    {
      "auxiliary_loss_clip": 0.01190399,
      "auxiliary_loss_mlp": 0.0104157,
      "balance_loss_clip": 1.06458569,
      "balance_loss_mlp": 1.03144932,
      "epoch": 0.13827932423495462,
      "flos": 17821855923840.0,
      "grad_norm": 2.088426942435251,
      "language_loss": 0.79490161,
      "learning_rate": 3.878376716863418e-06,
      "loss": 0.81722134,
      "num_input_tokens_seen": 24451420,
      "step": 1150,
      "time_per_iteration": 2.723644971847534
    },
    {
      "auxiliary_loss_clip": 0.01175368,
      "auxiliary_loss_mlp": 0.01035402,
      "balance_loss_clip": 1.0584271,
      "balance_loss_mlp": 1.02359402,
      "epoch": 0.1383995671255937,
      "flos": 19427098728960.0,
      "grad_norm": 2.0751127652622245,
      "language_loss": 0.71269315,
      "learning_rate": 3.878109073641219e-06,
      "loss": 0.73480093,
      "num_input_tokens_seen": 24470450,
      "step": 1151,
      "time_per_iteration": 2.81508207321167
    },
    {
      "auxiliary_loss_clip": 0.01189585,
      "auxiliary_loss_mlp": 0.01040629,
      "balance_loss_clip": 1.06223214,
      "balance_loss_mlp": 1.02917242,
      "epoch": 0.13851981001623279,
      "flos": 28296603331200.0,
      "grad_norm": 1.733612162520954,
      "language_loss": 0.8143059,
      "learning_rate": 3.877841145511366e-06,
      "loss": 0.83660805,
      "num_input_tokens_seen": 24493190,
      "step": 1152,
      "time_per_iteration": 2.8183603286743164
    },
    {
      "auxiliary_loss_clip": 0.01197762,
      "auxiliary_loss_mlp": 0.01045614,
      "balance_loss_clip": 1.06616807,
      "balance_loss_mlp": 1.03394914,
      "epoch": 0.13864005290687187,
      "flos": 21213079793280.0,
      "grad_norm": 1.7593643272536545,
      "language_loss": 0.83085406,
      "learning_rate": 3.8775729325145035e-06,
      "loss": 0.85328782,
      "num_input_tokens_seen": 24512425,
      "step": 1153,
      "time_per_iteration": 2.757657051086426
    },
    {
      "auxiliary_loss_clip": 0.01089259,
      "auxiliary_loss_mlp": 0.01013399,
      "balance_loss_clip": 1.02821219,
      "balance_loss_mlp": 1.00951231,
      "epoch": 0.13876029579751098,
      "flos": 71653389413760.0,
      "grad_norm": 0.8064152467419025,
      "language_loss": 0.64749616,
      "learning_rate": 3.877304434691321e-06,
      "loss": 0.66852272,
      "num_input_tokens_seen": 24579275,
      "step": 1154,
      "time_per_iteration": 3.542799234390259
    },
    {
      "auxiliary_loss_clip": 0.012016,
      "auxiliary_loss_mlp": 0.01041892,
      "balance_loss_clip": 1.06656742,
      "balance_loss_mlp": 1.03107357,
      "epoch": 0.13888053868815006,
      "flos": 21941348042880.0,
      "grad_norm": 1.8066798283935495,
      "language_loss": 0.79694265,
      "learning_rate": 3.877035652082548e-06,
      "loss": 0.81937754,
      "num_input_tokens_seen": 24598720,
      "step": 1155,
      "time_per_iteration": 2.70920729637146
    },
    {
      "auxiliary_loss_clip": 0.0118245,
      "auxiliary_loss_mlp": 0.01038742,
      "balance_loss_clip": 1.06123459,
      "balance_loss_mlp": 1.02708328,
      "epoch": 0.13900078157878915,
      "flos": 19608627087360.0,
      "grad_norm": 1.9149652779565671,
      "language_loss": 0.85055828,
      "learning_rate": 3.87676658472896e-06,
      "loss": 0.87277019,
      "num_input_tokens_seen": 24617530,
      "step": 1156,
      "time_per_iteration": 2.8032336235046387
    },
    {
      "auxiliary_loss_clip": 0.01213461,
      "auxiliary_loss_mlp": 0.01044092,
      "balance_loss_clip": 1.06643152,
      "balance_loss_mlp": 1.03205156,
      "epoch": 0.13912102446942826,
      "flos": 22638051216000.0,
      "grad_norm": 2.476058542578211,
      "language_loss": 0.85448468,
      "learning_rate": 3.876497232671372e-06,
      "loss": 0.87706017,
      "num_input_tokens_seen": 24637485,
      "step": 1157,
      "time_per_iteration": 2.723301410675049
    },
    {
      "auxiliary_loss_clip": 0.01180643,
      "auxiliary_loss_mlp": 0.01042524,
      "balance_loss_clip": 1.06152749,
      "balance_loss_mlp": 1.03109229,
      "epoch": 0.13924126736006734,
      "flos": 29643324975360.0,
      "grad_norm": 2.1546281178377655,
      "language_loss": 0.83692217,
      "learning_rate": 3.876227595950647e-06,
      "loss": 0.85915387,
      "num_input_tokens_seen": 24656915,
      "step": 1158,
      "time_per_iteration": 3.7182774543762207
    },
    {
      "auxiliary_loss_clip": 0.01166152,
      "auxiliary_loss_mlp": 0.01046306,
      "balance_loss_clip": 1.05490041,
      "balance_loss_mlp": 1.0353384,
      "epoch": 0.13936151025070642,
      "flos": 27417653527680.0,
      "grad_norm": 1.7516174528790087,
      "language_loss": 0.79001486,
      "learning_rate": 3.875957674607686e-06,
      "loss": 0.81213945,
      "num_input_tokens_seen": 24679190,
      "step": 1159,
      "time_per_iteration": 2.8821070194244385
    },
    {
      "auxiliary_loss_clip": 0.01238225,
      "auxiliary_loss_mlp": 0.01039008,
      "balance_loss_clip": 1.0680865,
      "balance_loss_mlp": 1.02671123,
      "epoch": 0.1394817531413455,
      "flos": 16399326625920.0,
      "grad_norm": 2.168499512922471,
      "language_loss": 0.88021779,
      "learning_rate": 3.8756874686834386e-06,
      "loss": 0.9029901,
      "num_input_tokens_seen": 24697405,
      "step": 1160,
      "time_per_iteration": 3.572321891784668
    },
    {
      "auxiliary_loss_clip": 0.01226938,
      "auxiliary_loss_mlp": 0.00765662,
      "balance_loss_clip": 1.06416762,
      "balance_loss_mlp": 1.00070357,
      "epoch": 0.13960199603198462,
      "flos": 30922319525760.0,
      "grad_norm": 1.639409405448247,
      "language_loss": 0.80495965,
      "learning_rate": 3.875416978218893e-06,
      "loss": 0.82488561,
      "num_input_tokens_seen": 24720600,
      "step": 1161,
      "time_per_iteration": 2.8354623317718506
    },
    {
      "auxiliary_loss_clip": 0.01213503,
      "auxiliary_loss_mlp": 0.01043679,
      "balance_loss_clip": 1.06616247,
      "balance_loss_mlp": 1.0313468,
      "epoch": 0.1397222389226237,
      "flos": 18113773754880.0,
      "grad_norm": 1.9694655123817268,
      "language_loss": 0.8258037,
      "learning_rate": 3.8751462032550835e-06,
      "loss": 0.84837544,
      "num_input_tokens_seen": 24737605,
      "step": 1162,
      "time_per_iteration": 3.6149728298187256
    },
    {
      "auxiliary_loss_clip": 0.0117636,
      "auxiliary_loss_mlp": 0.01047173,
      "balance_loss_clip": 1.05941153,
      "balance_loss_mlp": 1.03668237,
      "epoch": 0.13984248181326278,
      "flos": 16872772815360.0,
      "grad_norm": 2.552432061153091,
      "language_loss": 0.82932079,
      "learning_rate": 3.874875143833085e-06,
      "loss": 0.85155606,
      "num_input_tokens_seen": 24755845,
      "step": 1163,
      "time_per_iteration": 2.723212242126465
    },
    {
      "auxiliary_loss_clip": 0.01158613,
      "auxiliary_loss_mlp": 0.0076532,
      "balance_loss_clip": 1.05270672,
      "balance_loss_mlp": 1.00061893,
      "epoch": 0.1399627247039019,
      "flos": 54121401267840.0,
      "grad_norm": 2.149472405512119,
      "language_loss": 0.68951905,
      "learning_rate": 3.874603799994019e-06,
      "loss": 0.70875841,
      "num_input_tokens_seen": 24779380,
      "step": 1164,
      "time_per_iteration": 3.0918073654174805
    },
    {
      "auxiliary_loss_clip": 0.01216706,
      "auxiliary_loss_mlp": 0.01048994,
      "balance_loss_clip": 1.06820345,
      "balance_loss_mlp": 1.03794336,
      "epoch": 0.14008296759454097,
      "flos": 11765521618560.0,
      "grad_norm": 2.2712218267304127,
      "language_loss": 0.87047911,
      "learning_rate": 3.874332171779046e-06,
      "loss": 0.89313614,
      "num_input_tokens_seen": 24794260,
      "step": 1165,
      "time_per_iteration": 3.5794975757598877
    },
    {
      "auxiliary_loss_clip": 0.01221359,
      "auxiliary_loss_mlp": 0.01045551,
      "balance_loss_clip": 1.061692,
      "balance_loss_mlp": 1.0342257,
      "epoch": 0.14020321048518006,
      "flos": 22017514832640.0,
      "grad_norm": 1.710334877182831,
      "language_loss": 0.75866526,
      "learning_rate": 3.874060259229373e-06,
      "loss": 0.7813344,
      "num_input_tokens_seen": 24815835,
      "step": 1166,
      "time_per_iteration": 2.749516725540161
    },
    {
      "auxiliary_loss_clip": 0.0118695,
      "auxiliary_loss_mlp": 0.01046186,
      "balance_loss_clip": 1.06572723,
      "balance_loss_mlp": 1.03461671,
      "epoch": 0.14032345337581917,
      "flos": 23404313076480.0,
      "grad_norm": 5.454348693490914,
      "language_loss": 0.93630487,
      "learning_rate": 3.873788062386249e-06,
      "loss": 0.95863622,
      "num_input_tokens_seen": 24834095,
      "step": 1167,
      "time_per_iteration": 2.8020496368408203
    },
    {
      "auxiliary_loss_clip": 0.01245953,
      "auxiliary_loss_mlp": 0.01046988,
      "balance_loss_clip": 1.07267606,
      "balance_loss_mlp": 1.03548408,
      "epoch": 0.14044369626645825,
      "flos": 29645767100160.0,
      "grad_norm": 1.7569167604531435,
      "language_loss": 0.81802076,
      "learning_rate": 3.873515581290965e-06,
      "loss": 0.84095013,
      "num_input_tokens_seen": 24858900,
      "step": 1168,
      "time_per_iteration": 2.7053585052490234
    },
    {
      "auxiliary_loss_clip": 0.01231436,
      "auxiliary_loss_mlp": 0.01043489,
      "balance_loss_clip": 1.06982327,
      "balance_loss_mlp": 1.03162813,
      "epoch": 0.14056393915709733,
      "flos": 18332972501760.0,
      "grad_norm": 3.182121105468663,
      "language_loss": 0.75484252,
      "learning_rate": 3.8732428159848575e-06,
      "loss": 0.77759176,
      "num_input_tokens_seen": 24877875,
      "step": 1169,
      "time_per_iteration": 2.7138776779174805
    },
    {
      "auxiliary_loss_clip": 0.01179243,
      "auxiliary_loss_mlp": 0.01047097,
      "balance_loss_clip": 1.06455874,
      "balance_loss_mlp": 1.03508711,
      "epoch": 0.14068418204773642,
      "flos": 26687517770880.0,
      "grad_norm": 1.9613025217237245,
      "language_loss": 0.78585088,
      "learning_rate": 3.872969766509304e-06,
      "loss": 0.80811429,
      "num_input_tokens_seen": 24898430,
      "step": 1170,
      "time_per_iteration": 2.7638437747955322
    },
    {
      "auxiliary_loss_clip": 0.01068826,
      "auxiliary_loss_mlp": 0.01001997,
      "balance_loss_clip": 1.02901006,
      "balance_loss_mlp": 0.99827766,
      "epoch": 0.14080442493837553,
      "flos": 65259314501760.0,
      "grad_norm": 0.7629333309330887,
      "language_loss": 0.55670357,
      "learning_rate": 3.872696432905726e-06,
      "loss": 0.57741177,
      "num_input_tokens_seen": 24959250,
      "step": 1171,
      "time_per_iteration": 3.363835096359253
    },
    {
      "auxiliary_loss_clip": 0.01203714,
      "auxiliary_loss_mlp": 0.01042517,
      "balance_loss_clip": 1.06206012,
      "balance_loss_mlp": 1.03092992,
      "epoch": 0.1409246678290146,
      "flos": 25776715582080.0,
      "grad_norm": 3.352937575561993,
      "language_loss": 0.71733201,
      "learning_rate": 3.872422815215589e-06,
      "loss": 0.73979431,
      "num_input_tokens_seen": 24978330,
      "step": 1172,
      "time_per_iteration": 2.6973602771759033
    },
    {
      "auxiliary_loss_clip": 0.0117519,
      "auxiliary_loss_mlp": 0.01043131,
      "balance_loss_clip": 1.05719566,
      "balance_loss_mlp": 1.03090572,
      "epoch": 0.1410449107196537,
      "flos": 21868521217920.0,
      "grad_norm": 1.8576800906753579,
      "language_loss": 0.74240685,
      "learning_rate": 3.8721489134803994e-06,
      "loss": 0.76459014,
      "num_input_tokens_seen": 24997120,
      "step": 1173,
      "time_per_iteration": 2.8288915157318115
    },
    {
      "auxiliary_loss_clip": 0.01179965,
      "auxiliary_loss_mlp": 0.01043008,
      "balance_loss_clip": 1.05973375,
      "balance_loss_mlp": 1.03179622,
      "epoch": 0.1411651536102928,
      "flos": 16684133564160.0,
      "grad_norm": 2.2525646759964766,
      "language_loss": 0.72483802,
      "learning_rate": 3.871874727741707e-06,
      "loss": 0.74706775,
      "num_input_tokens_seen": 25014350,
      "step": 1174,
      "time_per_iteration": 2.7903292179107666
    },
    {
      "auxiliary_loss_clip": 0.01182548,
      "auxiliary_loss_mlp": 0.01047813,
      "balance_loss_clip": 1.06321669,
      "balance_loss_mlp": 1.037287,
      "epoch": 0.1412853965009319,
      "flos": 20992264934400.0,
      "grad_norm": 1.833637482475246,
      "language_loss": 0.96682179,
      "learning_rate": 3.871600258041108e-06,
      "loss": 0.98912543,
      "num_input_tokens_seen": 25033875,
      "step": 1175,
      "time_per_iteration": 2.7811946868896484
    },
    {
      "auxiliary_loss_clip": 0.01208202,
      "auxiliary_loss_mlp": 0.01044852,
      "balance_loss_clip": 1.06853819,
      "balance_loss_mlp": 1.03376508,
      "epoch": 0.14140563939157097,
      "flos": 20335279224960.0,
      "grad_norm": 2.5698115088398197,
      "language_loss": 0.85813129,
      "learning_rate": 3.871325504420238e-06,
      "loss": 0.8806619,
      "num_input_tokens_seen": 25052865,
      "step": 1176,
      "time_per_iteration": 2.724039316177368
    },
    {
      "auxiliary_loss_clip": 0.01190675,
      "auxiliary_loss_mlp": 0.01050738,
      "balance_loss_clip": 1.06590176,
      "balance_loss_mlp": 1.03993201,
      "epoch": 0.14152588228221005,
      "flos": 21068826773760.0,
      "grad_norm": 1.8515606876978106,
      "language_loss": 0.81620145,
      "learning_rate": 3.871050466920776e-06,
      "loss": 0.83861554,
      "num_input_tokens_seen": 25072770,
      "step": 1177,
      "time_per_iteration": 2.797941207885742
    },
    {
      "auxiliary_loss_clip": 0.01184067,
      "auxiliary_loss_mlp": 0.01045135,
      "balance_loss_clip": 1.06313157,
      "balance_loss_mlp": 1.03496671,
      "epoch": 0.14164612517284916,
      "flos": 18223157646720.0,
      "grad_norm": 1.9222834164469982,
      "language_loss": 0.79755533,
      "learning_rate": 3.870775145584447e-06,
      "loss": 0.81984735,
      "num_input_tokens_seen": 25090550,
      "step": 1178,
      "time_per_iteration": 2.861231803894043
    },
    {
      "auxiliary_loss_clip": 0.01249482,
      "auxiliary_loss_mlp": 0.00765261,
      "balance_loss_clip": 1.07328737,
      "balance_loss_mlp": 1.00054407,
      "epoch": 0.14176636806348825,
      "flos": 22744454279040.0,
      "grad_norm": 3.1948213666758143,
      "language_loss": 0.64863825,
      "learning_rate": 3.8704995404530145e-06,
      "loss": 0.66878569,
      "num_input_tokens_seen": 25106175,
      "step": 1179,
      "time_per_iteration": 2.6615514755249023
    },
    {
      "auxiliary_loss_clip": 0.0119776,
      "auxiliary_loss_mlp": 0.00764399,
      "balance_loss_clip": 1.05882847,
      "balance_loss_mlp": 1.00049448,
      "epoch": 0.14188661095412733,
      "flos": 22091095843200.0,
      "grad_norm": 2.098124654663375,
      "language_loss": 0.85137898,
      "learning_rate": 3.87022365156829e-06,
      "loss": 0.87100053,
      "num_input_tokens_seen": 25126890,
      "step": 1180,
      "time_per_iteration": 2.8688642978668213
    },
    {
      "auxiliary_loss_clip": 0.0124066,
      "auxiliary_loss_mlp": 0.01041617,
      "balance_loss_clip": 1.06980872,
      "balance_loss_mlp": 1.03066182,
      "epoch": 0.14200685384476644,
      "flos": 24352390604160.0,
      "grad_norm": 2.0601590226437945,
      "language_loss": 0.80981427,
      "learning_rate": 3.869947478972123e-06,
      "loss": 0.83263707,
      "num_input_tokens_seen": 25147915,
      "step": 1181,
      "time_per_iteration": 2.636692523956299
    },
    {
      "auxiliary_loss_clip": 0.01195812,
      "auxiliary_loss_mlp": 0.01031677,
      "balance_loss_clip": 1.06509328,
      "balance_loss_mlp": 1.02093017,
      "epoch": 0.14212709673540552,
      "flos": 24022048199040.0,
      "grad_norm": 2.422863164991537,
      "language_loss": 0.82279241,
      "learning_rate": 3.869671022706412e-06,
      "loss": 0.84506726,
      "num_input_tokens_seen": 25166645,
      "step": 1182,
      "time_per_iteration": 2.810215950012207
    },
    {
      "auxiliary_loss_clip": 0.01161604,
      "auxiliary_loss_mlp": 0.01041729,
      "balance_loss_clip": 1.05975568,
      "balance_loss_mlp": 1.03114319,
      "epoch": 0.1422473396260446,
      "flos": 26431797870720.0,
      "grad_norm": 2.192156782475494,
      "language_loss": 0.64771914,
      "learning_rate": 3.869394282813092e-06,
      "loss": 0.66975248,
      "num_input_tokens_seen": 25185845,
      "step": 1183,
      "time_per_iteration": 2.7931368350982666
    },
    {
      "auxiliary_loss_clip": 0.01227887,
      "auxiliary_loss_mlp": 0.01043873,
      "balance_loss_clip": 1.06954908,
      "balance_loss_mlp": 1.03226185,
      "epoch": 0.1423675825166837,
      "flos": 17055306754560.0,
      "grad_norm": 2.7473190067636533,
      "language_loss": 0.8947469,
      "learning_rate": 3.869117259334147e-06,
      "loss": 0.91746449,
      "num_input_tokens_seen": 25203770,
      "step": 1184,
      "time_per_iteration": 3.5457699298858643
    },
    {
      "auxiliary_loss_clip": 0.01161021,
      "auxiliary_loss_mlp": 0.01047176,
      "balance_loss_clip": 1.05612898,
      "balance_loss_mlp": 1.03632784,
      "epoch": 0.1424878254073228,
      "flos": 17929480049280.0,
      "grad_norm": 1.861331190756515,
      "language_loss": 0.82095921,
      "learning_rate": 3.868839952311599e-06,
      "loss": 0.84304118,
      "num_input_tokens_seen": 25221725,
      "step": 1185,
      "time_per_iteration": 3.742766857147217
    },
    {
      "auxiliary_loss_clip": 0.01240248,
      "auxiliary_loss_mlp": 0.01041317,
      "balance_loss_clip": 1.07005513,
      "balance_loss_mlp": 1.02993238,
      "epoch": 0.14260806829796188,
      "flos": 20303606407680.0,
      "grad_norm": 2.4194674360768746,
      "language_loss": 0.80086285,
      "learning_rate": 3.868562361787516e-06,
      "loss": 0.82367849,
      "num_input_tokens_seen": 25240855,
      "step": 1186,
      "time_per_iteration": 2.6647567749023438
    },
    {
      "auxiliary_loss_clip": 0.01188943,
      "auxiliary_loss_mlp": 0.01039481,
      "balance_loss_clip": 1.06524074,
      "balance_loss_mlp": 1.02863276,
      "epoch": 0.14272831118860096,
      "flos": 23185724860800.0,
      "grad_norm": 2.0859033769894535,
      "language_loss": 0.69308007,
      "learning_rate": 3.868284487804009e-06,
      "loss": 0.71536434,
      "num_input_tokens_seen": 25260085,
      "step": 1187,
      "time_per_iteration": 2.734628200531006
    },
    {
      "auxiliary_loss_clip": 0.01230827,
      "auxiliary_loss_mlp": 0.01048871,
      "balance_loss_clip": 1.07183218,
      "balance_loss_mlp": 1.03782058,
      "epoch": 0.14284855407924008,
      "flos": 27232210586880.0,
      "grad_norm": 2.5546782705690005,
      "language_loss": 0.77958894,
      "learning_rate": 3.86800633040323e-06,
      "loss": 0.80238593,
      "num_input_tokens_seen": 25280675,
      "step": 1188,
      "time_per_iteration": 2.710831642150879
    },
    {
      "auxiliary_loss_clip": 0.01201487,
      "auxiliary_loss_mlp": 0.01045737,
      "balance_loss_clip": 1.06133747,
      "balance_loss_mlp": 1.03465688,
      "epoch": 0.14296879696987916,
      "flos": 28184202696960.0,
      "grad_norm": 2.2504353758461897,
      "language_loss": 0.78808743,
      "learning_rate": 3.867727889627376e-06,
      "loss": 0.81055969,
      "num_input_tokens_seen": 25300290,
      "step": 1189,
      "time_per_iteration": 3.5972349643707275
    },
    {
      "auxiliary_loss_clip": 0.01229239,
      "auxiliary_loss_mlp": 0.01036988,
      "balance_loss_clip": 1.06996477,
      "balance_loss_mlp": 1.02611637,
      "epoch": 0.14308903986051824,
      "flos": 19390290266880.0,
      "grad_norm": 2.1700220397608927,
      "language_loss": 0.78406221,
      "learning_rate": 3.867449165518687e-06,
      "loss": 0.80672449,
      "num_input_tokens_seen": 25316760,
      "step": 1190,
      "time_per_iteration": 2.7054429054260254
    },
    {
      "auxiliary_loss_clip": 0.01226387,
      "auxiliary_loss_mlp": 0.01040366,
      "balance_loss_clip": 1.06726432,
      "balance_loss_mlp": 1.02877343,
      "epoch": 0.14320928275115732,
      "flos": 17457506317440.0,
      "grad_norm": 2.107057485173385,
      "language_loss": 0.71119195,
      "learning_rate": 3.867170158119444e-06,
      "loss": 0.73385954,
      "num_input_tokens_seen": 25335760,
      "step": 1191,
      "time_per_iteration": 3.664645195007324
    },
    {
      "auxiliary_loss_clip": 0.01204623,
      "auxiliary_loss_mlp": 0.01038357,
      "balance_loss_clip": 1.0644033,
      "balance_loss_mlp": 1.02694321,
      "epoch": 0.14332952564179643,
      "flos": 21466070259840.0,
      "grad_norm": 1.931581157420706,
      "language_loss": 0.75339365,
      "learning_rate": 3.866890867471972e-06,
      "loss": 0.77582347,
      "num_input_tokens_seen": 25354230,
      "step": 1192,
      "time_per_iteration": 2.699232339859009
    },
    {
      "auxiliary_loss_clip": 0.0121682,
      "auxiliary_loss_mlp": 0.01039495,
      "balance_loss_clip": 1.06709909,
      "balance_loss_mlp": 1.02755642,
      "epoch": 0.14344976853243552,
      "flos": 16396992241920.0,
      "grad_norm": 3.483571256028305,
      "language_loss": 0.89468586,
      "learning_rate": 3.86661129361864e-06,
      "loss": 0.91724908,
      "num_input_tokens_seen": 25368720,
      "step": 1193,
      "time_per_iteration": 2.691516399383545
    },
    {
      "auxiliary_loss_clip": 0.01229417,
      "auxiliary_loss_mlp": 0.01037262,
      "balance_loss_clip": 1.0658747,
      "balance_loss_mlp": 1.02629471,
      "epoch": 0.1435700114230746,
      "flos": 18916736336640.0,
      "grad_norm": 1.9745932137774294,
      "language_loss": 0.86215913,
      "learning_rate": 3.866331436601859e-06,
      "loss": 0.88482594,
      "num_input_tokens_seen": 25386715,
      "step": 1194,
      "time_per_iteration": 2.787811279296875
    },
    {
      "auxiliary_loss_clip": 0.01227181,
      "auxiliary_loss_mlp": 0.00764814,
      "balance_loss_clip": 1.06645012,
      "balance_loss_mlp": 1.00058508,
      "epoch": 0.1436902543137137,
      "flos": 19755394058880.0,
      "grad_norm": 2.662081448708821,
      "language_loss": 0.74044049,
      "learning_rate": 3.866051296464083e-06,
      "loss": 0.76036048,
      "num_input_tokens_seen": 25405550,
      "step": 1195,
      "time_per_iteration": 2.84523868560791
    },
    {
      "auxiliary_loss_clip": 0.0120401,
      "auxiliary_loss_mlp": 0.01043644,
      "balance_loss_clip": 1.06282067,
      "balance_loss_mlp": 1.0312525,
      "epoch": 0.1438104972043528,
      "flos": 14684807669760.0,
      "grad_norm": 2.0354477631499415,
      "language_loss": 0.85391283,
      "learning_rate": 3.86577087324781e-06,
      "loss": 0.87638927,
      "num_input_tokens_seen": 25422040,
      "step": 1196,
      "time_per_iteration": 2.6646270751953125
    },
    {
      "auxiliary_loss_clip": 0.01189706,
      "auxiliary_loss_mlp": 0.01042793,
      "balance_loss_clip": 1.06378222,
      "balance_loss_mlp": 1.0322845,
      "epoch": 0.14393074009499188,
      "flos": 17092330698240.0,
      "grad_norm": 2.5933887231484487,
      "language_loss": 0.77410793,
      "learning_rate": 3.865490166995578e-06,
      "loss": 0.79643291,
      "num_input_tokens_seen": 25440270,
      "step": 1197,
      "time_per_iteration": 2.7797188758850098
    },
    {
      "auxiliary_loss_clip": 0.01162381,
      "auxiliary_loss_mlp": 0.0104797,
      "balance_loss_clip": 1.05557203,
      "balance_loss_mlp": 1.03641868,
      "epoch": 0.144050982985631,
      "flos": 30476200608000.0,
      "grad_norm": 5.084718621164245,
      "language_loss": 0.84029156,
      "learning_rate": 3.86520917774997e-06,
      "loss": 0.86239511,
      "num_input_tokens_seen": 25459705,
      "step": 1198,
      "time_per_iteration": 2.8325605392456055
    },
    {
      "auxiliary_loss_clip": 0.01170095,
      "auxiliary_loss_mlp": 0.01036957,
      "balance_loss_clip": 1.05876637,
      "balance_loss_mlp": 1.02665126,
      "epoch": 0.14417122587627007,
      "flos": 17858484817920.0,
      "grad_norm": 2.2250016413895612,
      "language_loss": 0.74954432,
      "learning_rate": 3.864927905553614e-06,
      "loss": 0.77161485,
      "num_input_tokens_seen": 25477615,
      "step": 1199,
      "time_per_iteration": 2.8456482887268066
    },
    {
      "auxiliary_loss_clip": 0.01199658,
      "auxiliary_loss_mlp": 0.01036524,
      "balance_loss_clip": 1.06078029,
      "balance_loss_mlp": 1.02569985,
      "epoch": 0.14429146876690915,
      "flos": 21613914639360.0,
      "grad_norm": 1.914887598813734,
      "language_loss": 0.88945305,
      "learning_rate": 3.8646463504491765e-06,
      "loss": 0.91181481,
      "num_input_tokens_seen": 25497750,
      "step": 1200,
      "time_per_iteration": 2.7463133335113525
    },
    {
      "auxiliary_loss_clip": 0.01213527,
      "auxiliary_loss_mlp": 0.01046387,
      "balance_loss_clip": 1.06623244,
      "balance_loss_mlp": 1.0348357,
      "epoch": 0.14441171165754824,
      "flos": 23258120722560.0,
      "grad_norm": 1.730900113612014,
      "language_loss": 0.83222961,
      "learning_rate": 3.8643645124793705e-06,
      "loss": 0.85482872,
      "num_input_tokens_seen": 25516650,
      "step": 1201,
      "time_per_iteration": 2.6695313453674316
    },
    {
      "auxiliary_loss_clip": 0.012051,
      "auxiliary_loss_mlp": 0.01040289,
      "balance_loss_clip": 1.06143594,
      "balance_loss_mlp": 1.029423,
      "epoch": 0.14453195454818735,
      "flos": 42854213963520.0,
      "grad_norm": 1.6805112722746316,
      "language_loss": 0.74786735,
      "learning_rate": 3.8640823916869515e-06,
      "loss": 0.77032131,
      "num_input_tokens_seen": 25540960,
      "step": 1202,
      "time_per_iteration": 2.9278323650360107
    },
    {
      "auxiliary_loss_clip": 0.01208274,
      "auxiliary_loss_mlp": 0.01036172,
      "balance_loss_clip": 1.06713963,
      "balance_loss_mlp": 1.02578294,
      "epoch": 0.14465219743882643,
      "flos": 27235873774080.0,
      "grad_norm": 1.72495765454861,
      "language_loss": 0.78285581,
      "learning_rate": 3.863799988114714e-06,
      "loss": 0.8053003,
      "num_input_tokens_seen": 25562990,
      "step": 1203,
      "time_per_iteration": 2.737952709197998
    },
    {
      "auxiliary_loss_clip": 0.01246531,
      "auxiliary_loss_mlp": 0.01039199,
      "balance_loss_clip": 1.07244039,
      "balance_loss_mlp": 1.02811289,
      "epoch": 0.1447724403294655,
      "flos": 16690705752960.0,
      "grad_norm": 2.310849913953314,
      "language_loss": 0.70221436,
      "learning_rate": 3.863517301805502e-06,
      "loss": 0.72507167,
      "num_input_tokens_seen": 25581380,
      "step": 1204,
      "time_per_iteration": 2.666224479675293
    },
    {
      "auxiliary_loss_clip": 0.01185077,
      "auxiliary_loss_mlp": 0.01043728,
      "balance_loss_clip": 1.06536305,
      "balance_loss_mlp": 1.03270102,
      "epoch": 0.14489268322010462,
      "flos": 20073741321600.0,
      "grad_norm": 2.4460658444674297,
      "language_loss": 0.96655941,
      "learning_rate": 3.863234332802196e-06,
      "loss": 0.98884749,
      "num_input_tokens_seen": 25593585,
      "step": 1205,
      "time_per_iteration": 2.7246968746185303
    },
    {
      "auxiliary_loss_clip": 0.01157811,
      "auxiliary_loss_mlp": 0.01047062,
      "balance_loss_clip": 1.05807996,
      "balance_loss_mlp": 1.03634512,
      "epoch": 0.1450129261107437,
      "flos": 27125627955840.0,
      "grad_norm": 5.706346481590045,
      "language_loss": 0.73726785,
      "learning_rate": 3.862951081147723e-06,
      "loss": 0.75931656,
      "num_input_tokens_seen": 25613750,
      "step": 1206,
      "time_per_iteration": 2.7926626205444336
    },
    {
      "auxiliary_loss_clip": 0.01210978,
      "auxiliary_loss_mlp": 0.01036187,
      "balance_loss_clip": 1.0678544,
      "balance_loss_mlp": 1.02570879,
      "epoch": 0.1451331690013828,
      "flos": 25702344472320.0,
      "grad_norm": 2.4193016260258737,
      "language_loss": 0.78340673,
      "learning_rate": 3.862667546885053e-06,
      "loss": 0.8058784,
      "num_input_tokens_seen": 25632300,
      "step": 1207,
      "time_per_iteration": 2.7086234092712402
    },
    {
      "auxiliary_loss_clip": 0.01141472,
      "auxiliary_loss_mlp": 0.01043583,
      "balance_loss_clip": 1.05101538,
      "balance_loss_mlp": 1.03204322,
      "epoch": 0.14525341189202187,
      "flos": 25737393168000.0,
      "grad_norm": 2.6483156988803875,
      "language_loss": 0.73579204,
      "learning_rate": 3.8623837300571965e-06,
      "loss": 0.75764269,
      "num_input_tokens_seen": 25651285,
      "step": 1208,
      "time_per_iteration": 2.871870994567871
    },
    {
      "auxiliary_loss_clip": 0.01186593,
      "auxiliary_loss_mlp": 0.01047857,
      "balance_loss_clip": 1.06063271,
      "balance_loss_mlp": 1.03563213,
      "epoch": 0.14537365478266098,
      "flos": 23073898844160.0,
      "grad_norm": 3.529282883060105,
      "language_loss": 0.84002173,
      "learning_rate": 3.8620996307072085e-06,
      "loss": 0.8623662,
      "num_input_tokens_seen": 25671990,
      "step": 1209,
      "time_per_iteration": 2.809213638305664
    },
    {
      "auxiliary_loss_clip": 0.01162875,
      "auxiliary_loss_mlp": 0.0104238,
      "balance_loss_clip": 1.05152452,
      "balance_loss_mlp": 1.0316515,
      "epoch": 0.14549389767330007,
      "flos": 20595021448320.0,
      "grad_norm": 2.0836828969331584,
      "language_loss": 0.64749944,
      "learning_rate": 3.861815248878188e-06,
      "loss": 0.66955203,
      "num_input_tokens_seen": 25689475,
      "step": 1210,
      "time_per_iteration": 3.6570234298706055
    },
    {
      "auxiliary_loss_clip": 0.01197667,
      "auxiliary_loss_mlp": 0.0104083,
      "balance_loss_clip": 1.06542015,
      "balance_loss_mlp": 1.03028595,
      "epoch": 0.14561414056393915,
      "flos": 15121804533120.0,
      "grad_norm": 2.754570223004999,
      "language_loss": 0.79912424,
      "learning_rate": 3.861530584613274e-06,
      "loss": 0.82150918,
      "num_input_tokens_seen": 25707475,
      "step": 1211,
      "time_per_iteration": 3.7738935947418213
    },
    {
      "auxiliary_loss_clip": 0.01196353,
      "auxiliary_loss_mlp": 0.01043223,
      "balance_loss_clip": 1.06616545,
      "balance_loss_mlp": 1.03213036,
      "epoch": 0.14573438345457826,
      "flos": 19427493778560.0,
      "grad_norm": 2.8722248288343977,
      "language_loss": 0.82521844,
      "learning_rate": 3.86124563795565e-06,
      "loss": 0.84761417,
      "num_input_tokens_seen": 25726290,
      "step": 1212,
      "time_per_iteration": 2.7378175258636475
    },
    {
      "auxiliary_loss_clip": 0.01206481,
      "auxiliary_loss_mlp": 0.01037803,
      "balance_loss_clip": 1.06765687,
      "balance_loss_mlp": 1.02662086,
      "epoch": 0.14585462634521734,
      "flos": 24828422572800.0,
      "grad_norm": 1.8461150237486856,
      "language_loss": 0.70346928,
      "learning_rate": 3.860960408948543e-06,
      "loss": 0.72591203,
      "num_input_tokens_seen": 25748040,
      "step": 1213,
      "time_per_iteration": 2.7574586868286133
    },
    {
      "auxiliary_loss_clip": 0.01188925,
      "auxiliary_loss_mlp": 0.01046027,
      "balance_loss_clip": 1.06602252,
      "balance_loss_mlp": 1.03532791,
      "epoch": 0.14597486923585642,
      "flos": 15448627405440.0,
      "grad_norm": 3.421766419116699,
      "language_loss": 0.89558876,
      "learning_rate": 3.860674897635222e-06,
      "loss": 0.91793823,
      "num_input_tokens_seen": 25764525,
      "step": 1214,
      "time_per_iteration": 3.6840388774871826
    },
    {
      "auxiliary_loss_clip": 0.01215329,
      "auxiliary_loss_mlp": 0.01038254,
      "balance_loss_clip": 1.06868958,
      "balance_loss_mlp": 1.02711987,
      "epoch": 0.1460951121264955,
      "flos": 16655154266880.0,
      "grad_norm": 1.9814544028767456,
      "language_loss": 0.83127975,
      "learning_rate": 3.860389104058998e-06,
      "loss": 0.85381556,
      "num_input_tokens_seen": 25782755,
      "step": 1215,
      "time_per_iteration": 2.6750543117523193
    },
    {
      "auxiliary_loss_clip": 0.01199679,
      "auxiliary_loss_mlp": 0.01044576,
      "balance_loss_clip": 1.06488299,
      "balance_loss_mlp": 1.03351951,
      "epoch": 0.14621535501713462,
      "flos": 24863291700480.0,
      "grad_norm": 2.023405377999328,
      "language_loss": 0.72447026,
      "learning_rate": 3.860103028263227e-06,
      "loss": 0.7469129,
      "num_input_tokens_seen": 25805860,
      "step": 1216,
      "time_per_iteration": 2.7583911418914795
    },
    {
      "auxiliary_loss_clip": 0.01205319,
      "auxiliary_loss_mlp": 0.01039518,
      "balance_loss_clip": 1.05983901,
      "balance_loss_mlp": 1.0287652,
      "epoch": 0.1463355979077737,
      "flos": 25228000442880.0,
      "grad_norm": 3.4572832993162286,
      "language_loss": 0.7033996,
      "learning_rate": 3.859816670291304e-06,
      "loss": 0.72584796,
      "num_input_tokens_seen": 25824955,
      "step": 1217,
      "time_per_iteration": 3.596620559692383
    },
    {
      "auxiliary_loss_clip": 0.01214128,
      "auxiliary_loss_mlp": 0.01032618,
      "balance_loss_clip": 1.06687891,
      "balance_loss_mlp": 1.02162147,
      "epoch": 0.14645584079841278,
      "flos": 22054143726720.0,
      "grad_norm": 2.5014279618540853,
      "language_loss": 0.90105748,
      "learning_rate": 3.859530030186672e-06,
      "loss": 0.92352492,
      "num_input_tokens_seen": 25841965,
      "step": 1218,
      "time_per_iteration": 2.671746253967285
    },
    {
      "auxiliary_loss_clip": 0.01209315,
      "auxiliary_loss_mlp": 0.0103737,
      "balance_loss_clip": 1.06670499,
      "balance_loss_mlp": 1.02673638,
      "epoch": 0.1465760836890519,
      "flos": 23623870959360.0,
      "grad_norm": 2.796232627736643,
      "language_loss": 0.82706738,
      "learning_rate": 3.859243107992813e-06,
      "loss": 0.84953427,
      "num_input_tokens_seen": 25860770,
      "step": 1219,
      "time_per_iteration": 2.719111680984497
    },
    {
      "auxiliary_loss_clip": 0.01208008,
      "auxiliary_loss_mlp": 0.01042248,
      "balance_loss_clip": 1.06745446,
      "balance_loss_mlp": 1.03049409,
      "epoch": 0.14669632657969098,
      "flos": 37407893356800.0,
      "grad_norm": 2.87010269876934,
      "language_loss": 0.78030205,
      "learning_rate": 3.858955903753252e-06,
      "loss": 0.80280459,
      "num_input_tokens_seen": 25879410,
      "step": 1220,
      "time_per_iteration": 2.791734457015991
    },
    {
      "auxiliary_loss_clip": 0.0121267,
      "auxiliary_loss_mlp": 0.00763916,
      "balance_loss_clip": 1.06622839,
      "balance_loss_mlp": 1.0004648,
      "epoch": 0.14681656947033006,
      "flos": 28365910623360.0,
      "grad_norm": 1.5500907260095331,
      "language_loss": 0.83705282,
      "learning_rate": 3.858668417511559e-06,
      "loss": 0.85681868,
      "num_input_tokens_seen": 25902160,
      "step": 1221,
      "time_per_iteration": 2.8251023292541504
    },
    {
      "auxiliary_loss_clip": 0.01166013,
      "auxiliary_loss_mlp": 0.01045539,
      "balance_loss_clip": 1.05783093,
      "balance_loss_mlp": 1.03398156,
      "epoch": 0.14693681236096917,
      "flos": 18479488078080.0,
      "grad_norm": 2.590108929473409,
      "language_loss": 0.76366693,
      "learning_rate": 3.8583806493113445e-06,
      "loss": 0.78578246,
      "num_input_tokens_seen": 25920505,
      "step": 1222,
      "time_per_iteration": 2.746364116668701
    },
    {
      "auxiliary_loss_clip": 0.01226458,
      "auxiliary_loss_mlp": 0.01042838,
      "balance_loss_clip": 1.06773782,
      "balance_loss_mlp": 1.03107178,
      "epoch": 0.14705705525160825,
      "flos": 20777806782720.0,
      "grad_norm": 2.103466716951821,
      "language_loss": 0.82520723,
      "learning_rate": 3.858092599196263e-06,
      "loss": 0.84790015,
      "num_input_tokens_seen": 25938460,
      "step": 1223,
      "time_per_iteration": 2.655397415161133
    },
    {
      "auxiliary_loss_clip": 0.01211119,
      "auxiliary_loss_mlp": 0.01042037,
      "balance_loss_clip": 1.06770337,
      "balance_loss_mlp": 1.030509,
      "epoch": 0.14717729814224734,
      "flos": 29932944336000.0,
      "grad_norm": 2.3581255008623643,
      "language_loss": 0.82869202,
      "learning_rate": 3.857804267210012e-06,
      "loss": 0.85122359,
      "num_input_tokens_seen": 25957760,
      "step": 1224,
      "time_per_iteration": 3.012885808944702
    },
    {
      "auxiliary_loss_clip": 0.01211289,
      "auxiliary_loss_mlp": 0.01040758,
      "balance_loss_clip": 1.06518853,
      "balance_loss_mlp": 1.02948117,
      "epoch": 0.14729754103288642,
      "flos": 20047491457920.0,
      "grad_norm": 2.189883804548403,
      "language_loss": 0.88357973,
      "learning_rate": 3.857515653396331e-06,
      "loss": 0.90610021,
      "num_input_tokens_seen": 25974970,
      "step": 1225,
      "time_per_iteration": 2.690035581588745
    },
    {
      "auxiliary_loss_clip": 0.01224889,
      "auxiliary_loss_mlp": 0.01041123,
      "balance_loss_clip": 1.0698154,
      "balance_loss_mlp": 1.02970839,
      "epoch": 0.14741778392352553,
      "flos": 19281516906240.0,
      "grad_norm": 2.6089342175886223,
      "language_loss": 0.86726385,
      "learning_rate": 3.857226757799002e-06,
      "loss": 0.88992393,
      "num_input_tokens_seen": 25992525,
      "step": 1226,
      "time_per_iteration": 2.7004644870758057
    },
    {
      "auxiliary_loss_clip": 0.0122445,
      "auxiliary_loss_mlp": 0.0104824,
      "balance_loss_clip": 1.06777668,
      "balance_loss_mlp": 1.03646779,
      "epoch": 0.1475380268141646,
      "flos": 25411108999680.0,
      "grad_norm": 2.9477704289225706,
      "language_loss": 0.74460888,
      "learning_rate": 3.85693758046185e-06,
      "loss": 0.76733577,
      "num_input_tokens_seen": 26010815,
      "step": 1227,
      "time_per_iteration": 2.6282927989959717
    },
    {
      "auxiliary_loss_clip": 0.01242806,
      "auxiliary_loss_mlp": 0.01044334,
      "balance_loss_clip": 1.07431614,
      "balance_loss_mlp": 1.03375387,
      "epoch": 0.1476582697048037,
      "flos": 20847652778880.0,
      "grad_norm": 1.9450090592410496,
      "language_loss": 0.82427114,
      "learning_rate": 3.8566481214287435e-06,
      "loss": 0.84714258,
      "num_input_tokens_seen": 26028935,
      "step": 1228,
      "time_per_iteration": 2.6100995540618896
    },
    {
      "auxiliary_loss_clip": 0.01192842,
      "auxiliary_loss_mlp": 0.01043142,
      "balance_loss_clip": 1.06471789,
      "balance_loss_mlp": 1.03233612,
      "epoch": 0.1477785125954428,
      "flos": 14028109269120.0,
      "grad_norm": 2.1215883117819785,
      "language_loss": 0.90291411,
      "learning_rate": 3.8563583807435935e-06,
      "loss": 0.92527395,
      "num_input_tokens_seen": 26045080,
      "step": 1229,
      "time_per_iteration": 2.6920886039733887
    },
    {
      "auxiliary_loss_clip": 0.0118167,
      "auxiliary_loss_mlp": 0.00765077,
      "balance_loss_clip": 1.06310749,
      "balance_loss_mlp": 1.00039959,
      "epoch": 0.1478987554860819,
      "flos": 20516699842560.0,
      "grad_norm": 2.2780364790878482,
      "language_loss": 0.77807891,
      "learning_rate": 3.856068358450353e-06,
      "loss": 0.79754639,
      "num_input_tokens_seen": 26065030,
      "step": 1230,
      "time_per_iteration": 2.8031270503997803
    },
    {
      "auxiliary_loss_clip": 0.01241437,
      "auxiliary_loss_mlp": 0.010441,
      "balance_loss_clip": 1.07158804,
      "balance_loss_mlp": 1.03287613,
      "epoch": 0.14801899837672097,
      "flos": 17857012360320.0,
      "grad_norm": 1.8347944408691907,
      "language_loss": 0.85864723,
      "learning_rate": 3.8557780545930186e-06,
      "loss": 0.88150257,
      "num_input_tokens_seen": 26083445,
      "step": 1231,
      "time_per_iteration": 2.7358665466308594
    },
    {
      "auxiliary_loss_clip": 0.01174723,
      "auxiliary_loss_mlp": 0.01044976,
      "balance_loss_clip": 1.05914879,
      "balance_loss_mlp": 1.03329337,
      "epoch": 0.14813924126736006,
      "flos": 20881408584960.0,
      "grad_norm": 2.2020066673776193,
      "language_loss": 0.79322517,
      "learning_rate": 3.855487469215628e-06,
      "loss": 0.81542218,
      "num_input_tokens_seen": 26102375,
      "step": 1232,
      "time_per_iteration": 2.80078387260437
    },
    {
      "auxiliary_loss_clip": 0.01192992,
      "auxiliary_loss_mlp": 0.00764221,
      "balance_loss_clip": 1.06837988,
      "balance_loss_mlp": 1.0004307,
      "epoch": 0.14825948415799917,
      "flos": 37414070496000.0,
      "grad_norm": 2.1349286902279188,
      "language_loss": 0.72691071,
      "learning_rate": 3.855196602362264e-06,
      "loss": 0.74648285,
      "num_input_tokens_seen": 26125295,
      "step": 1233,
      "time_per_iteration": 2.88732647895813
    },
    {
      "auxiliary_loss_clip": 0.01237689,
      "auxiliary_loss_mlp": 0.01048628,
      "balance_loss_clip": 1.06882119,
      "balance_loss_mlp": 1.03725505,
      "epoch": 0.14837972704863825,
      "flos": 22014641744640.0,
      "grad_norm": 3.3242541208234577,
      "language_loss": 0.94068587,
      "learning_rate": 3.854905454077051e-06,
      "loss": 0.96354902,
      "num_input_tokens_seen": 26142905,
      "step": 1234,
      "time_per_iteration": 2.659409999847412
    },
    {
      "auxiliary_loss_clip": 0.01181739,
      "auxiliary_loss_mlp": 0.01041134,
      "balance_loss_clip": 1.0644908,
      "balance_loss_mlp": 1.02943397,
      "epoch": 0.14849996993927733,
      "flos": 20996323171200.0,
      "grad_norm": 2.1450802122037347,
      "language_loss": 0.88296491,
      "learning_rate": 3.854614024404155e-06,
      "loss": 0.90519363,
      "num_input_tokens_seen": 26161215,
      "step": 1235,
      "time_per_iteration": 3.7362024784088135
    },
    {
      "auxiliary_loss_clip": 0.0123884,
      "auxiliary_loss_mlp": 0.00765222,
      "balance_loss_clip": 1.0693965,
      "balance_loss_mlp": 1.0003686,
      "epoch": 0.14862021282991644,
      "flos": 20047994248320.0,
      "grad_norm": 1.9198583271860918,
      "language_loss": 0.89587504,
      "learning_rate": 3.8543223133877865e-06,
      "loss": 0.91591567,
      "num_input_tokens_seen": 26179810,
      "step": 1236,
      "time_per_iteration": 2.9461777210235596
    },
    {
      "auxiliary_loss_clip": 0.01242428,
      "auxiliary_loss_mlp": 0.01041429,
      "balance_loss_clip": 1.06953597,
      "balance_loss_mlp": 1.03050947,
      "epoch": 0.14874045572055553,
      "flos": 22712027276160.0,
      "grad_norm": 1.694816126302321,
      "language_loss": 0.88276559,
      "learning_rate": 3.854030321072198e-06,
      "loss": 0.90560418,
      "num_input_tokens_seen": 26199715,
      "step": 1237,
      "time_per_iteration": 3.8578426837921143
    },
    {
      "auxiliary_loss_clip": 0.01166577,
      "auxiliary_loss_mlp": 0.01042525,
      "balance_loss_clip": 1.05939722,
      "balance_loss_mlp": 1.03056216,
      "epoch": 0.1488606986111946,
      "flos": 25411288567680.0,
      "grad_norm": 3.334826140700753,
      "language_loss": 0.7319302,
      "learning_rate": 3.853738047501682e-06,
      "loss": 0.75402117,
      "num_input_tokens_seen": 26220275,
      "step": 1238,
      "time_per_iteration": 2.9547245502471924
    },
    {
      "auxiliary_loss_clip": 0.01199396,
      "auxiliary_loss_mlp": 0.01044595,
      "balance_loss_clip": 1.06236279,
      "balance_loss_mlp": 1.03326392,
      "epoch": 0.1489809415018337,
      "flos": 17018749687680.0,
      "grad_norm": 1.9102094817518265,
      "language_loss": 0.77833605,
      "learning_rate": 3.85344549272058e-06,
      "loss": 0.800776,
      "num_input_tokens_seen": 26238255,
      "step": 1239,
      "time_per_iteration": 2.6371989250183105
    },
    {
      "auxiliary_loss_clip": 0.01222517,
      "auxiliary_loss_mlp": 0.01039038,
      "balance_loss_clip": 1.06909108,
      "balance_loss_mlp": 1.02722406,
      "epoch": 0.1491011843924728,
      "flos": 33659394860160.0,
      "grad_norm": 1.9078860531272235,
      "language_loss": 0.82485509,
      "learning_rate": 3.853152656773269e-06,
      "loss": 0.84747064,
      "num_input_tokens_seen": 26259690,
      "step": 1240,
      "time_per_iteration": 3.6783108711242676
    },
    {
      "auxiliary_loss_clip": 0.01236882,
      "auxiliary_loss_mlp": 0.01039892,
      "balance_loss_clip": 1.06856108,
      "balance_loss_mlp": 1.02864504,
      "epoch": 0.14922142728311188,
      "flos": 21179000764800.0,
      "grad_norm": 1.672603127741231,
      "language_loss": 0.85003448,
      "learning_rate": 3.852859539704174e-06,
      "loss": 0.8728022,
      "num_input_tokens_seen": 26278990,
      "step": 1241,
      "time_per_iteration": 2.6962952613830566
    },
    {
      "auxiliary_loss_clip": 0.01221293,
      "auxiliary_loss_mlp": 0.01041779,
      "balance_loss_clip": 1.06880963,
      "balance_loss_mlp": 1.03030539,
      "epoch": 0.14934167017375097,
      "flos": 29860548474240.0,
      "grad_norm": 2.7049017827466906,
      "language_loss": 0.76188838,
      "learning_rate": 3.85256614155776e-06,
      "loss": 0.78451908,
      "num_input_tokens_seen": 26299120,
      "step": 1242,
      "time_per_iteration": 2.6520137786865234
    },
    {
      "auxiliary_loss_clip": 0.01157512,
      "auxiliary_loss_mlp": 0.01044636,
      "balance_loss_clip": 1.05711102,
      "balance_loss_mlp": 1.03218412,
      "epoch": 0.14946191306439008,
      "flos": 17019216564480.0,
      "grad_norm": 2.1220034789111177,
      "language_loss": 0.74492705,
      "learning_rate": 3.852272462378535e-06,
      "loss": 0.76694858,
      "num_input_tokens_seen": 26316995,
      "step": 1243,
      "time_per_iteration": 3.6638271808624268
    },
    {
      "auxiliary_loss_clip": 0.01235644,
      "auxiliary_loss_mlp": 0.01043266,
      "balance_loss_clip": 1.0690074,
      "balance_loss_mlp": 1.03274035,
      "epoch": 0.14958215595502916,
      "flos": 15669047214720.0,
      "grad_norm": 1.8502839222124874,
      "language_loss": 0.77816427,
      "learning_rate": 3.85197850221105e-06,
      "loss": 0.80095339,
      "num_input_tokens_seen": 26333295,
      "step": 1244,
      "time_per_iteration": 2.6402182579040527
    },
    {
      "auxiliary_loss_clip": 0.01187789,
      "auxiliary_loss_mlp": 0.01045156,
      "balance_loss_clip": 1.06176603,
      "balance_loss_mlp": 1.03463602,
      "epoch": 0.14970239884566824,
      "flos": 33108560818560.0,
      "grad_norm": 1.934239014791806,
      "language_loss": 0.75782323,
      "learning_rate": 3.851684261099899e-06,
      "loss": 0.78015268,
      "num_input_tokens_seen": 26355035,
      "step": 1245,
      "time_per_iteration": 2.7688374519348145
    },
    {
      "auxiliary_loss_clip": 0.01202686,
      "auxiliary_loss_mlp": 0.01050058,
      "balance_loss_clip": 1.06574261,
      "balance_loss_mlp": 1.03773785,
      "epoch": 0.14982264173630733,
      "flos": 17821245392640.0,
      "grad_norm": 2.1562483463688564,
      "language_loss": 0.86794078,
      "learning_rate": 3.851389739089718e-06,
      "loss": 0.89046824,
      "num_input_tokens_seen": 26371655,
      "step": 1246,
      "time_per_iteration": 2.6131844520568848
    },
    {
      "auxiliary_loss_clip": 0.0122612,
      "auxiliary_loss_mlp": 0.01042536,
      "balance_loss_clip": 1.07127619,
      "balance_loss_mlp": 1.0309844,
      "epoch": 0.14994288462694644,
      "flos": 32409559175040.0,
      "grad_norm": 2.0476302998253875,
      "language_loss": 0.80268204,
      "learning_rate": 3.851094936225186e-06,
      "loss": 0.82536852,
      "num_input_tokens_seen": 26392540,
      "step": 1247,
      "time_per_iteration": 2.779221773147583
    },
    {
      "auxiliary_loss_clip": 0.01162291,
      "auxiliary_loss_mlp": 0.01038183,
      "balance_loss_clip": 1.05686796,
      "balance_loss_mlp": 1.02716208,
      "epoch": 0.15006312751758552,
      "flos": 31794661226880.0,
      "grad_norm": 1.5666960946039505,
      "language_loss": 0.76704705,
      "learning_rate": 3.850799852551024e-06,
      "loss": 0.78905183,
      "num_input_tokens_seen": 26414960,
      "step": 1248,
      "time_per_iteration": 2.8475799560546875
    },
    {
      "auxiliary_loss_clip": 0.01220475,
      "auxiliary_loss_mlp": 0.01041113,
      "balance_loss_clip": 1.06697381,
      "balance_loss_mlp": 1.03015161,
      "epoch": 0.1501833704082246,
      "flos": 16618022582400.0,
      "grad_norm": 2.3971825759884178,
      "language_loss": 0.86187381,
      "learning_rate": 3.850504488111995e-06,
      "loss": 0.88448972,
      "num_input_tokens_seen": 26431635,
      "step": 1249,
      "time_per_iteration": 2.651089906692505
    },
    {
      "auxiliary_loss_clip": 0.01236491,
      "auxiliary_loss_mlp": 0.01038989,
      "balance_loss_clip": 1.06956649,
      "balance_loss_mlp": 1.02901149,
      "epoch": 0.15030361329886371,
      "flos": 23471178243840.0,
      "grad_norm": 1.8777689400892177,
      "language_loss": 0.82689667,
      "learning_rate": 3.850208842952907e-06,
      "loss": 0.84965146,
      "num_input_tokens_seen": 26450440,
      "step": 1250,
      "time_per_iteration": 2.6812965869903564
    },
    {
      "auxiliary_loss_clip": 0.01201514,
      "auxiliary_loss_mlp": 0.01037703,
      "balance_loss_clip": 1.06671023,
      "balance_loss_mlp": 1.02670646,
      "epoch": 0.1504238561895028,
      "flos": 25629409906560.0,
      "grad_norm": 3.2500672290370556,
      "language_loss": 0.79077482,
      "learning_rate": 3.849912917118608e-06,
      "loss": 0.8131671,
      "num_input_tokens_seen": 26471480,
      "step": 1251,
      "time_per_iteration": 2.6895267963409424
    },
    {
      "auxiliary_loss_clip": 0.01114894,
      "auxiliary_loss_mlp": 0.01010473,
      "balance_loss_clip": 1.04155147,
      "balance_loss_mlp": 1.00720692,
      "epoch": 0.15054409908014188,
      "flos": 52095146129280.0,
      "grad_norm": 1.0360261896250826,
      "language_loss": 0.59312749,
      "learning_rate": 3.849616710653992e-06,
      "loss": 0.61438113,
      "num_input_tokens_seen": 26532950,
      "step": 1252,
      "time_per_iteration": 3.288886070251465
    },
    {
      "auxiliary_loss_clip": 0.01206881,
      "auxiliary_loss_mlp": 0.01033201,
      "balance_loss_clip": 1.06584597,
      "balance_loss_mlp": 1.02161431,
      "epoch": 0.150664341970781,
      "flos": 18880251096960.0,
      "grad_norm": 1.7748559327366196,
      "language_loss": 0.74580073,
      "learning_rate": 3.84932022360399e-06,
      "loss": 0.76820159,
      "num_input_tokens_seen": 26551615,
      "step": 1253,
      "time_per_iteration": 2.6920347213745117
    },
    {
      "auxiliary_loss_clip": 0.01228919,
      "auxiliary_loss_mlp": 0.01045377,
      "balance_loss_clip": 1.06899333,
      "balance_loss_mlp": 1.03467155,
      "epoch": 0.15078458486142007,
      "flos": 22163240309760.0,
      "grad_norm": 2.5534406835331884,
      "language_loss": 0.84322083,
      "learning_rate": 3.849023456013581e-06,
      "loss": 0.86596382,
      "num_input_tokens_seen": 26569175,
      "step": 1254,
      "time_per_iteration": 2.7251176834106445
    },
    {
      "auxiliary_loss_clip": 0.01212143,
      "auxiliary_loss_mlp": 0.01052374,
      "balance_loss_clip": 1.06569016,
      "balance_loss_mlp": 1.04150248,
      "epoch": 0.15090482775205916,
      "flos": 26651894457600.0,
      "grad_norm": 2.51262081531877,
      "language_loss": 0.6266765,
      "learning_rate": 3.848726407927784e-06,
      "loss": 0.64932168,
      "num_input_tokens_seen": 26589560,
      "step": 1255,
      "time_per_iteration": 2.6904397010803223
    },
    {
      "auxiliary_loss_clip": 0.01180387,
      "auxiliary_loss_mlp": 0.00765085,
      "balance_loss_clip": 1.06609094,
      "balance_loss_mlp": 1.0003252,
      "epoch": 0.15102507064269824,
      "flos": 21798998444160.0,
      "grad_norm": 2.422101776440795,
      "language_loss": 0.86848551,
      "learning_rate": 3.84842907939166e-06,
      "loss": 0.88794017,
      "num_input_tokens_seen": 26608785,
      "step": 1256,
      "time_per_iteration": 2.7976911067962646
    },
    {
      "auxiliary_loss_clip": 0.01182777,
      "auxiliary_loss_mlp": 0.01048904,
      "balance_loss_clip": 1.06057131,
      "balance_loss_mlp": 1.03875339,
      "epoch": 0.15114531353333735,
      "flos": 22820908377600.0,
      "grad_norm": 2.3689895658337052,
      "language_loss": 0.70775634,
      "learning_rate": 3.8481314704503146e-06,
      "loss": 0.73007309,
      "num_input_tokens_seen": 26628615,
      "step": 1257,
      "time_per_iteration": 2.7760143280029297
    },
    {
      "auxiliary_loss_clip": 0.01178069,
      "auxiliary_loss_mlp": 0.01041941,
      "balance_loss_clip": 1.06201708,
      "balance_loss_mlp": 1.03027654,
      "epoch": 0.15126555642397643,
      "flos": 19682674974720.0,
      "grad_norm": 2.510494947952776,
      "language_loss": 0.87997782,
      "learning_rate": 3.847833581148895e-06,
      "loss": 0.90217793,
      "num_input_tokens_seen": 26647525,
      "step": 1258,
      "time_per_iteration": 2.75046443939209
    },
    {
      "auxiliary_loss_clip": 0.01223989,
      "auxiliary_loss_mlp": 0.0104275,
      "balance_loss_clip": 1.06480289,
      "balance_loss_mlp": 1.0321995,
      "epoch": 0.15138579931461552,
      "flos": 28726022424960.0,
      "grad_norm": 5.793566016812556,
      "language_loss": 0.8120141,
      "learning_rate": 3.84753541153259e-06,
      "loss": 0.83468145,
      "num_input_tokens_seen": 26667095,
      "step": 1259,
      "time_per_iteration": 2.7946462631225586
    },
    {
      "auxiliary_loss_clip": 0.01141446,
      "auxiliary_loss_mlp": 0.01039274,
      "balance_loss_clip": 1.05905426,
      "balance_loss_mlp": 1.02788997,
      "epoch": 0.15150604220525463,
      "flos": 22127006465280.0,
      "grad_norm": 2.511823105845486,
      "language_loss": 0.83309245,
      "learning_rate": 3.847236961646633e-06,
      "loss": 0.85489964,
      "num_input_tokens_seen": 26686075,
      "step": 1260,
      "time_per_iteration": 2.852447271347046
    },
    {
      "auxiliary_loss_clip": 0.01207105,
      "auxiliary_loss_mlp": 0.01044124,
      "balance_loss_clip": 1.06633258,
      "balance_loss_mlp": 1.03307319,
      "epoch": 0.1516262850958937,
      "flos": 12968708515200.0,
      "grad_norm": 2.3811764412057004,
      "language_loss": 0.78246218,
      "learning_rate": 3.846938231536296e-06,
      "loss": 0.80497444,
      "num_input_tokens_seen": 26701695,
      "step": 1261,
      "time_per_iteration": 3.6185057163238525
    },
    {
      "auxiliary_loss_clip": 0.01193116,
      "auxiliary_loss_mlp": 0.01044357,
      "balance_loss_clip": 1.06687903,
      "balance_loss_mlp": 1.03394377,
      "epoch": 0.1517465279865328,
      "flos": 21797130936960.0,
      "grad_norm": 2.281428878245092,
      "language_loss": 0.81149209,
      "learning_rate": 3.8466392212468995e-06,
      "loss": 0.83386678,
      "num_input_tokens_seen": 26721885,
      "step": 1262,
      "time_per_iteration": 2.719754934310913
    },
    {
      "auxiliary_loss_clip": 0.01112766,
      "auxiliary_loss_mlp": 0.01007489,
      "balance_loss_clip": 1.04055524,
      "balance_loss_mlp": 1.00439,
      "epoch": 0.15186677087717187,
      "flos": 58174569901440.0,
      "grad_norm": 0.8170170496742122,
      "language_loss": 0.61931378,
      "learning_rate": 3.8463399308238e-06,
      "loss": 0.64051634,
      "num_input_tokens_seen": 26780990,
      "step": 1263,
      "time_per_iteration": 4.20847749710083
    },
    {
      "auxiliary_loss_clip": 0.01226246,
      "auxiliary_loss_mlp": 0.01037783,
      "balance_loss_clip": 1.07309341,
      "balance_loss_mlp": 1.02707243,
      "epoch": 0.15198701376781099,
      "flos": 32669696448000.0,
      "grad_norm": 2.1052143610277136,
      "language_loss": 0.64234376,
      "learning_rate": 3.846040360312402e-06,
      "loss": 0.66498411,
      "num_input_tokens_seen": 26804250,
      "step": 1264,
      "time_per_iteration": 2.757939577102661
    },
    {
      "auxiliary_loss_clip": 0.0121231,
      "auxiliary_loss_mlp": 0.01043055,
      "balance_loss_clip": 1.06629205,
      "balance_loss_mlp": 1.03217173,
      "epoch": 0.15210725665845007,
      "flos": 28402575431040.0,
      "grad_norm": 2.066804565953042,
      "language_loss": 0.81270349,
      "learning_rate": 3.8457405097581485e-06,
      "loss": 0.83525711,
      "num_input_tokens_seen": 26823240,
      "step": 1265,
      "time_per_iteration": 3.7417423725128174
    },
    {
      "auxiliary_loss_clip": 0.01207043,
      "auxiliary_loss_mlp": 0.01050845,
      "balance_loss_clip": 1.06825483,
      "balance_loss_mlp": 1.03991365,
      "epoch": 0.15222749954908915,
      "flos": 19938179393280.0,
      "grad_norm": 2.136614430701606,
      "language_loss": 0.7795521,
      "learning_rate": 3.8454403792065275e-06,
      "loss": 0.802131,
      "num_input_tokens_seen": 26842060,
      "step": 1266,
      "time_per_iteration": 2.6942152976989746
    },
    {
      "auxiliary_loss_clip": 0.01176906,
      "auxiliary_loss_mlp": 0.01045056,
      "balance_loss_clip": 1.06425738,
      "balance_loss_mlp": 1.0347271,
      "epoch": 0.15234774243972826,
      "flos": 21324223451520.0,
      "grad_norm": 2.1237373206721513,
      "language_loss": 0.86135507,
      "learning_rate": 3.845139968703068e-06,
      "loss": 0.88357466,
      "num_input_tokens_seen": 26859580,
      "step": 1267,
      "time_per_iteration": 2.8268821239471436
    },
    {
      "auxiliary_loss_clip": 0.01182489,
      "auxiliary_loss_mlp": 0.01042381,
      "balance_loss_clip": 1.06136358,
      "balance_loss_mlp": 1.03105021,
      "epoch": 0.15246798533036734,
      "flos": 25957812977280.0,
      "grad_norm": 8.369214401205346,
      "language_loss": 0.83016145,
      "learning_rate": 3.844839278293342e-06,
      "loss": 0.8524102,
      "num_input_tokens_seen": 26880430,
      "step": 1268,
      "time_per_iteration": 2.8021090030670166
    },
    {
      "auxiliary_loss_clip": 0.01145626,
      "auxiliary_loss_mlp": 0.00764702,
      "balance_loss_clip": 1.05310321,
      "balance_loss_mlp": 1.00034618,
      "epoch": 0.15258822822100643,
      "flos": 25811907932160.0,
      "grad_norm": 2.319221860196665,
      "language_loss": 0.76467812,
      "learning_rate": 3.8445383080229654e-06,
      "loss": 0.78378141,
      "num_input_tokens_seen": 26896445,
      "step": 1269,
      "time_per_iteration": 3.7860140800476074
    },
    {
      "auxiliary_loss_clip": 0.01211664,
      "auxiliary_loss_mlp": 0.01043209,
      "balance_loss_clip": 1.06505263,
      "balance_loss_mlp": 1.03256989,
      "epoch": 0.1527084711116455,
      "flos": 25265455349760.0,
      "grad_norm": 2.031912106581231,
      "language_loss": 0.73659229,
      "learning_rate": 3.844237057937593e-06,
      "loss": 0.75914109,
      "num_input_tokens_seen": 26915450,
      "step": 1270,
      "time_per_iteration": 2.7049927711486816
    },
    {
      "auxiliary_loss_clip": 0.01211384,
      "auxiliary_loss_mlp": 0.0104178,
      "balance_loss_clip": 1.0646919,
      "balance_loss_mlp": 1.03087258,
      "epoch": 0.15282871400228462,
      "flos": 29240227572480.0,
      "grad_norm": 3.2138665321018367,
      "language_loss": 0.77997005,
      "learning_rate": 3.843935528082926e-06,
      "loss": 0.80250168,
      "num_input_tokens_seen": 26936475,
      "step": 1271,
      "time_per_iteration": 2.7010464668273926
    },
    {
      "auxiliary_loss_clip": 0.01168978,
      "auxiliary_loss_mlp": 0.01035531,
      "balance_loss_clip": 1.05847812,
      "balance_loss_mlp": 1.02517748,
      "epoch": 0.1529489568929237,
      "flos": 20882952869760.0,
      "grad_norm": 2.0592938450147913,
      "language_loss": 0.85227573,
      "learning_rate": 3.843633718504704e-06,
      "loss": 0.87432081,
      "num_input_tokens_seen": 26954920,
      "step": 1272,
      "time_per_iteration": 2.7234394550323486
    },
    {
      "auxiliary_loss_clip": 0.01241034,
      "auxiliary_loss_mlp": 0.01041073,
      "balance_loss_clip": 1.07173336,
      "balance_loss_mlp": 1.0308094,
      "epoch": 0.1530691997835628,
      "flos": 20083833043200.0,
      "grad_norm": 4.692656744508234,
      "language_loss": 0.90400052,
      "learning_rate": 3.843331629248715e-06,
      "loss": 0.92682165,
      "num_input_tokens_seen": 26972520,
      "step": 1273,
      "time_per_iteration": 2.619450807571411
    },
    {
      "auxiliary_loss_clip": 0.01173766,
      "auxiliary_loss_mlp": 0.01039355,
      "balance_loss_clip": 1.06105661,
      "balance_loss_mlp": 1.02839375,
      "epoch": 0.1531894426742019,
      "flos": 28759814144640.0,
      "grad_norm": 2.248622511986419,
      "language_loss": 0.76734382,
      "learning_rate": 3.843029260360782e-06,
      "loss": 0.78947508,
      "num_input_tokens_seen": 26990890,
      "step": 1274,
      "time_per_iteration": 2.785214424133301
    },
    {
      "auxiliary_loss_clip": 0.01238023,
      "auxiliary_loss_mlp": 0.01042177,
      "balance_loss_clip": 1.0728662,
      "balance_loss_mlp": 1.03233075,
      "epoch": 0.15330968556484098,
      "flos": 22236282616320.0,
      "grad_norm": 1.992100827685223,
      "language_loss": 0.78926992,
      "learning_rate": 3.8427266118867755e-06,
      "loss": 0.81207192,
      "num_input_tokens_seen": 27010640,
      "step": 1275,
      "time_per_iteration": 2.63594651222229
    },
    {
      "auxiliary_loss_clip": 0.01180488,
      "auxiliary_loss_mlp": 0.01036443,
      "balance_loss_clip": 1.06255651,
      "balance_loss_mlp": 1.02557731,
      "epoch": 0.15342992845548006,
      "flos": 27527504296320.0,
      "grad_norm": 2.394024913997373,
      "language_loss": 0.83351558,
      "learning_rate": 3.842423683872608e-06,
      "loss": 0.85568488,
      "num_input_tokens_seen": 27031215,
      "step": 1276,
      "time_per_iteration": 2.7893123626708984
    },
    {
      "auxiliary_loss_clip": 0.01174409,
      "auxiliary_loss_mlp": 0.01040146,
      "balance_loss_clip": 1.06155539,
      "balance_loss_mlp": 1.02936983,
      "epoch": 0.15355017134611917,
      "flos": 19609596754560.0,
      "grad_norm": 3.7113732542515874,
      "language_loss": 0.77996171,
      "learning_rate": 3.842120476364232e-06,
      "loss": 0.80210727,
      "num_input_tokens_seen": 27049665,
      "step": 1277,
      "time_per_iteration": 2.7462363243103027
    },
    {
      "auxiliary_loss_clip": 0.01216034,
      "auxiliary_loss_mlp": 0.01039996,
      "balance_loss_clip": 1.06671023,
      "balance_loss_mlp": 1.02873707,
      "epoch": 0.15367041423675826,
      "flos": 18478590238080.0,
      "grad_norm": 2.883325371398332,
      "language_loss": 0.83351398,
      "learning_rate": 3.841816989407644e-06,
      "loss": 0.85607433,
      "num_input_tokens_seen": 27065155,
      "step": 1278,
      "time_per_iteration": 2.6669135093688965
    },
    {
      "auxiliary_loss_clip": 0.01189415,
      "auxiliary_loss_mlp": 0.01037659,
      "balance_loss_clip": 1.06355786,
      "balance_loss_mlp": 1.02766919,
      "epoch": 0.15379065712739734,
      "flos": 41427662342400.0,
      "grad_norm": 2.1576481628857556,
      "language_loss": 0.76837903,
      "learning_rate": 3.841513223048884e-06,
      "loss": 0.79064983,
      "num_input_tokens_seen": 27085840,
      "step": 1279,
      "time_per_iteration": 2.960766315460205
    },
    {
      "auxiliary_loss_clip": 0.01238242,
      "auxiliary_loss_mlp": 0.01043883,
      "balance_loss_clip": 1.07038248,
      "balance_loss_mlp": 1.03368509,
      "epoch": 0.15391090001803642,
      "flos": 22054215553920.0,
      "grad_norm": 2.346070115133,
      "language_loss": 0.7868036,
      "learning_rate": 3.841209177334031e-06,
      "loss": 0.80962485,
      "num_input_tokens_seen": 27104200,
      "step": 1280,
      "time_per_iteration": 2.622391939163208
    },
    {
      "auxiliary_loss_clip": 0.01184936,
      "auxiliary_loss_mlp": 0.01035119,
      "balance_loss_clip": 1.06544709,
      "balance_loss_mlp": 1.02551103,
      "epoch": 0.15403114290867553,
      "flos": 15450351258240.0,
      "grad_norm": 1.8439632454002177,
      "language_loss": 0.7479642,
      "learning_rate": 3.84090485230921e-06,
      "loss": 0.77016473,
      "num_input_tokens_seen": 27122440,
      "step": 1281,
      "time_per_iteration": 2.7456328868865967
    },
    {
      "auxiliary_loss_clip": 0.0118809,
      "auxiliary_loss_mlp": 0.01043553,
      "balance_loss_clip": 1.06273317,
      "balance_loss_mlp": 1.03325963,
      "epoch": 0.15415138579931462,
      "flos": 17929156826880.0,
      "grad_norm": 3.0274763624636494,
      "language_loss": 0.75962532,
      "learning_rate": 3.840600248020588e-06,
      "loss": 0.78194171,
      "num_input_tokens_seen": 27139380,
      "step": 1282,
      "time_per_iteration": 2.7326769828796387
    },
    {
      "auxiliary_loss_clip": 0.0122309,
      "auxiliary_loss_mlp": 0.01034179,
      "balance_loss_clip": 1.06713641,
      "balance_loss_mlp": 1.02356923,
      "epoch": 0.1542716286899537,
      "flos": 11429325296640.0,
      "grad_norm": 2.3326456674441776,
      "language_loss": 0.80066806,
      "learning_rate": 3.840295364514371e-06,
      "loss": 0.82324076,
      "num_input_tokens_seen": 27156760,
      "step": 1283,
      "time_per_iteration": 2.656437397003174
    },
    {
      "auxiliary_loss_clip": 0.01161295,
      "auxiliary_loss_mlp": 0.01037407,
      "balance_loss_clip": 1.05858636,
      "balance_loss_mlp": 1.02657032,
      "epoch": 0.1543918715805928,
      "flos": 17420338719360.0,
      "grad_norm": 2.871062975628808,
      "language_loss": 0.78455293,
      "learning_rate": 3.83999020183681e-06,
      "loss": 0.80653989,
      "num_input_tokens_seen": 27175455,
      "step": 1284,
      "time_per_iteration": 2.709066152572632
    },
    {
      "auxiliary_loss_clip": 0.01206191,
      "auxiliary_loss_mlp": 0.01045148,
      "balance_loss_clip": 1.06768751,
      "balance_loss_mlp": 1.03532493,
      "epoch": 0.1545121144712319,
      "flos": 17786376264960.0,
      "grad_norm": 3.1028134792181623,
      "language_loss": 0.78566796,
      "learning_rate": 3.839684760034199e-06,
      "loss": 0.80818135,
      "num_input_tokens_seen": 27193660,
      "step": 1285,
      "time_per_iteration": 2.793631076812744
    },
    {
      "auxiliary_loss_clip": 0.01197225,
      "auxiliary_loss_mlp": 0.01038704,
      "balance_loss_clip": 1.06319427,
      "balance_loss_mlp": 1.02816033,
      "epoch": 0.15463235736187098,
      "flos": 28220185146240.0,
      "grad_norm": 3.1764611160668212,
      "language_loss": 0.65199888,
      "learning_rate": 3.8393790391528716e-06,
      "loss": 0.67435813,
      "num_input_tokens_seen": 27214355,
      "step": 1286,
      "time_per_iteration": 2.7229502201080322
    },
    {
      "auxiliary_loss_clip": 0.01199393,
      "auxiliary_loss_mlp": 0.01041421,
      "balance_loss_clip": 1.06176805,
      "balance_loss_mlp": 1.03211069,
      "epoch": 0.15475260025251006,
      "flos": 22856890826880.0,
      "grad_norm": 3.4641701470187143,
      "language_loss": 0.88846678,
      "learning_rate": 3.8390730392392075e-06,
      "loss": 0.91087496,
      "num_input_tokens_seen": 27234335,
      "step": 1287,
      "time_per_iteration": 3.710505485534668
    },
    {
      "auxiliary_loss_clip": 0.01186954,
      "auxiliary_loss_mlp": 0.01043771,
      "balance_loss_clip": 1.06405234,
      "balance_loss_mlp": 1.03292906,
      "epoch": 0.15487284314314917,
      "flos": 17602872658560.0,
      "grad_norm": 3.408049886047457,
      "language_loss": 0.79389983,
      "learning_rate": 3.838766760339626e-06,
      "loss": 0.81620705,
      "num_input_tokens_seen": 27252860,
      "step": 1288,
      "time_per_iteration": 3.6651203632354736
    },
    {
      "auxiliary_loss_clip": 0.0118984,
      "auxiliary_loss_mlp": 0.01035324,
      "balance_loss_clip": 1.05928969,
      "balance_loss_mlp": 1.02482736,
      "epoch": 0.15499308603378825,
      "flos": 20082037363200.0,
      "grad_norm": 2.9856895477227487,
      "language_loss": 0.79144126,
      "learning_rate": 3.838460202500587e-06,
      "loss": 0.81369293,
      "num_input_tokens_seen": 27268650,
      "step": 1289,
      "time_per_iteration": 2.724111795425415
    },
    {
      "auxiliary_loss_clip": 0.01225118,
      "auxiliary_loss_mlp": 0.01039554,
      "balance_loss_clip": 1.06950831,
      "balance_loss_mlp": 1.0290513,
      "epoch": 0.15511332892442733,
      "flos": 15918051271680.0,
      "grad_norm": 2.051626177679508,
      "language_loss": 0.73449349,
      "learning_rate": 3.838153365768599e-06,
      "loss": 0.75714016,
      "num_input_tokens_seen": 27285160,
      "step": 1290,
      "time_per_iteration": 2.6997575759887695
    },
    {
      "auxiliary_loss_clip": 0.01183814,
      "auxiliary_loss_mlp": 0.01040433,
      "balance_loss_clip": 1.06731057,
      "balance_loss_mlp": 1.03081918,
      "epoch": 0.15523357181506645,
      "flos": 41282475569280.0,
      "grad_norm": 3.539608028173898,
      "language_loss": 0.75323218,
      "learning_rate": 3.837846250190206e-06,
      "loss": 0.77547467,
      "num_input_tokens_seen": 27308025,
      "step": 1291,
      "time_per_iteration": 2.824558734893799
    },
    {
      "auxiliary_loss_clip": 0.01177233,
      "auxiliary_loss_mlp": 0.00763566,
      "balance_loss_clip": 1.06400037,
      "balance_loss_mlp": 1.00029707,
      "epoch": 0.15535381470570553,
      "flos": 18478769806080.0,
      "grad_norm": 1.8855089560271818,
      "language_loss": 0.76977473,
      "learning_rate": 3.837538855811998e-06,
      "loss": 0.78918272,
      "num_input_tokens_seen": 27326200,
      "step": 1292,
      "time_per_iteration": 3.6285769939422607
    },
    {
      "auxiliary_loss_clip": 0.0119787,
      "auxiliary_loss_mlp": 0.01038562,
      "balance_loss_clip": 1.06666422,
      "balance_loss_mlp": 1.02914405,
      "epoch": 0.1554740575963446,
      "flos": 13918150759680.0,
      "grad_norm": 3.3668060777057485,
      "language_loss": 0.71046722,
      "learning_rate": 3.837231182680606e-06,
      "loss": 0.73283154,
      "num_input_tokens_seen": 27344165,
      "step": 1293,
      "time_per_iteration": 2.7683424949645996
    },
    {
      "auxiliary_loss_clip": 0.01166003,
      "auxiliary_loss_mlp": 0.01037571,
      "balance_loss_clip": 1.06018245,
      "balance_loss_mlp": 1.02796924,
      "epoch": 0.1555943004869837,
      "flos": 20847078161280.0,
      "grad_norm": 1.707917166792781,
      "language_loss": 0.76320803,
      "learning_rate": 3.836923230842706e-06,
      "loss": 0.78524375,
      "num_input_tokens_seen": 27363280,
      "step": 1294,
      "time_per_iteration": 2.7304301261901855
    },
    {
      "auxiliary_loss_clip": 0.0115072,
      "auxiliary_loss_mlp": 0.00763744,
      "balance_loss_clip": 1.06227279,
      "balance_loss_mlp": 1.00044036,
      "epoch": 0.1557145433776228,
      "flos": 22085888371200.0,
      "grad_norm": 3.072550388495152,
      "language_loss": 0.80833805,
      "learning_rate": 3.836615000345011e-06,
      "loss": 0.8274827,
      "num_input_tokens_seen": 27381460,
      "step": 1295,
      "time_per_iteration": 2.8480701446533203
    },
    {
      "auxiliary_loss_clip": 0.01235721,
      "auxiliary_loss_mlp": 0.01041385,
      "balance_loss_clip": 1.07141972,
      "balance_loss_mlp": 1.03225327,
      "epoch": 0.1558347862682619,
      "flos": 19791987039360.0,
      "grad_norm": 2.3965308453992655,
      "language_loss": 0.77999741,
      "learning_rate": 3.836306491234282e-06,
      "loss": 0.80276847,
      "num_input_tokens_seen": 27399310,
      "step": 1296,
      "time_per_iteration": 3.7891488075256348
    },
    {
      "auxiliary_loss_clip": 0.01238589,
      "auxiliary_loss_mlp": 0.01039867,
      "balance_loss_clip": 1.07521343,
      "balance_loss_mlp": 1.03033626,
      "epoch": 0.15595502915890097,
      "flos": 17237086508160.0,
      "grad_norm": 2.8552420843595008,
      "language_loss": 0.75559843,
      "learning_rate": 3.835997703557317e-06,
      "loss": 0.77838302,
      "num_input_tokens_seen": 27416050,
      "step": 1297,
      "time_per_iteration": 2.6595566272735596
    },
    {
      "auxiliary_loss_clip": 0.01213977,
      "auxiliary_loss_mlp": 0.01041615,
      "balance_loss_clip": 1.06985641,
      "balance_loss_mlp": 1.03237593,
      "epoch": 0.15607527204954008,
      "flos": 19719519350400.0,
      "grad_norm": 2.3712664749970087,
      "language_loss": 0.80003577,
      "learning_rate": 3.83568863736096e-06,
      "loss": 0.82259166,
      "num_input_tokens_seen": 27434920,
      "step": 1298,
      "time_per_iteration": 2.6806721687316895
    },
    {
      "auxiliary_loss_clip": 0.01224838,
      "auxiliary_loss_mlp": 0.00763343,
      "balance_loss_clip": 1.06827521,
      "balance_loss_mlp": 1.00035703,
      "epoch": 0.15619551494017916,
      "flos": 18515650095360.0,
      "grad_norm": 2.206928021923627,
      "language_loss": 0.89223379,
      "learning_rate": 3.8353792926920975e-06,
      "loss": 0.91211563,
      "num_input_tokens_seen": 27453570,
      "step": 1299,
      "time_per_iteration": 2.6099510192871094
    },
    {
      "auxiliary_loss_clip": 0.01191211,
      "auxiliary_loss_mlp": 0.01048363,
      "balance_loss_clip": 1.06177092,
      "balance_loss_mlp": 1.03771138,
      "epoch": 0.15631575783081825,
      "flos": 19902125116800.0,
      "grad_norm": 3.3156082154932465,
      "language_loss": 0.82018125,
      "learning_rate": 3.835069669597655e-06,
      "loss": 0.84257698,
      "num_input_tokens_seen": 27471960,
      "step": 1300,
      "time_per_iteration": 2.7503626346588135
    },
    {
      "auxiliary_loss_clip": 0.01192445,
      "auxiliary_loss_mlp": 0.00763938,
      "balance_loss_clip": 1.06550753,
      "balance_loss_mlp": 1.00035071,
      "epoch": 0.15643600072145733,
      "flos": 20777663128320.0,
      "grad_norm": 3.252783212021064,
      "language_loss": 0.79968965,
      "learning_rate": 3.834759768124603e-06,
      "loss": 0.8192535,
      "num_input_tokens_seen": 27490835,
      "step": 1301,
      "time_per_iteration": 2.741694450378418
    },
    {
      "auxiliary_loss_clip": 0.01244297,
      "auxiliary_loss_mlp": 0.01042449,
      "balance_loss_clip": 1.07565367,
      "balance_loss_mlp": 1.03239965,
      "epoch": 0.15655624361209644,
      "flos": 18546389159040.0,
      "grad_norm": 8.077176566306326,
      "language_loss": 0.76607323,
      "learning_rate": 3.834449588319953e-06,
      "loss": 0.78894067,
      "num_input_tokens_seen": 27508870,
      "step": 1302,
      "time_per_iteration": 2.616501808166504
    },
    {
      "auxiliary_loss_clip": 0.01199838,
      "auxiliary_loss_mlp": 0.01037638,
      "balance_loss_clip": 1.06786895,
      "balance_loss_mlp": 1.02872658,
      "epoch": 0.15667648650273552,
      "flos": 25229544727680.0,
      "grad_norm": 2.049725546662428,
      "language_loss": 0.85001844,
      "learning_rate": 3.834139130230758e-06,
      "loss": 0.87239319,
      "num_input_tokens_seen": 27528175,
      "step": 1303,
      "time_per_iteration": 2.7028040885925293
    },
    {
      "auxiliary_loss_clip": 0.01180219,
      "auxiliary_loss_mlp": 0.01037349,
      "balance_loss_clip": 1.06214499,
      "balance_loss_mlp": 1.02642369,
      "epoch": 0.1567967293933746,
      "flos": 24827093769600.0,
      "grad_norm": 1.7256389355930801,
      "language_loss": 0.80874825,
      "learning_rate": 3.833828393904117e-06,
      "loss": 0.83092391,
      "num_input_tokens_seen": 27548455,
      "step": 1304,
      "time_per_iteration": 2.8241395950317383
    },
    {
      "auxiliary_loss_clip": 0.01223159,
      "auxiliary_loss_mlp": 0.01039916,
      "balance_loss_clip": 1.07161236,
      "balance_loss_mlp": 1.0294559,
      "epoch": 0.15691697228401372,
      "flos": 19164555244800.0,
      "grad_norm": 2.22033365465348,
      "language_loss": 0.77733433,
      "learning_rate": 3.833517379387165e-06,
      "loss": 0.79996502,
      "num_input_tokens_seen": 27564910,
      "step": 1305,
      "time_per_iteration": 2.62599515914917
    },
    {
      "auxiliary_loss_clip": 0.01221572,
      "auxiliary_loss_mlp": 0.01039708,
      "balance_loss_clip": 1.0717783,
      "balance_loss_mlp": 1.02928936,
      "epoch": 0.1570372151746528,
      "flos": 24790931752320.0,
      "grad_norm": 3.037830496126954,
      "language_loss": 0.8858555,
      "learning_rate": 3.833206086727085e-06,
      "loss": 0.90846825,
      "num_input_tokens_seen": 27584260,
      "step": 1306,
      "time_per_iteration": 2.7506775856018066
    },
    {
      "auxiliary_loss_clip": 0.01204465,
      "auxiliary_loss_mlp": 0.01030235,
      "balance_loss_clip": 1.06423438,
      "balance_loss_mlp": 1.02019179,
      "epoch": 0.15715745806529188,
      "flos": 24863650836480.0,
      "grad_norm": 3.869648813960186,
      "language_loss": 0.70784014,
      "learning_rate": 3.8328945159710994e-06,
      "loss": 0.73018712,
      "num_input_tokens_seen": 27604440,
      "step": 1307,
      "time_per_iteration": 2.766730785369873
    },
    {
      "auxiliary_loss_clip": 0.01177866,
      "auxiliary_loss_mlp": 0.01044882,
      "balance_loss_clip": 1.06547415,
      "balance_loss_mlp": 1.03502953,
      "epoch": 0.157277700955931,
      "flos": 21872148491520.0,
      "grad_norm": 2.262120510321677,
      "language_loss": 0.88503373,
      "learning_rate": 3.832582667166473e-06,
      "loss": 0.90726119,
      "num_input_tokens_seen": 27624250,
      "step": 1308,
      "time_per_iteration": 2.726186752319336
    },
    {
      "auxiliary_loss_clip": 0.01209422,
      "auxiliary_loss_mlp": 0.01035739,
      "balance_loss_clip": 1.06963575,
      "balance_loss_mlp": 1.02523124,
      "epoch": 0.15739794384657008,
      "flos": 24533344344960.0,
      "grad_norm": 1.7497483175707327,
      "language_loss": 0.81769884,
      "learning_rate": 3.8322705403605125e-06,
      "loss": 0.84015048,
      "num_input_tokens_seen": 27644595,
      "step": 1309,
      "time_per_iteration": 2.8373959064483643
    },
    {
      "auxiliary_loss_clip": 0.01198129,
      "auxiliary_loss_mlp": 0.01041192,
      "balance_loss_clip": 1.0637064,
      "balance_loss_mlp": 1.03125024,
      "epoch": 0.15751818673720916,
      "flos": 17745329998080.0,
      "grad_norm": 3.3710327863980347,
      "language_loss": 0.81332433,
      "learning_rate": 3.831958135600568e-06,
      "loss": 0.83571756,
      "num_input_tokens_seen": 27662145,
      "step": 1310,
      "time_per_iteration": 2.6619887351989746
    },
    {
      "auxiliary_loss_clip": 0.01174399,
      "auxiliary_loss_mlp": 0.01041769,
      "balance_loss_clip": 1.06119967,
      "balance_loss_mlp": 1.03254855,
      "epoch": 0.15763842962784824,
      "flos": 17858520731520.0,
      "grad_norm": 1.9507519189167593,
      "language_loss": 0.79185605,
      "learning_rate": 3.831645452934032e-06,
      "loss": 0.81401771,
      "num_input_tokens_seen": 27680575,
      "step": 1311,
      "time_per_iteration": 2.786480665206909
    },
    {
      "auxiliary_loss_clip": 0.01182966,
      "auxiliary_loss_mlp": 0.01054224,
      "balance_loss_clip": 1.06709886,
      "balance_loss_mlp": 1.04332209,
      "epoch": 0.15775867251848735,
      "flos": 26980908059520.0,
      "grad_norm": 3.250403955669631,
      "language_loss": 0.80104673,
      "learning_rate": 3.831332492408336e-06,
      "loss": 0.82341868,
      "num_input_tokens_seen": 27701985,
      "step": 1312,
      "time_per_iteration": 2.8427894115448
    },
    {
      "auxiliary_loss_clip": 0.01170259,
      "auxiliary_loss_mlp": 0.01037103,
      "balance_loss_clip": 1.05998826,
      "balance_loss_mlp": 1.02719688,
      "epoch": 0.15787891540912644,
      "flos": 19240398812160.0,
      "grad_norm": 2.2062540656139418,
      "language_loss": 0.69118774,
      "learning_rate": 3.831019254070957e-06,
      "loss": 0.71326137,
      "num_input_tokens_seen": 27719770,
      "step": 1313,
      "time_per_iteration": 3.6954362392425537
    },
    {
      "auxiliary_loss_clip": 0.01184759,
      "auxiliary_loss_mlp": 0.01044468,
      "balance_loss_clip": 1.06569088,
      "balance_loss_mlp": 1.03440642,
      "epoch": 0.15799915829976552,
      "flos": 27271102037760.0,
      "grad_norm": 3.0110098786573043,
      "language_loss": 0.95101535,
      "learning_rate": 3.8307057379694135e-06,
      "loss": 0.97330767,
      "num_input_tokens_seen": 27739105,
      "step": 1314,
      "time_per_iteration": 2.847996950149536
    },
    {
      "auxiliary_loss_clip": 0.01177721,
      "auxiliary_loss_mlp": 0.01039135,
      "balance_loss_clip": 1.06402636,
      "balance_loss_mlp": 1.02815628,
      "epoch": 0.15811940119040463,
      "flos": 20405520270720.0,
      "grad_norm": 2.215821726652674,
      "language_loss": 0.82584238,
      "learning_rate": 3.830391944151264e-06,
      "loss": 0.84801102,
      "num_input_tokens_seen": 27754985,
      "step": 1315,
      "time_per_iteration": 3.693669319152832
    },
    {
      "auxiliary_loss_clip": 0.01210107,
      "auxiliary_loss_mlp": 0.01049868,
      "balance_loss_clip": 1.06939483,
      "balance_loss_mlp": 1.03946698,
      "epoch": 0.1582396440810437,
      "flos": 32599347661440.0,
      "grad_norm": 2.0920737992075273,
      "language_loss": 0.6760143,
      "learning_rate": 3.830077872664114e-06,
      "loss": 0.69861412,
      "num_input_tokens_seen": 27776110,
      "step": 1316,
      "time_per_iteration": 2.8186042308807373
    },
    {
      "auxiliary_loss_clip": 0.01210833,
      "auxiliary_loss_mlp": 0.01042464,
      "balance_loss_clip": 1.07153201,
      "balance_loss_mlp": 1.03162825,
      "epoch": 0.1583598869716828,
      "flos": 33800559310080.0,
      "grad_norm": 5.295303892752628,
      "language_loss": 0.73125529,
      "learning_rate": 3.829763523555604e-06,
      "loss": 0.75378823,
      "num_input_tokens_seen": 27796510,
      "step": 1317,
      "time_per_iteration": 3.7038321495056152
    },
    {
      "auxiliary_loss_clip": 0.01176084,
      "auxiliary_loss_mlp": 0.01042338,
      "balance_loss_clip": 1.0689857,
      "balance_loss_mlp": 1.03236651,
      "epoch": 0.15848012986232188,
      "flos": 24681332378880.0,
      "grad_norm": 2.139825293286583,
      "language_loss": 0.78035152,
      "learning_rate": 3.829448896873423e-06,
      "loss": 0.80253577,
      "num_input_tokens_seen": 27815610,
      "step": 1318,
      "time_per_iteration": 2.7624056339263916
    },
    {
      "auxiliary_loss_clip": 0.01215793,
      "auxiliary_loss_mlp": 0.01045858,
      "balance_loss_clip": 1.07057261,
      "balance_loss_mlp": 1.03477776,
      "epoch": 0.158600372752961,
      "flos": 22602068766720.0,
      "grad_norm": 1.8629985264079933,
      "language_loss": 0.78956717,
      "learning_rate": 3.829133992665299e-06,
      "loss": 0.81218368,
      "num_input_tokens_seen": 27834735,
      "step": 1319,
      "time_per_iteration": 2.730086088180542
    },
    {
      "auxiliary_loss_clip": 0.0116917,
      "auxiliary_loss_mlp": 0.01042625,
      "balance_loss_clip": 1.06350422,
      "balance_loss_mlp": 1.0327065,
      "epoch": 0.15872061564360007,
      "flos": 27927944092800.0,
      "grad_norm": 2.3833201408827804,
      "language_loss": 0.88854575,
      "learning_rate": 3.828818810979002e-06,
      "loss": 0.91066372,
      "num_input_tokens_seen": 27853065,
      "step": 1320,
      "time_per_iteration": 2.794381618499756
    },
    {
      "auxiliary_loss_clip": 0.01242708,
      "auxiliary_loss_mlp": 0.01041111,
      "balance_loss_clip": 1.0777123,
      "balance_loss_mlp": 1.03103805,
      "epoch": 0.15884085853423915,
      "flos": 23696805525120.0,
      "grad_norm": 19.798948113247814,
      "language_loss": 0.80768442,
      "learning_rate": 3.8285033518623454e-06,
      "loss": 0.83052266,
      "num_input_tokens_seen": 27873315,
      "step": 1321,
      "time_per_iteration": 3.635406255722046
    },
    {
      "auxiliary_loss_clip": 0.01244514,
      "auxiliary_loss_mlp": 0.01046989,
      "balance_loss_clip": 1.0755465,
      "balance_loss_mlp": 1.03542614,
      "epoch": 0.15896110142487826,
      "flos": 23112359331840.0,
      "grad_norm": 3.5174642266491136,
      "language_loss": 0.81609797,
      "learning_rate": 3.8281876153631845e-06,
      "loss": 0.83901304,
      "num_input_tokens_seen": 27890070,
      "step": 1322,
      "time_per_iteration": 2.7053568363189697
    },
    {
      "auxiliary_loss_clip": 0.01203944,
      "auxiliary_loss_mlp": 0.01040459,
      "balance_loss_clip": 1.06794596,
      "balance_loss_mlp": 1.0299921,
      "epoch": 0.15908134431551735,
      "flos": 14685238632960.0,
      "grad_norm": 1.7646443297823555,
      "language_loss": 0.64355087,
      "learning_rate": 3.827871601529416e-06,
      "loss": 0.66599488,
      "num_input_tokens_seen": 27908590,
      "step": 1323,
      "time_per_iteration": 2.658862590789795
    },
    {
      "auxiliary_loss_clip": 0.01225334,
      "auxiliary_loss_mlp": 0.01036544,
      "balance_loss_clip": 1.07016385,
      "balance_loss_mlp": 1.02612519,
      "epoch": 0.15920158720615643,
      "flos": 20193611984640.0,
      "grad_norm": 1.7393843351641038,
      "language_loss": 0.80554122,
      "learning_rate": 3.827555310408979e-06,
      "loss": 0.82816005,
      "num_input_tokens_seen": 27927985,
      "step": 1324,
      "time_per_iteration": 2.724809408187866
    },
    {
      "auxiliary_loss_clip": 0.01184913,
      "auxiliary_loss_mlp": 0.01042595,
      "balance_loss_clip": 1.0696516,
      "balance_loss_mlp": 1.03192616,
      "epoch": 0.1593218300967955,
      "flos": 24826626892800.0,
      "grad_norm": 1.7413182992264888,
      "language_loss": 0.8261404,
      "learning_rate": 3.827238742049854e-06,
      "loss": 0.84841549,
      "num_input_tokens_seen": 27948280,
      "step": 1325,
      "time_per_iteration": 2.773714065551758
    },
    {
      "auxiliary_loss_clip": 0.01220727,
      "auxiliary_loss_mlp": 0.01031993,
      "balance_loss_clip": 1.06910205,
      "balance_loss_mlp": 1.02150226,
      "epoch": 0.15944207298743462,
      "flos": 28328707111680.0,
      "grad_norm": 2.604729534843354,
      "language_loss": 0.51504719,
      "learning_rate": 3.826921896500066e-06,
      "loss": 0.53757435,
      "num_input_tokens_seen": 27969565,
      "step": 1326,
      "time_per_iteration": 2.781895399093628
    },
    {
      "auxiliary_loss_clip": 0.01216377,
      "auxiliary_loss_mlp": 0.01039901,
      "balance_loss_clip": 1.07386649,
      "balance_loss_mlp": 1.02907681,
      "epoch": 0.1595623158780737,
      "flos": 22964838174720.0,
      "grad_norm": 2.120228412654577,
      "language_loss": 0.78463459,
      "learning_rate": 3.826604773807678e-06,
      "loss": 0.80719739,
      "num_input_tokens_seen": 27987540,
      "step": 1327,
      "time_per_iteration": 2.6955485343933105
    },
    {
      "auxiliary_loss_clip": 0.01228059,
      "auxiliary_loss_mlp": 0.01045774,
      "balance_loss_clip": 1.07438171,
      "balance_loss_mlp": 1.03438413,
      "epoch": 0.1596825587687128,
      "flos": 19710540950400.0,
      "grad_norm": 2.877019888263574,
      "language_loss": 0.73695219,
      "learning_rate": 3.826287374020798e-06,
      "loss": 0.75969052,
      "num_input_tokens_seen": 28002345,
      "step": 1328,
      "time_per_iteration": 2.7012670040130615
    },
    {
      "auxiliary_loss_clip": 0.0123083,
      "auxiliary_loss_mlp": 0.01044184,
      "balance_loss_clip": 1.07883334,
      "balance_loss_mlp": 1.03307343,
      "epoch": 0.1598028016593519,
      "flos": 22637727993600.0,
      "grad_norm": 2.766551687917594,
      "language_loss": 0.81772685,
      "learning_rate": 3.825969697187575e-06,
      "loss": 0.84047699,
      "num_input_tokens_seen": 28021675,
      "step": 1329,
      "time_per_iteration": 2.6940767765045166
    },
    {
      "auxiliary_loss_clip": 0.01226964,
      "auxiliary_loss_mlp": 0.01040337,
      "balance_loss_clip": 1.07740855,
      "balance_loss_mlp": 1.02973342,
      "epoch": 0.15992304454999098,
      "flos": 20482908122880.0,
      "grad_norm": 2.249445227439106,
      "language_loss": 0.69227314,
      "learning_rate": 3.8256517433562015e-06,
      "loss": 0.71494621,
      "num_input_tokens_seen": 28039615,
      "step": 1330,
      "time_per_iteration": 2.662257194519043
    },
    {
      "auxiliary_loss_clip": 0.01214566,
      "auxiliary_loss_mlp": 0.00763759,
      "balance_loss_clip": 1.07261181,
      "balance_loss_mlp": 1.00053334,
      "epoch": 0.16004328744063007,
      "flos": 17676094533120.0,
      "grad_norm": 2.8332517129359926,
      "language_loss": 0.91109848,
      "learning_rate": 3.82533351257491e-06,
      "loss": 0.93088174,
      "num_input_tokens_seen": 28057565,
      "step": 1331,
      "time_per_iteration": 2.7850165367126465
    },
    {
      "auxiliary_loss_clip": 0.01149246,
      "auxiliary_loss_mlp": 0.0103825,
      "balance_loss_clip": 1.05795097,
      "balance_loss_mlp": 1.02848101,
      "epoch": 0.16016353033126918,
      "flos": 24098717779200.0,
      "grad_norm": 2.989050067849571,
      "language_loss": 0.88409317,
      "learning_rate": 3.825015004891975e-06,
      "loss": 0.90596813,
      "num_input_tokens_seen": 28076305,
      "step": 1332,
      "time_per_iteration": 2.86081862449646
    },
    {
      "auxiliary_loss_clip": 0.01170541,
      "auxiliary_loss_mlp": 0.01041403,
      "balance_loss_clip": 1.06577921,
      "balance_loss_mlp": 1.03097856,
      "epoch": 0.16028377322190826,
      "flos": 27634841112960.0,
      "grad_norm": 2.1090690514033175,
      "language_loss": 0.75792754,
      "learning_rate": 3.824696220355716e-06,
      "loss": 0.78004706,
      "num_input_tokens_seen": 28097895,
      "step": 1333,
      "time_per_iteration": 2.9024746417999268
    },
    {
      "auxiliary_loss_clip": 0.01131243,
      "auxiliary_loss_mlp": 0.01038034,
      "balance_loss_clip": 1.05109453,
      "balance_loss_mlp": 1.02764511,
      "epoch": 0.16040401611254734,
      "flos": 20961202648320.0,
      "grad_norm": 3.4284767348407237,
      "language_loss": 0.79038,
      "learning_rate": 3.824377159014491e-06,
      "loss": 0.81207275,
      "num_input_tokens_seen": 28118790,
      "step": 1334,
      "time_per_iteration": 2.9366936683654785
    },
    {
      "auxiliary_loss_clip": 0.012079,
      "auxiliary_loss_mlp": 0.0103914,
      "balance_loss_clip": 1.07225943,
      "balance_loss_mlp": 1.02916837,
      "epoch": 0.16052425900318643,
      "flos": 21247051080960.0,
      "grad_norm": 2.0725870082922215,
      "language_loss": 0.84827816,
      "learning_rate": 3.824057820916702e-06,
      "loss": 0.87074852,
      "num_input_tokens_seen": 28135995,
      "step": 1335,
      "time_per_iteration": 2.8178834915161133
    },
    {
      "auxiliary_loss_clip": 0.01196639,
      "auxiliary_loss_mlp": 0.01039657,
      "balance_loss_clip": 1.06694007,
      "balance_loss_mlp": 1.0286417,
      "epoch": 0.16064450189382554,
      "flos": 15524004096000.0,
      "grad_norm": 2.2333534633057655,
      "language_loss": 0.71469414,
      "learning_rate": 3.8237382061107904e-06,
      "loss": 0.73705709,
      "num_input_tokens_seen": 28152715,
      "step": 1336,
      "time_per_iteration": 2.7475035190582275
    },
    {
      "auxiliary_loss_clip": 0.01192082,
      "auxiliary_loss_mlp": 0.01043621,
      "balance_loss_clip": 1.06698489,
      "balance_loss_mlp": 1.03295755,
      "epoch": 0.16076474478446462,
      "flos": 21178497974400.0,
      "grad_norm": 1.9952534210986612,
      "language_loss": 0.78799605,
      "learning_rate": 3.823418314645243e-06,
      "loss": 0.8103531,
      "num_input_tokens_seen": 28171590,
      "step": 1337,
      "time_per_iteration": 2.7843141555786133
    },
    {
      "auxiliary_loss_clip": 0.01211746,
      "auxiliary_loss_mlp": 0.01041603,
      "balance_loss_clip": 1.06998801,
      "balance_loss_mlp": 1.03083205,
      "epoch": 0.1608849876751037,
      "flos": 18366476912640.0,
      "grad_norm": 2.477141918698906,
      "language_loss": 0.75844598,
      "learning_rate": 3.823098146568588e-06,
      "loss": 0.78097951,
      "num_input_tokens_seen": 28191295,
      "step": 1338,
      "time_per_iteration": 2.7326500415802
    },
    {
      "auxiliary_loss_clip": 0.01181269,
      "auxiliary_loss_mlp": 0.01039744,
      "balance_loss_clip": 1.06453419,
      "balance_loss_mlp": 1.02918243,
      "epoch": 0.1610052305657428,
      "flos": 29497024880640.0,
      "grad_norm": 1.8803205039817321,
      "language_loss": 0.71215272,
      "learning_rate": 3.822777701929394e-06,
      "loss": 0.73436284,
      "num_input_tokens_seen": 28213120,
      "step": 1339,
      "time_per_iteration": 3.6894595623016357
    },
    {
      "auxiliary_loss_clip": 0.01192874,
      "auxiliary_loss_mlp": 0.01039668,
      "balance_loss_clip": 1.06594372,
      "balance_loss_mlp": 1.02975023,
      "epoch": 0.1611254734563819,
      "flos": 26797871329920.0,
      "grad_norm": 1.8072681641421873,
      "language_loss": 0.73269594,
      "learning_rate": 3.8224569807762714e-06,
      "loss": 0.75502133,
      "num_input_tokens_seen": 28232440,
      "step": 1340,
      "time_per_iteration": 2.843238353729248
    },
    {
      "auxiliary_loss_clip": 0.01213023,
      "auxiliary_loss_mlp": 0.01044711,
      "balance_loss_clip": 1.07001936,
      "balance_loss_mlp": 1.03420901,
      "epoch": 0.16124571634702098,
      "flos": 22419570741120.0,
      "grad_norm": 2.205374005223593,
      "language_loss": 0.76309204,
      "learning_rate": 3.822135983157873e-06,
      "loss": 0.78566945,
      "num_input_tokens_seen": 28251715,
      "step": 1341,
      "time_per_iteration": 3.6165006160736084
    },
    {
      "auxiliary_loss_clip": 0.01170759,
      "auxiliary_loss_mlp": 0.01031527,
      "balance_loss_clip": 1.06181002,
      "balance_loss_mlp": 1.02178168,
      "epoch": 0.16136595923766006,
      "flos": 10999116103680.0,
      "grad_norm": 1.9694264373378594,
      "language_loss": 0.83960873,
      "learning_rate": 3.821814709122896e-06,
      "loss": 0.86163157,
      "num_input_tokens_seen": 28269765,
      "step": 1342,
      "time_per_iteration": 2.7520394325256348
    },
    {
      "auxiliary_loss_clip": 0.01220928,
      "auxiliary_loss_mlp": 0.01039147,
      "balance_loss_clip": 1.0739634,
      "balance_loss_mlp": 1.02924061,
      "epoch": 0.16148620212829917,
      "flos": 21214983214080.0,
      "grad_norm": 2.2178300021935797,
      "language_loss": 0.84462023,
      "learning_rate": 3.821493158720076e-06,
      "loss": 0.867221,
      "num_input_tokens_seen": 28288870,
      "step": 1343,
      "time_per_iteration": 3.6539344787597656
    },
    {
      "auxiliary_loss_clip": 0.01209338,
      "auxiliary_loss_mlp": 0.01038094,
      "balance_loss_clip": 1.06893063,
      "balance_loss_mlp": 1.02635181,
      "epoch": 0.16160644501893826,
      "flos": 16758468760320.0,
      "grad_norm": 4.110273010357713,
      "language_loss": 0.73083299,
      "learning_rate": 3.821171331998191e-06,
      "loss": 0.75330728,
      "num_input_tokens_seen": 28305400,
      "step": 1344,
      "time_per_iteration": 2.7857062816619873
    },
    {
      "auxiliary_loss_clip": 0.01125828,
      "auxiliary_loss_mlp": 0.01002108,
      "balance_loss_clip": 1.05289745,
      "balance_loss_mlp": 0.99940223,
      "epoch": 0.16172668790957734,
      "flos": 64444967308800.0,
      "grad_norm": 0.7110607129221049,
      "language_loss": 0.54475403,
      "learning_rate": 3.820849229006064e-06,
      "loss": 0.56603336,
      "num_input_tokens_seen": 28373150,
      "step": 1345,
      "time_per_iteration": 3.4289584159851074
    },
    {
      "auxiliary_loss_clip": 0.01141907,
      "auxiliary_loss_mlp": 0.01043736,
      "balance_loss_clip": 1.05999732,
      "balance_loss_mlp": 1.03287625,
      "epoch": 0.16184693080021645,
      "flos": 23257689759360.0,
      "grad_norm": 3.0198389423680063,
      "language_loss": 0.70496601,
      "learning_rate": 3.8205268497925564e-06,
      "loss": 0.7268225,
      "num_input_tokens_seen": 28393620,
      "step": 1346,
      "time_per_iteration": 3.864487409591675
    },
    {
      "auxiliary_loss_clip": 0.01226772,
      "auxiliary_loss_mlp": 0.01042875,
      "balance_loss_clip": 1.07302296,
      "balance_loss_mlp": 1.03160393,
      "epoch": 0.16196717369085553,
      "flos": 17451113696640.0,
      "grad_norm": 4.311440412929312,
      "language_loss": 0.78833592,
      "learning_rate": 3.8202041944065725e-06,
      "loss": 0.81103235,
      "num_input_tokens_seen": 28409440,
      "step": 1347,
      "time_per_iteration": 2.716261863708496
    },
    {
      "auxiliary_loss_clip": 0.01225864,
      "auxiliary_loss_mlp": 0.01053606,
      "balance_loss_clip": 1.07361639,
      "balance_loss_mlp": 1.04321098,
      "epoch": 0.16208741658149461,
      "flos": 23873377806720.0,
      "grad_norm": 2.6809140318305014,
      "language_loss": 0.73911572,
      "learning_rate": 3.819881262897061e-06,
      "loss": 0.76191044,
      "num_input_tokens_seen": 28427575,
      "step": 1348,
      "time_per_iteration": 2.7973580360412598
    },
    {
      "auxiliary_loss_clip": 0.01165384,
      "auxiliary_loss_mlp": 0.01045823,
      "balance_loss_clip": 1.0666039,
      "balance_loss_mlp": 1.03454626,
      "epoch": 0.1622076594721337,
      "flos": 25884806584320.0,
      "grad_norm": 9.766165060884148,
      "language_loss": 0.73494023,
      "learning_rate": 3.819558055313008e-06,
      "loss": 0.7570523,
      "num_input_tokens_seen": 28448260,
      "step": 1349,
      "time_per_iteration": 2.8439908027648926
    },
    {
      "auxiliary_loss_clip": 0.01187279,
      "auxiliary_loss_mlp": 0.01048726,
      "balance_loss_clip": 1.06554735,
      "balance_loss_mlp": 1.0379492,
      "epoch": 0.1623279023627728,
      "flos": 21539759011200.0,
      "grad_norm": 2.795476301767974,
      "language_loss": 0.77399701,
      "learning_rate": 3.819234571703444e-06,
      "loss": 0.79635704,
      "num_input_tokens_seen": 28467085,
      "step": 1350,
      "time_per_iteration": 2.8044135570526123
    },
    {
      "auxiliary_loss_clip": 0.01224932,
      "auxiliary_loss_mlp": 0.01043968,
      "balance_loss_clip": 1.07149911,
      "balance_loss_mlp": 1.03276801,
      "epoch": 0.1624481452534119,
      "flos": 22085421494400.0,
      "grad_norm": 1.860563525031552,
      "language_loss": 0.85695893,
      "learning_rate": 3.8189108121174435e-06,
      "loss": 0.87964791,
      "num_input_tokens_seen": 28486850,
      "step": 1351,
      "time_per_iteration": 2.725402593612671
    },
    {
      "auxiliary_loss_clip": 0.01191085,
      "auxiliary_loss_mlp": 0.01048157,
      "balance_loss_clip": 1.06926489,
      "balance_loss_mlp": 1.03648663,
      "epoch": 0.16256838814405097,
      "flos": 27087490690560.0,
      "grad_norm": 1.640271228891387,
      "language_loss": 0.83470201,
      "learning_rate": 3.818586776604118e-06,
      "loss": 0.85709441,
      "num_input_tokens_seen": 28507490,
      "step": 1352,
      "time_per_iteration": 2.7918500900268555
    },
    {
      "auxiliary_loss_clip": 0.0119391,
      "auxiliary_loss_mlp": 0.01049434,
      "balance_loss_clip": 1.0656383,
      "balance_loss_mlp": 1.03846681,
      "epoch": 0.16268863103469008,
      "flos": 20120354196480.0,
      "grad_norm": 2.0187691126875422,
      "language_loss": 0.61723363,
      "learning_rate": 3.818262465212625e-06,
      "loss": 0.63966703,
      "num_input_tokens_seen": 28527615,
      "step": 1353,
      "time_per_iteration": 2.716280698776245
    },
    {
      "auxiliary_loss_clip": 0.01215807,
      "auxiliary_loss_mlp": 0.01045816,
      "balance_loss_clip": 1.07156491,
      "balance_loss_mlp": 1.03484309,
      "epoch": 0.16280887392532917,
      "flos": 18332792933760.0,
      "grad_norm": 4.634854756480655,
      "language_loss": 0.76968503,
      "learning_rate": 3.817937877992161e-06,
      "loss": 0.7923013,
      "num_input_tokens_seen": 28544910,
      "step": 1354,
      "time_per_iteration": 2.65285587310791
    },
    {
      "auxiliary_loss_clip": 0.01122763,
      "auxiliary_loss_mlp": 0.01047075,
      "balance_loss_clip": 1.05544305,
      "balance_loss_mlp": 1.03633451,
      "epoch": 0.16292911681596825,
      "flos": 11874330892800.0,
      "grad_norm": 2.504659251154707,
      "language_loss": 0.85042846,
      "learning_rate": 3.817613014991967e-06,
      "loss": 0.87212688,
      "num_input_tokens_seen": 28561050,
      "step": 1355,
      "time_per_iteration": 2.849651336669922
    },
    {
      "auxiliary_loss_clip": 0.01199315,
      "auxiliary_loss_mlp": 0.01037454,
      "balance_loss_clip": 1.06680024,
      "balance_loss_mlp": 1.02593899,
      "epoch": 0.16304935970660733,
      "flos": 26103466627200.0,
      "grad_norm": 1.972678483467851,
      "language_loss": 0.76902425,
      "learning_rate": 3.817287876261323e-06,
      "loss": 0.79139191,
      "num_input_tokens_seen": 28581385,
      "step": 1356,
      "time_per_iteration": 2.736440658569336
    },
    {
      "auxiliary_loss_clip": 0.01194463,
      "auxiliary_loss_mlp": 0.01037294,
      "balance_loss_clip": 1.06500936,
      "balance_loss_mlp": 1.02657676,
      "epoch": 0.16316960259724644,
      "flos": 29351945848320.0,
      "grad_norm": 4.810281947427564,
      "language_loss": 0.80060101,
      "learning_rate": 3.816962461849553e-06,
      "loss": 0.82291853,
      "num_input_tokens_seen": 28603255,
      "step": 1357,
      "time_per_iteration": 2.8114283084869385
    },
    {
      "auxiliary_loss_clip": 0.01242412,
      "auxiliary_loss_mlp": 0.01049839,
      "balance_loss_clip": 1.0759263,
      "balance_loss_mlp": 1.03855014,
      "epoch": 0.16328984548788553,
      "flos": 20886759711360.0,
      "grad_norm": 2.6218383526713143,
      "language_loss": 0.84446502,
      "learning_rate": 3.8166367718060235e-06,
      "loss": 0.86738753,
      "num_input_tokens_seen": 28623145,
      "step": 1358,
      "time_per_iteration": 2.636829376220703
    },
    {
      "auxiliary_loss_clip": 0.01234404,
      "auxiliary_loss_mlp": 0.01043841,
      "balance_loss_clip": 1.07013047,
      "balance_loss_mlp": 1.0327369,
      "epoch": 0.1634100883785246,
      "flos": 18041090584320.0,
      "grad_norm": 2.6881533296006395,
      "language_loss": 0.76808119,
      "learning_rate": 3.816310806180139e-06,
      "loss": 0.79086363,
      "num_input_tokens_seen": 28641555,
      "step": 1359,
      "time_per_iteration": 2.6682662963867188
    },
    {
      "auxiliary_loss_clip": 0.01145071,
      "auxiliary_loss_mlp": 0.01049554,
      "balance_loss_clip": 1.06024098,
      "balance_loss_mlp": 1.03815722,
      "epoch": 0.16353033126916372,
      "flos": 24572128055040.0,
      "grad_norm": 2.3795167538584967,
      "language_loss": 0.8094517,
      "learning_rate": 3.81598456502135e-06,
      "loss": 0.83139795,
      "num_input_tokens_seen": 28661575,
      "step": 1360,
      "time_per_iteration": 2.812697172164917
    },
    {
      "auxiliary_loss_clip": 0.01178488,
      "auxiliary_loss_mlp": 0.01041395,
      "balance_loss_clip": 1.06500936,
      "balance_loss_mlp": 1.03044021,
      "epoch": 0.1636505741598028,
      "flos": 19892895321600.0,
      "grad_norm": 3.2976413973537997,
      "language_loss": 0.87060928,
      "learning_rate": 3.8156580483791455e-06,
      "loss": 0.89280808,
      "num_input_tokens_seen": 28676765,
      "step": 1361,
      "time_per_iteration": 2.7810261249542236
    },
    {
      "auxiliary_loss_clip": 0.01179723,
      "auxiliary_loss_mlp": 0.01045949,
      "balance_loss_clip": 1.05943298,
      "balance_loss_mlp": 1.03566766,
      "epoch": 0.16377081705044189,
      "flos": 28402611344640.0,
      "grad_norm": 2.754503514898632,
      "language_loss": 0.76979089,
      "learning_rate": 3.815331256303059e-06,
      "loss": 0.79204762,
      "num_input_tokens_seen": 28696795,
      "step": 1362,
      "time_per_iteration": 2.851433277130127
    },
    {
      "auxiliary_loss_clip": 0.01229608,
      "auxiliary_loss_mlp": 0.010474,
      "balance_loss_clip": 1.07287109,
      "balance_loss_mlp": 1.03686225,
      "epoch": 0.163891059941081,
      "flos": 21908059113600.0,
      "grad_norm": 2.2378937373546903,
      "language_loss": 0.76927662,
      "learning_rate": 3.815004188842665e-06,
      "loss": 0.79204673,
      "num_input_tokens_seen": 28714835,
      "step": 1363,
      "time_per_iteration": 2.7375855445861816
    },
    {
      "auxiliary_loss_clip": 0.01196052,
      "auxiliary_loss_mlp": 0.0103859,
      "balance_loss_clip": 1.06594419,
      "balance_loss_mlp": 1.02789712,
      "epoch": 0.16401130283172008,
      "flos": 26797619934720.0,
      "grad_norm": 2.1638244584892568,
      "language_loss": 0.79875541,
      "learning_rate": 3.814676846047578e-06,
      "loss": 0.82110184,
      "num_input_tokens_seen": 28735710,
      "step": 1364,
      "time_per_iteration": 3.7507052421569824
    },
    {
      "auxiliary_loss_clip": 0.01212184,
      "auxiliary_loss_mlp": 0.0104387,
      "balance_loss_clip": 1.07194424,
      "balance_loss_mlp": 1.03274202,
      "epoch": 0.16413154572235916,
      "flos": 32997417160320.0,
      "grad_norm": 6.844459258574313,
      "language_loss": 0.69340253,
      "learning_rate": 3.8143492279674565e-06,
      "loss": 0.71596313,
      "num_input_tokens_seen": 28758405,
      "step": 1365,
      "time_per_iteration": 2.8040096759796143
    },
    {
      "auxiliary_loss_clip": 0.01122743,
      "auxiliary_loss_mlp": 0.01011309,
      "balance_loss_clip": 1.04231131,
      "balance_loss_mlp": 1.00875843,
      "epoch": 0.16425178861299825,
      "flos": 40113622074240.0,
      "grad_norm": 0.8353485920304119,
      "language_loss": 0.58354813,
      "learning_rate": 3.8140213346519997e-06,
      "loss": 0.60488868,
      "num_input_tokens_seen": 28809000,
      "step": 1366,
      "time_per_iteration": 3.031022787094116
    },
    {
      "auxiliary_loss_clip": 0.01203816,
      "auxiliary_loss_mlp": 0.01043396,
      "balance_loss_clip": 1.06322908,
      "balance_loss_mlp": 1.03227925,
      "epoch": 0.16437203150363736,
      "flos": 25447486498560.0,
      "grad_norm": 2.441060423435663,
      "language_loss": 0.76804399,
      "learning_rate": 3.813693166150948e-06,
      "loss": 0.79051614,
      "num_input_tokens_seen": 28829210,
      "step": 1367,
      "time_per_iteration": 3.6873435974121094
    },
    {
      "auxiliary_loss_clip": 0.0116581,
      "auxiliary_loss_mlp": 0.01043477,
      "balance_loss_clip": 1.06437624,
      "balance_loss_mlp": 1.03288555,
      "epoch": 0.16449227439427644,
      "flos": 23476888506240.0,
      "grad_norm": 3.0549082411264536,
      "language_loss": 0.85558426,
      "learning_rate": 3.813364722514086e-06,
      "loss": 0.87767714,
      "num_input_tokens_seen": 28847545,
      "step": 1368,
      "time_per_iteration": 2.8001441955566406
    },
    {
      "auxiliary_loss_clip": 0.01225041,
      "auxiliary_loss_mlp": 0.01048435,
      "balance_loss_clip": 1.07138324,
      "balance_loss_mlp": 1.03802204,
      "epoch": 0.16461251728491552,
      "flos": 13545217802880.0,
      "grad_norm": 2.0402265509114526,
      "language_loss": 0.80489886,
      "learning_rate": 3.8130360037912368e-06,
      "loss": 0.82763362,
      "num_input_tokens_seen": 28863990,
      "step": 1369,
      "time_per_iteration": 3.5950400829315186
    },
    {
      "auxiliary_loss_clip": 0.01196265,
      "auxiliary_loss_mlp": 0.0103216,
      "balance_loss_clip": 1.06617284,
      "balance_loss_mlp": 1.02198529,
      "epoch": 0.16473276017555463,
      "flos": 23003298662400.0,
      "grad_norm": 2.9160355240415052,
      "language_loss": 0.81837249,
      "learning_rate": 3.812707010032268e-06,
      "loss": 0.84065676,
      "num_input_tokens_seen": 28883045,
      "step": 1370,
      "time_per_iteration": 2.800161123275757
    },
    {
      "auxiliary_loss_clip": 0.01226887,
      "auxiliary_loss_mlp": 0.01040578,
      "balance_loss_clip": 1.07085407,
      "balance_loss_mlp": 1.03003943,
      "epoch": 0.16485300306619372,
      "flos": 24790680357120.0,
      "grad_norm": 2.1171271352331638,
      "language_loss": 0.79413545,
      "learning_rate": 3.8123777412870863e-06,
      "loss": 0.81681013,
      "num_input_tokens_seen": 28902545,
      "step": 1371,
      "time_per_iteration": 2.6268064975738525
    },
    {
      "auxiliary_loss_clip": 0.01192848,
      "auxiliary_loss_mlp": 0.01044655,
      "balance_loss_clip": 1.06644642,
      "balance_loss_mlp": 1.03412247,
      "epoch": 0.1649732459568328,
      "flos": 21106497162240.0,
      "grad_norm": 3.7667673668019437,
      "language_loss": 0.77885604,
      "learning_rate": 3.812048197605643e-06,
      "loss": 0.80123103,
      "num_input_tokens_seen": 28921440,
      "step": 1372,
      "time_per_iteration": 3.6824162006378174
    },
    {
      "auxiliary_loss_clip": 0.0120765,
      "auxiliary_loss_mlp": 0.01037432,
      "balance_loss_clip": 1.06988466,
      "balance_loss_mlp": 1.02720404,
      "epoch": 0.16509348884747188,
      "flos": 20266726118400.0,
      "grad_norm": 2.891681924196691,
      "language_loss": 0.81321782,
      "learning_rate": 3.8117183790379277e-06,
      "loss": 0.83566868,
      "num_input_tokens_seen": 28939890,
      "step": 1373,
      "time_per_iteration": 2.742400884628296
    },
    {
      "auxiliary_loss_clip": 0.0119572,
      "auxiliary_loss_mlp": 0.01037875,
      "balance_loss_clip": 1.06294513,
      "balance_loss_mlp": 1.02761126,
      "epoch": 0.165213731738111,
      "flos": 11035493602560.0,
      "grad_norm": 3.3577099859829227,
      "language_loss": 0.93968827,
      "learning_rate": 3.811388285633976e-06,
      "loss": 0.96202421,
      "num_input_tokens_seen": 28955875,
      "step": 1374,
      "time_per_iteration": 2.703166961669922
    },
    {
      "auxiliary_loss_clip": 0.01178613,
      "auxiliary_loss_mlp": 0.01037619,
      "balance_loss_clip": 1.06195891,
      "balance_loss_mlp": 1.02643704,
      "epoch": 0.16533397462875007,
      "flos": 29972051268480.0,
      "grad_norm": 6.069163927117445,
      "language_loss": 0.61868393,
      "learning_rate": 3.811057917443861e-06,
      "loss": 0.64084625,
      "num_input_tokens_seen": 28975140,
      "step": 1375,
      "time_per_iteration": 2.968756914138794
    },
    {
      "auxiliary_loss_clip": 0.01091852,
      "auxiliary_loss_mlp": 0.01000558,
      "balance_loss_clip": 1.05226362,
      "balance_loss_mlp": 0.99806672,
      "epoch": 0.16545421751938916,
      "flos": 65556763027200.0,
      "grad_norm": 0.8473767868618682,
      "language_loss": 0.68203336,
      "learning_rate": 3.8107272745177e-06,
      "loss": 0.70295745,
      "num_input_tokens_seen": 29047470,
      "step": 1376,
      "time_per_iteration": 3.6078577041625977
    },
    {
      "auxiliary_loss_clip": 0.01216311,
      "auxiliary_loss_mlp": 0.0104078,
      "balance_loss_clip": 1.06830215,
      "balance_loss_mlp": 1.03055763,
      "epoch": 0.16557446041002827,
      "flos": 22492361652480.0,
      "grad_norm": 1.8761221814502427,
      "language_loss": 0.78773642,
      "learning_rate": 3.8103963569056513e-06,
      "loss": 0.81030732,
      "num_input_tokens_seen": 29066605,
      "step": 1377,
      "time_per_iteration": 3.2116615772247314
    },
    {
      "auxiliary_loss_clip": 0.01180655,
      "auxiliary_loss_mlp": 0.01041116,
      "balance_loss_clip": 1.06426549,
      "balance_loss_mlp": 1.03105474,
      "epoch": 0.16569470330066735,
      "flos": 24602723464320.0,
      "grad_norm": 1.8379872166471487,
      "language_loss": 0.88198179,
      "learning_rate": 3.8100651646579146e-06,
      "loss": 0.90419948,
      "num_input_tokens_seen": 29085815,
      "step": 1378,
      "time_per_iteration": 2.801145076751709
    },
    {
      "auxiliary_loss_clip": 0.01208533,
      "auxiliary_loss_mlp": 0.01040582,
      "balance_loss_clip": 1.06622934,
      "balance_loss_mlp": 1.03014505,
      "epoch": 0.16581494619130643,
      "flos": 15006207588480.0,
      "grad_norm": 2.685035914671899,
      "language_loss": 0.92834866,
      "learning_rate": 3.8097336978247317e-06,
      "loss": 0.9508397,
      "num_input_tokens_seen": 29102520,
      "step": 1379,
      "time_per_iteration": 2.7311248779296875
    },
    {
      "auxiliary_loss_clip": 0.0117249,
      "auxiliary_loss_mlp": 0.01037976,
      "balance_loss_clip": 1.06392694,
      "balance_loss_mlp": 1.02719998,
      "epoch": 0.16593518908194552,
      "flos": 17420338719360.0,
      "grad_norm": 2.462941189599797,
      "language_loss": 0.88916302,
      "learning_rate": 3.8094019564563854e-06,
      "loss": 0.91126764,
      "num_input_tokens_seen": 29119450,
      "step": 1380,
      "time_per_iteration": 2.763680934906006
    },
    {
      "auxiliary_loss_clip": 0.01218625,
      "auxiliary_loss_mlp": 0.01044076,
      "balance_loss_clip": 1.06776357,
      "balance_loss_mlp": 1.034307,
      "epoch": 0.16605543197258463,
      "flos": 20412631163520.0,
      "grad_norm": 7.8191058914117715,
      "language_loss": 0.75027919,
      "learning_rate": 3.809069940603201e-06,
      "loss": 0.77290618,
      "num_input_tokens_seen": 29137405,
      "step": 1381,
      "time_per_iteration": 2.6742916107177734
    },
    {
      "auxiliary_loss_clip": 0.0120561,
      "auxiliary_loss_mlp": 0.01031336,
      "balance_loss_clip": 1.06688344,
      "balance_loss_mlp": 1.02100098,
      "epoch": 0.1661756748632237,
      "flos": 14209745368320.0,
      "grad_norm": 2.9723122073136055,
      "language_loss": 0.77936625,
      "learning_rate": 3.8087376503155452e-06,
      "loss": 0.8017357,
      "num_input_tokens_seen": 29154890,
      "step": 1382,
      "time_per_iteration": 2.750574827194214
    },
    {
      "auxiliary_loss_clip": 0.01123072,
      "auxiliary_loss_mlp": 0.01011359,
      "balance_loss_clip": 1.04043937,
      "balance_loss_mlp": 1.00893867,
      "epoch": 0.1662959177538628,
      "flos": 66080877350400.0,
      "grad_norm": 0.9085022852445599,
      "language_loss": 0.56192756,
      "learning_rate": 3.808405085643826e-06,
      "loss": 0.58327186,
      "num_input_tokens_seen": 29219770,
      "step": 1383,
      "time_per_iteration": 3.3186733722686768
    },
    {
      "auxiliary_loss_clip": 0.01194882,
      "auxiliary_loss_mlp": 0.01043305,
      "balance_loss_clip": 1.06884503,
      "balance_loss_mlp": 1.03277278,
      "epoch": 0.1664161606445019,
      "flos": 20740567357440.0,
      "grad_norm": 5.451937161949042,
      "language_loss": 0.89114922,
      "learning_rate": 3.8080722466384925e-06,
      "loss": 0.91353112,
      "num_input_tokens_seen": 29237620,
      "step": 1384,
      "time_per_iteration": 2.762697219848633
    },
    {
      "auxiliary_loss_clip": 0.01204354,
      "auxiliary_loss_mlp": 0.01037654,
      "balance_loss_clip": 1.06530142,
      "balance_loss_mlp": 1.02653813,
      "epoch": 0.166536403535141,
      "flos": 25260930236160.0,
      "grad_norm": 2.478530537661697,
      "language_loss": 0.70915508,
      "learning_rate": 3.8077391333500376e-06,
      "loss": 0.73157525,
      "num_input_tokens_seen": 29256760,
      "step": 1385,
      "time_per_iteration": 2.7823374271392822
    },
    {
      "auxiliary_loss_clip": 0.01204215,
      "auxiliary_loss_mlp": 0.01045856,
      "balance_loss_clip": 1.07110524,
      "balance_loss_mlp": 1.03544879,
      "epoch": 0.16665664642578007,
      "flos": 25447450584960.0,
      "grad_norm": 2.032593178827267,
      "language_loss": 0.7654376,
      "learning_rate": 3.8074057458289934e-06,
      "loss": 0.7879383,
      "num_input_tokens_seen": 29277450,
      "step": 1386,
      "time_per_iteration": 2.7575531005859375
    },
    {
      "auxiliary_loss_clip": 0.01172206,
      "auxiliary_loss_mlp": 0.01041386,
      "balance_loss_clip": 1.06466973,
      "balance_loss_mlp": 1.03140807,
      "epoch": 0.16677688931641918,
      "flos": 22200767043840.0,
      "grad_norm": 2.46272870824204,
      "language_loss": 0.82689261,
      "learning_rate": 3.807072084125934e-06,
      "loss": 0.84902847,
      "num_input_tokens_seen": 29299300,
      "step": 1387,
      "time_per_iteration": 2.912964344024658
    },
    {
      "auxiliary_loss_clip": 0.0121139,
      "auxiliary_loss_mlp": 0.01042646,
      "balance_loss_clip": 1.06939924,
      "balance_loss_mlp": 1.03189301,
      "epoch": 0.16689713220705826,
      "flos": 16945958776320.0,
      "grad_norm": 2.7817084198708106,
      "language_loss": 0.80565715,
      "learning_rate": 3.806738148291477e-06,
      "loss": 0.82819754,
      "num_input_tokens_seen": 29316125,
      "step": 1388,
      "time_per_iteration": 2.6938555240631104
    },
    {
      "auxiliary_loss_clip": 0.01188994,
      "auxiliary_loss_mlp": 0.01034706,
      "balance_loss_clip": 1.06494904,
      "balance_loss_mlp": 1.02406037,
      "epoch": 0.16701737509769735,
      "flos": 36244423923840.0,
      "grad_norm": 3.4483274740175385,
      "language_loss": 0.71136773,
      "learning_rate": 3.8064039383762793e-06,
      "loss": 0.73360473,
      "num_input_tokens_seen": 29338490,
      "step": 1389,
      "time_per_iteration": 2.9130825996398926
    },
    {
      "auxiliary_loss_clip": 0.01200899,
      "auxiliary_loss_mlp": 0.01038611,
      "balance_loss_clip": 1.06773996,
      "balance_loss_mlp": 1.02679753,
      "epoch": 0.16713761798833643,
      "flos": 23258659426560.0,
      "grad_norm": 3.1735444803251363,
      "language_loss": 0.77252591,
      "learning_rate": 3.8060694544310396e-06,
      "loss": 0.79492092,
      "num_input_tokens_seen": 29357000,
      "step": 1390,
      "time_per_iteration": 3.745972156524658
    },
    {
      "auxiliary_loss_clip": 0.012105,
      "auxiliary_loss_mlp": 0.01038391,
      "balance_loss_clip": 1.06933141,
      "balance_loss_mlp": 1.02751279,
      "epoch": 0.16725786087897554,
      "flos": 25302515207040.0,
      "grad_norm": 2.1764662139867457,
      "language_loss": 0.78423381,
      "learning_rate": 3.8057346965065006e-06,
      "loss": 0.80672276,
      "num_input_tokens_seen": 29378230,
      "step": 1391,
      "time_per_iteration": 2.7679543495178223
    },
    {
      "auxiliary_loss_clip": 0.01183399,
      "auxiliary_loss_mlp": 0.01039846,
      "balance_loss_clip": 1.06752419,
      "balance_loss_mlp": 1.02989805,
      "epoch": 0.16737810376961462,
      "flos": 31831541516160.0,
      "grad_norm": 6.445711947127164,
      "language_loss": 0.84587324,
      "learning_rate": 3.805399664653443e-06,
      "loss": 0.86810577,
      "num_input_tokens_seen": 29400370,
      "step": 1392,
      "time_per_iteration": 2.8518900871276855
    },
    {
      "auxiliary_loss_clip": 0.01213619,
      "auxiliary_loss_mlp": 0.01042881,
      "balance_loss_clip": 1.07076049,
      "balance_loss_mlp": 1.03196192,
      "epoch": 0.1674983466602537,
      "flos": 27961843553280.0,
      "grad_norm": 6.098649075227957,
      "language_loss": 0.74567533,
      "learning_rate": 3.805064358922692e-06,
      "loss": 0.76824033,
      "num_input_tokens_seen": 29418660,
      "step": 1393,
      "time_per_iteration": 3.9386250972747803
    },
    {
      "auxiliary_loss_clip": 0.01229861,
      "auxiliary_loss_mlp": 0.01044747,
      "balance_loss_clip": 1.06942558,
      "balance_loss_mlp": 1.03370845,
      "epoch": 0.16761858955089282,
      "flos": 21762656858880.0,
      "grad_norm": 2.3561599122630743,
      "language_loss": 0.81306326,
      "learning_rate": 3.8047287793651136e-06,
      "loss": 0.83580935,
      "num_input_tokens_seen": 29440105,
      "step": 1394,
      "time_per_iteration": 2.7486627101898193
    },
    {
      "auxiliary_loss_clip": 0.01158951,
      "auxiliary_loss_mlp": 0.01038142,
      "balance_loss_clip": 1.06073904,
      "balance_loss_mlp": 1.02714539,
      "epoch": 0.1677388324415319,
      "flos": 23805507058560.0,
      "grad_norm": 5.02441286085763,
      "language_loss": 0.88768041,
      "learning_rate": 3.8043929260316137e-06,
      "loss": 0.90965128,
      "num_input_tokens_seen": 29458260,
      "step": 1395,
      "time_per_iteration": 3.892900228500366
    },
    {
      "auxiliary_loss_clip": 0.01168813,
      "auxiliary_loss_mlp": 0.0104143,
      "balance_loss_clip": 1.06689978,
      "balance_loss_mlp": 1.03078425,
      "epoch": 0.16785907533217098,
      "flos": 20558859431040.0,
      "grad_norm": 2.0071994459517004,
      "language_loss": 0.8352651,
      "learning_rate": 3.8040567989731417e-06,
      "loss": 0.85736752,
      "num_input_tokens_seen": 29476205,
      "step": 1396,
      "time_per_iteration": 2.774975061416626
    },
    {
      "auxiliary_loss_clip": 0.0121813,
      "auxiliary_loss_mlp": 0.01037203,
      "balance_loss_clip": 1.07227349,
      "balance_loss_mlp": 1.02758908,
      "epoch": 0.16797931822281006,
      "flos": 15669657745920.0,
      "grad_norm": 2.4792549172147247,
      "language_loss": 0.79627955,
      "learning_rate": 3.8037203982406876e-06,
      "loss": 0.81883287,
      "num_input_tokens_seen": 29494370,
      "step": 1397,
      "time_per_iteration": 2.84226393699646
    },
    {
      "auxiliary_loss_clip": 0.01179858,
      "auxiliary_loss_mlp": 0.01045297,
      "balance_loss_clip": 1.06691039,
      "balance_loss_mlp": 1.03429973,
      "epoch": 0.16809956111344918,
      "flos": 16541101607040.0,
      "grad_norm": 2.3453538366172473,
      "language_loss": 0.72929847,
      "learning_rate": 3.8033837238852835e-06,
      "loss": 0.75155002,
      "num_input_tokens_seen": 29511070,
      "step": 1398,
      "time_per_iteration": 2.837017297744751
    },
    {
      "auxiliary_loss_clip": 0.01186184,
      "auxiliary_loss_mlp": 0.01045998,
      "balance_loss_clip": 1.06667995,
      "balance_loss_mlp": 1.03569818,
      "epoch": 0.16821980400408826,
      "flos": 23258084808960.0,
      "grad_norm": 3.6682111572111316,
      "language_loss": 0.69711393,
      "learning_rate": 3.8030467759580017e-06,
      "loss": 0.71943569,
      "num_input_tokens_seen": 29531990,
      "step": 1399,
      "time_per_iteration": 3.7477190494537354
    },
    {
      "auxiliary_loss_clip": 0.01225797,
      "auxiliary_loss_mlp": 0.00764866,
      "balance_loss_clip": 1.07265401,
      "balance_loss_mlp": 1.00065684,
      "epoch": 0.16834004689472734,
      "flos": 20774754126720.0,
      "grad_norm": 5.038360856136953,
      "language_loss": 0.87290996,
      "learning_rate": 3.802709554509958e-06,
      "loss": 0.8928166,
      "num_input_tokens_seen": 29549790,
      "step": 1400,
      "time_per_iteration": 2.7569782733917236
    },
    {
      "auxiliary_loss_clip": 0.01232425,
      "auxiliary_loss_mlp": 0.01030318,
      "balance_loss_clip": 1.06935573,
      "balance_loss_mlp": 1.02007842,
      "epoch": 0.16846028978536645,
      "flos": 26687302289280.0,
      "grad_norm": 3.3332805670301684,
      "language_loss": 0.79361618,
      "learning_rate": 3.8023720595923083e-06,
      "loss": 0.81624359,
      "num_input_tokens_seen": 29569045,
      "step": 1401,
      "time_per_iteration": 2.757394552230835
    },
    {
      "auxiliary_loss_clip": 0.0118356,
      "auxiliary_loss_mlp": 0.0103835,
      "balance_loss_clip": 1.06546712,
      "balance_loss_mlp": 1.02800226,
      "epoch": 0.16858053267600553,
      "flos": 18843298980480.0,
      "grad_norm": 2.6976087785179437,
      "language_loss": 0.87844598,
      "learning_rate": 3.80203429125625e-06,
      "loss": 0.90066504,
      "num_input_tokens_seen": 29587220,
      "step": 1402,
      "time_per_iteration": 2.75529408454895
    },
    {
      "auxiliary_loss_clip": 0.01204927,
      "auxiliary_loss_mlp": 0.01040459,
      "balance_loss_clip": 1.07070041,
      "balance_loss_mlp": 1.03058267,
      "epoch": 0.16870077556664462,
      "flos": 27744548227200.0,
      "grad_norm": 3.0349757102830224,
      "language_loss": 0.70050991,
      "learning_rate": 3.8016962495530225e-06,
      "loss": 0.72296381,
      "num_input_tokens_seen": 29606410,
      "step": 1403,
      "time_per_iteration": 2.8485209941864014
    },
    {
      "auxiliary_loss_clip": 0.01209769,
      "auxiliary_loss_mlp": 0.01041759,
      "balance_loss_clip": 1.06859422,
      "balance_loss_mlp": 1.03145325,
      "epoch": 0.1688210184572837,
      "flos": 13730768484480.0,
      "grad_norm": 8.196736205766603,
      "language_loss": 0.76754606,
      "learning_rate": 3.8013579345339063e-06,
      "loss": 0.79006135,
      "num_input_tokens_seen": 29621275,
      "step": 1404,
      "time_per_iteration": 2.684225082397461
    },
    {
      "auxiliary_loss_clip": 0.01141348,
      "auxiliary_loss_mlp": 0.01046869,
      "balance_loss_clip": 1.05719697,
      "balance_loss_mlp": 1.03591919,
      "epoch": 0.1689412613479228,
      "flos": 26468785900800.0,
      "grad_norm": 3.7903777062691204,
      "language_loss": 0.6950047,
      "learning_rate": 3.801019346250224e-06,
      "loss": 0.71688682,
      "num_input_tokens_seen": 29641420,
      "step": 1405,
      "time_per_iteration": 2.876307964324951
    },
    {
      "auxiliary_loss_clip": 0.01206432,
      "auxiliary_loss_mlp": 0.01031802,
      "balance_loss_clip": 1.06955171,
      "balance_loss_mlp": 1.02244425,
      "epoch": 0.1690615042385619,
      "flos": 21138852337920.0,
      "grad_norm": 4.577575072586073,
      "language_loss": 0.83769482,
      "learning_rate": 3.8006804847533395e-06,
      "loss": 0.86007714,
      "num_input_tokens_seen": 29660935,
      "step": 1406,
      "time_per_iteration": 2.6605517864227295
    },
    {
      "auxiliary_loss_clip": 0.01171156,
      "auxiliary_loss_mlp": 0.00764173,
      "balance_loss_clip": 1.05953681,
      "balance_loss_mlp": 1.00062156,
      "epoch": 0.16918174712920098,
      "flos": 20849340718080.0,
      "grad_norm": 2.889300259915884,
      "language_loss": 0.85630476,
      "learning_rate": 3.8003413500946556e-06,
      "loss": 0.87565804,
      "num_input_tokens_seen": 29681045,
      "step": 1407,
      "time_per_iteration": 2.7373006343841553
    },
    {
      "auxiliary_loss_clip": 0.0119031,
      "auxiliary_loss_mlp": 0.00763996,
      "balance_loss_clip": 1.06067693,
      "balance_loss_mlp": 1.00047541,
      "epoch": 0.1693019900198401,
      "flos": 16983270028800.0,
      "grad_norm": 3.4408470598793817,
      "language_loss": 0.83414209,
      "learning_rate": 3.8000019423256216e-06,
      "loss": 0.85368514,
      "num_input_tokens_seen": 29698810,
      "step": 1408,
      "time_per_iteration": 2.620224714279175
    },
    {
      "auxiliary_loss_clip": 0.01135107,
      "auxiliary_loss_mlp": 0.01039115,
      "balance_loss_clip": 1.05606151,
      "balance_loss_mlp": 1.0282433,
      "epoch": 0.16942223291047917,
      "flos": 26796901662720.0,
      "grad_norm": 7.580537907506638,
      "language_loss": 0.88130701,
      "learning_rate": 3.7996622614977234e-06,
      "loss": 0.90304923,
      "num_input_tokens_seen": 29720000,
      "step": 1409,
      "time_per_iteration": 2.7744174003601074
    },
    {
      "auxiliary_loss_clip": 0.01158196,
      "auxiliary_loss_mlp": 0.01036258,
      "balance_loss_clip": 1.06132078,
      "balance_loss_mlp": 1.02606595,
      "epoch": 0.16954247580111825,
      "flos": 18583700411520.0,
      "grad_norm": 3.5385134755444376,
      "language_loss": 0.7909255,
      "learning_rate": 3.799322307662492e-06,
      "loss": 0.81287003,
      "num_input_tokens_seen": 29737820,
      "step": 1410,
      "time_per_iteration": 2.751068115234375
    },
    {
      "auxiliary_loss_clip": 0.01207259,
      "auxiliary_loss_mlp": 0.01040248,
      "balance_loss_clip": 1.07045436,
      "balance_loss_mlp": 1.02988851,
      "epoch": 0.16966271869175734,
      "flos": 13983651210240.0,
      "grad_norm": 3.2930848777790547,
      "language_loss": 0.83798748,
      "learning_rate": 3.798982080871496e-06,
      "loss": 0.86046255,
      "num_input_tokens_seen": 29752960,
      "step": 1411,
      "time_per_iteration": 2.6859681606292725
    },
    {
      "auxiliary_loss_clip": 0.01212127,
      "auxiliary_loss_mlp": 0.01039724,
      "balance_loss_clip": 1.0707233,
      "balance_loss_mlp": 1.02841055,
      "epoch": 0.16978296158239645,
      "flos": 37487328284160.0,
      "grad_norm": 2.4775186960914097,
      "language_loss": 0.67938179,
      "learning_rate": 3.798641581176349e-06,
      "loss": 0.70190036,
      "num_input_tokens_seen": 29775240,
      "step": 1412,
      "time_per_iteration": 2.7892863750457764
    },
    {
      "auxiliary_loss_clip": 0.01183697,
      "auxiliary_loss_mlp": 0.00764483,
      "balance_loss_clip": 1.06404352,
      "balance_loss_mlp": 1.00059235,
      "epoch": 0.16990320447303553,
      "flos": 28328958506880.0,
      "grad_norm": 2.058317676738656,
      "language_loss": 0.74700397,
      "learning_rate": 3.7983008086287044e-06,
      "loss": 0.76648581,
      "num_input_tokens_seen": 29796560,
      "step": 1413,
      "time_per_iteration": 2.8707828521728516
    },
    {
      "auxiliary_loss_clip": 0.01208977,
      "auxiliary_loss_mlp": 0.01037193,
      "balance_loss_clip": 1.07220602,
      "balance_loss_mlp": 1.02580905,
      "epoch": 0.1700234473636746,
      "flos": 20188189031040.0,
      "grad_norm": 4.922008874562053,
      "language_loss": 0.79224181,
      "learning_rate": 3.797959763280257e-06,
      "loss": 0.81470346,
      "num_input_tokens_seen": 29815245,
      "step": 1414,
      "time_per_iteration": 2.772036552429199
    },
    {
      "auxiliary_loss_clip": 0.01182179,
      "auxiliary_loss_mlp": 0.01037159,
      "balance_loss_clip": 1.0651294,
      "balance_loss_mlp": 1.02618539,
      "epoch": 0.17014369025431372,
      "flos": 24858658846080.0,
      "grad_norm": 1.9950908838112718,
      "language_loss": 0.78905231,
      "learning_rate": 3.797618445182743e-06,
      "loss": 0.81124568,
      "num_input_tokens_seen": 29836640,
      "step": 1415,
      "time_per_iteration": 2.759680986404419
    },
    {
      "auxiliary_loss_clip": 0.01223602,
      "auxiliary_loss_mlp": 0.01040154,
      "balance_loss_clip": 1.07425237,
      "balance_loss_mlp": 1.02970529,
      "epoch": 0.1702639331449528,
      "flos": 16467233287680.0,
      "grad_norm": 5.39555679092393,
      "language_loss": 0.85149944,
      "learning_rate": 3.79727685438794e-06,
      "loss": 0.87413704,
      "num_input_tokens_seen": 29850830,
      "step": 1416,
      "time_per_iteration": 4.031467437744141
    },
    {
      "auxiliary_loss_clip": 0.01087673,
      "auxiliary_loss_mlp": 0.01014184,
      "balance_loss_clip": 1.04050684,
      "balance_loss_mlp": 1.01160896,
      "epoch": 0.1703841760355919,
      "flos": 52508870979840.0,
      "grad_norm": 0.8752154260683245,
      "language_loss": 0.61604369,
      "learning_rate": 3.796934990947667e-06,
      "loss": 0.63706219,
      "num_input_tokens_seen": 29912515,
      "step": 1417,
      "time_per_iteration": 3.4027512073516846
    },
    {
      "auxiliary_loss_clip": 0.0113742,
      "auxiliary_loss_mlp": 0.0075664,
      "balance_loss_clip": 1.04802895,
      "balance_loss_mlp": 1.00108075,
      "epoch": 0.170504418926231,
      "flos": 49370637576960.0,
      "grad_norm": 0.8752084057240928,
      "language_loss": 0.62484872,
      "learning_rate": 3.7965928549137854e-06,
      "loss": 0.64378929,
      "num_input_tokens_seen": 29969330,
      "step": 1418,
      "time_per_iteration": 4.1655354499816895
    },
    {
      "auxiliary_loss_clip": 0.01204183,
      "auxiliary_loss_mlp": 0.01034112,
      "balance_loss_clip": 1.06576443,
      "balance_loss_mlp": 1.02250719,
      "epoch": 0.17062466181687008,
      "flos": 25849219184640.0,
      "grad_norm": 2.64325636702312,
      "language_loss": 0.77895975,
      "learning_rate": 3.7962504463381953e-06,
      "loss": 0.80134273,
      "num_input_tokens_seen": 29990820,
      "step": 1419,
      "time_per_iteration": 2.873469591140747
    },
    {
      "auxiliary_loss_clip": 0.01244627,
      "auxiliary_loss_mlp": 0.01039546,
      "balance_loss_clip": 1.07874966,
      "balance_loss_mlp": 1.02890062,
      "epoch": 0.17074490470750917,
      "flos": 20960412549120.0,
      "grad_norm": 2.708585944572838,
      "language_loss": 0.78855252,
      "learning_rate": 3.7959077652728412e-06,
      "loss": 0.81139421,
      "num_input_tokens_seen": 30009275,
      "step": 1420,
      "time_per_iteration": 2.6869442462921143
    },
    {
      "auxiliary_loss_clip": 0.01177491,
      "auxiliary_loss_mlp": 0.0103669,
      "balance_loss_clip": 1.06076813,
      "balance_loss_mlp": 1.02525198,
      "epoch": 0.17086514759814825,
      "flos": 20959766104320.0,
      "grad_norm": 4.672191397173583,
      "language_loss": 0.77096224,
      "learning_rate": 3.795564811769707e-06,
      "loss": 0.79310405,
      "num_input_tokens_seen": 30027630,
      "step": 1421,
      "time_per_iteration": 3.6727352142333984
    },
    {
      "auxiliary_loss_clip": 0.01229657,
      "auxiliary_loss_mlp": 0.01049111,
      "balance_loss_clip": 1.07919455,
      "balance_loss_mlp": 1.03758335,
      "epoch": 0.17098539048878736,
      "flos": 28474073452800.0,
      "grad_norm": 2.7108599676314022,
      "language_loss": 0.78235686,
      "learning_rate": 3.795221585880818e-06,
      "loss": 0.80514455,
      "num_input_tokens_seen": 30048310,
      "step": 1422,
      "time_per_iteration": 2.8074679374694824
    },
    {
      "auxiliary_loss_clip": 0.0120188,
      "auxiliary_loss_mlp": 0.01044258,
      "balance_loss_clip": 1.07047486,
      "balance_loss_mlp": 1.03327894,
      "epoch": 0.17110563337942644,
      "flos": 16290014561280.0,
      "grad_norm": 2.7079599127262077,
      "language_loss": 0.90972483,
      "learning_rate": 3.794878087658242e-06,
      "loss": 0.93218613,
      "num_input_tokens_seen": 30066080,
      "step": 1423,
      "time_per_iteration": 2.7628579139709473
    },
    {
      "auxiliary_loss_clip": 0.01195971,
      "auxiliary_loss_mlp": 0.01040425,
      "balance_loss_clip": 1.06708682,
      "balance_loss_mlp": 1.02995205,
      "epoch": 0.17122587627006552,
      "flos": 29674207693440.0,
      "grad_norm": 3.073756781212297,
      "language_loss": 0.7834602,
      "learning_rate": 3.7945343171540873e-06,
      "loss": 0.8058241,
      "num_input_tokens_seen": 30086955,
      "step": 1424,
      "time_per_iteration": 3.7522692680358887
    },
    {
      "auxiliary_loss_clip": 0.01161263,
      "auxiliary_loss_mlp": 0.01034459,
      "balance_loss_clip": 1.06013346,
      "balance_loss_mlp": 1.0234201,
      "epoch": 0.17134611916070464,
      "flos": 25338389915520.0,
      "grad_norm": 2.4235497449467225,
      "language_loss": 0.79019403,
      "learning_rate": 3.7941902744205033e-06,
      "loss": 0.81215125,
      "num_input_tokens_seen": 30107990,
      "step": 1425,
      "time_per_iteration": 2.864365816116333
    },
    {
      "auxiliary_loss_clip": 0.01246895,
      "auxiliary_loss_mlp": 0.01045162,
      "balance_loss_clip": 1.07733369,
      "balance_loss_mlp": 1.03359234,
      "epoch": 0.17146636205134372,
      "flos": 13953845900160.0,
      "grad_norm": 1.9196800672995453,
      "language_loss": 0.83191502,
      "learning_rate": 3.7938459595096817e-06,
      "loss": 0.85483551,
      "num_input_tokens_seen": 30126535,
      "step": 1426,
      "time_per_iteration": 2.5920050144195557
    },
    {
      "auxiliary_loss_clip": 0.01233782,
      "auxiliary_loss_mlp": 0.01036385,
      "balance_loss_clip": 1.07704926,
      "balance_loss_mlp": 1.02479792,
      "epoch": 0.1715866049419828,
      "flos": 23915214172800.0,
      "grad_norm": 2.1898407700384594,
      "language_loss": 0.86289984,
      "learning_rate": 3.7935013724738545e-06,
      "loss": 0.88560152,
      "num_input_tokens_seen": 30147035,
      "step": 1427,
      "time_per_iteration": 2.709157943725586
    },
    {
      "auxiliary_loss_clip": 0.01157317,
      "auxiliary_loss_mlp": 0.00764265,
      "balance_loss_clip": 1.06102109,
      "balance_loss_mlp": 1.0005132,
      "epoch": 0.17170684783262188,
      "flos": 22709369669760.0,
      "grad_norm": 2.051042487498941,
      "language_loss": 0.77789366,
      "learning_rate": 3.7931565133652945e-06,
      "loss": 0.79710948,
      "num_input_tokens_seen": 30167110,
      "step": 1428,
      "time_per_iteration": 2.7990479469299316
    },
    {
      "auxiliary_loss_clip": 0.0116524,
      "auxiliary_loss_mlp": 0.01042643,
      "balance_loss_clip": 1.0613842,
      "balance_loss_mlp": 1.03210449,
      "epoch": 0.171827090723261,
      "flos": 26613290315520.0,
      "grad_norm": 4.225976984265807,
      "language_loss": 0.67981195,
      "learning_rate": 3.792811382236317e-06,
      "loss": 0.70189083,
      "num_input_tokens_seen": 30185620,
      "step": 1429,
      "time_per_iteration": 2.9111709594726562
    },
    {
      "auxiliary_loss_clip": 0.01170845,
      "auxiliary_loss_mlp": 0.01047665,
      "balance_loss_clip": 1.06479383,
      "balance_loss_mlp": 1.0361371,
      "epoch": 0.17194733361390008,
      "flos": 28148507556480.0,
      "grad_norm": 2.1098299228173456,
      "language_loss": 0.78139031,
      "learning_rate": 3.792465979139279e-06,
      "loss": 0.8035754,
      "num_input_tokens_seen": 30208225,
      "step": 1430,
      "time_per_iteration": 2.861847162246704
    },
    {
      "auxiliary_loss_clip": 0.01109338,
      "auxiliary_loss_mlp": 0.01011195,
      "balance_loss_clip": 1.03784657,
      "balance_loss_mlp": 1.00848949,
      "epoch": 0.17206757650453916,
      "flos": 65530689753600.0,
      "grad_norm": 0.9241560851474658,
      "language_loss": 0.65627527,
      "learning_rate": 3.792120304126576e-06,
      "loss": 0.67748058,
      "num_input_tokens_seen": 30271600,
      "step": 1431,
      "time_per_iteration": 3.3688457012176514
    },
    {
      "auxiliary_loss_clip": 0.01195762,
      "auxiliary_loss_mlp": 0.010429,
      "balance_loss_clip": 1.07035804,
      "balance_loss_mlp": 1.03149748,
      "epoch": 0.17218781939517827,
      "flos": 22273486128000.0,
      "grad_norm": 1.7823488229768234,
      "language_loss": 0.83478701,
      "learning_rate": 3.791774357250649e-06,
      "loss": 0.85717356,
      "num_input_tokens_seen": 30290430,
      "step": 1432,
      "time_per_iteration": 2.731855630874634
    },
    {
      "auxiliary_loss_clip": 0.01171213,
      "auxiliary_loss_mlp": 0.01047589,
      "balance_loss_clip": 1.06195831,
      "balance_loss_mlp": 1.03657961,
      "epoch": 0.17230806228581735,
      "flos": 14137313592960.0,
      "grad_norm": 2.6386766867824814,
      "language_loss": 0.78814489,
      "learning_rate": 3.7914281385639757e-06,
      "loss": 0.81033289,
      "num_input_tokens_seen": 30308305,
      "step": 1433,
      "time_per_iteration": 2.768367052078247
    },
    {
      "auxiliary_loss_clip": 0.01239081,
      "auxiliary_loss_mlp": 0.01048033,
      "balance_loss_clip": 1.07409573,
      "balance_loss_mlp": 1.03738761,
      "epoch": 0.17242830517645644,
      "flos": 20704836303360.0,
      "grad_norm": 2.153180261101895,
      "language_loss": 0.79412097,
      "learning_rate": 3.7910816481190784e-06,
      "loss": 0.81699216,
      "num_input_tokens_seen": 30328120,
      "step": 1434,
      "time_per_iteration": 2.6597561836242676
    },
    {
      "auxiliary_loss_clip": 0.0112818,
      "auxiliary_loss_mlp": 0.01044497,
      "balance_loss_clip": 1.05704212,
      "balance_loss_mlp": 1.03339839,
      "epoch": 0.17254854806709552,
      "flos": 30774582887040.0,
      "grad_norm": 1.9718400820862954,
      "language_loss": 0.74669212,
      "learning_rate": 3.7907348859685193e-06,
      "loss": 0.76841891,
      "num_input_tokens_seen": 30349825,
      "step": 1435,
      "time_per_iteration": 2.953169107437134
    },
    {
      "auxiliary_loss_clip": 0.01238262,
      "auxiliary_loss_mlp": 0.0104549,
      "balance_loss_clip": 1.07568765,
      "balance_loss_mlp": 1.03429055,
      "epoch": 0.17266879095773463,
      "flos": 26614726859520.0,
      "grad_norm": 2.6533907086295194,
      "language_loss": 0.80078888,
      "learning_rate": 3.790387852164902e-06,
      "loss": 0.8236264,
      "num_input_tokens_seen": 30370555,
      "step": 1436,
      "time_per_iteration": 3.1151068210601807
    },
    {
      "auxiliary_loss_clip": 0.01168756,
      "auxiliary_loss_mlp": 0.01035901,
      "balance_loss_clip": 1.06202579,
      "balance_loss_mlp": 1.0255475,
      "epoch": 0.1727890338483737,
      "flos": 20266295155200.0,
      "grad_norm": 2.088388238306747,
      "language_loss": 0.76633084,
      "learning_rate": 3.7900405467608707e-06,
      "loss": 0.7883774,
      "num_input_tokens_seen": 30390100,
      "step": 1437,
      "time_per_iteration": 2.738334894180298
    },
    {
      "auxiliary_loss_clip": 0.01209323,
      "auxiliary_loss_mlp": 0.01050345,
      "balance_loss_clip": 1.06890297,
      "balance_loss_mlp": 1.03924704,
      "epoch": 0.1729092767390128,
      "flos": 18179812909440.0,
      "grad_norm": 4.06237003032756,
      "language_loss": 0.79409623,
      "learning_rate": 3.7896929698091114e-06,
      "loss": 0.81669289,
      "num_input_tokens_seen": 30402915,
      "step": 1438,
      "time_per_iteration": 2.6150009632110596
    },
    {
      "auxiliary_loss_clip": 0.01196626,
      "auxiliary_loss_mlp": 0.01048778,
      "balance_loss_clip": 1.07611585,
      "balance_loss_mlp": 1.0382576,
      "epoch": 0.1730295196296519,
      "flos": 26759518583040.0,
      "grad_norm": 3.312686505930049,
      "language_loss": 0.67958444,
      "learning_rate": 3.7893451213623518e-06,
      "loss": 0.70203853,
      "num_input_tokens_seen": 30420145,
      "step": 1439,
      "time_per_iteration": 2.8172547817230225
    },
    {
      "auxiliary_loss_clip": 0.01221354,
      "auxiliary_loss_mlp": 0.01035113,
      "balance_loss_clip": 1.07249188,
      "balance_loss_mlp": 1.02365744,
      "epoch": 0.173149762520291,
      "flos": 23842531002240.0,
      "grad_norm": 3.060614579690584,
      "language_loss": 0.81982332,
      "learning_rate": 3.7889970014733606e-06,
      "loss": 0.84238803,
      "num_input_tokens_seen": 30439250,
      "step": 1440,
      "time_per_iteration": 2.7433950901031494
    },
    {
      "auxiliary_loss_clip": 0.01178915,
      "auxiliary_loss_mlp": 0.01038623,
      "balance_loss_clip": 1.06417489,
      "balance_loss_mlp": 1.02788186,
      "epoch": 0.17327000541093007,
      "flos": 23368186972800.0,
      "grad_norm": 1.974385718132932,
      "language_loss": 0.77914274,
      "learning_rate": 3.7886486101949463e-06,
      "loss": 0.80131817,
      "num_input_tokens_seen": 30460430,
      "step": 1441,
      "time_per_iteration": 2.787954330444336
    },
    {
      "auxiliary_loss_clip": 0.01195835,
      "auxiliary_loss_mlp": 0.01043564,
      "balance_loss_clip": 1.07331455,
      "balance_loss_mlp": 1.03331816,
      "epoch": 0.17339024830156918,
      "flos": 18221290139520.0,
      "grad_norm": 2.862334899384068,
      "language_loss": 0.88238454,
      "learning_rate": 3.7882999475799594e-06,
      "loss": 0.90477854,
      "num_input_tokens_seen": 30478465,
      "step": 1442,
      "time_per_iteration": 3.888530731201172
    },
    {
      "auxiliary_loss_clip": 0.01216177,
      "auxiliary_loss_mlp": 0.01053739,
      "balance_loss_clip": 1.07244301,
      "balance_loss_mlp": 1.0423069,
      "epoch": 0.17351049119220827,
      "flos": 23332024955520.0,
      "grad_norm": 7.93894275245745,
      "language_loss": 0.81545383,
      "learning_rate": 3.787951013681293e-06,
      "loss": 0.838153,
      "num_input_tokens_seen": 30496510,
      "step": 1443,
      "time_per_iteration": 2.6795248985290527
    },
    {
      "auxiliary_loss_clip": 0.01149217,
      "auxiliary_loss_mlp": 0.01043112,
      "balance_loss_clip": 1.06181884,
      "balance_loss_mlp": 1.03172708,
      "epoch": 0.17363073408284735,
      "flos": 23803495896960.0,
      "grad_norm": 2.553619552562121,
      "language_loss": 0.7804991,
      "learning_rate": 3.787601808551879e-06,
      "loss": 0.8024224,
      "num_input_tokens_seen": 30516325,
      "step": 1444,
      "time_per_iteration": 3.923795223236084
    },
    {
      "auxiliary_loss_clip": 0.01210394,
      "auxiliary_loss_mlp": 0.01043497,
      "balance_loss_clip": 1.0693326,
      "balance_loss_mlp": 1.03284574,
      "epoch": 0.17375097697348643,
      "flos": 18515290959360.0,
      "grad_norm": 2.872087895293403,
      "language_loss": 0.84149951,
      "learning_rate": 3.7872523322446926e-06,
      "loss": 0.86403847,
      "num_input_tokens_seen": 30535210,
      "step": 1445,
      "time_per_iteration": 3.2552239894866943
    },
    {
      "auxiliary_loss_clip": 0.01154127,
      "auxiliary_loss_mlp": 0.01028854,
      "balance_loss_clip": 1.0588038,
      "balance_loss_mlp": 1.01709962,
      "epoch": 0.17387121986412554,
      "flos": 38877897456000.0,
      "grad_norm": 1.9587365118745765,
      "language_loss": 0.60326481,
      "learning_rate": 3.7869025848127478e-06,
      "loss": 0.62509459,
      "num_input_tokens_seen": 30559405,
      "step": 1446,
      "time_per_iteration": 3.926239252090454
    },
    {
      "auxiliary_loss_clip": 0.01211183,
      "auxiliary_loss_mlp": 0.01043854,
      "balance_loss_clip": 1.0656414,
      "balance_loss_mlp": 1.03338122,
      "epoch": 0.17399146275476463,
      "flos": 20375714960640.0,
      "grad_norm": 3.10199458944682,
      "language_loss": 0.80524135,
      "learning_rate": 3.786552566309102e-06,
      "loss": 0.82779169,
      "num_input_tokens_seen": 30577615,
      "step": 1447,
      "time_per_iteration": 2.6956863403320312
    },
    {
      "auxiliary_loss_clip": 0.01182084,
      "auxiliary_loss_mlp": 0.01046991,
      "balance_loss_clip": 1.06394339,
      "balance_loss_mlp": 1.03620303,
      "epoch": 0.1741117056454037,
      "flos": 19164339763200.0,
      "grad_norm": 3.0683255070697015,
      "language_loss": 0.86584687,
      "learning_rate": 3.7862022767868517e-06,
      "loss": 0.88813764,
      "num_input_tokens_seen": 30595205,
      "step": 1448,
      "time_per_iteration": 2.7574825286865234
    },
    {
      "auxiliary_loss_clip": 0.01149414,
      "auxiliary_loss_mlp": 0.01040534,
      "balance_loss_clip": 1.06403089,
      "balance_loss_mlp": 1.03062761,
      "epoch": 0.17423194853604282,
      "flos": 25374300537600.0,
      "grad_norm": 2.0788129298730187,
      "language_loss": 0.84399104,
      "learning_rate": 3.7858517162991367e-06,
      "loss": 0.8658905,
      "num_input_tokens_seen": 30615280,
      "step": 1449,
      "time_per_iteration": 2.8519210815429688
    },
    {
      "auxiliary_loss_clip": 0.01176174,
      "auxiliary_loss_mlp": 0.0104401,
      "balance_loss_clip": 1.06416953,
      "balance_loss_mlp": 1.03397882,
      "epoch": 0.1743521914266819,
      "flos": 25191874339200.0,
      "grad_norm": 3.6261900315175137,
      "language_loss": 0.60691655,
      "learning_rate": 3.7855008848991363e-06,
      "loss": 0.62911844,
      "num_input_tokens_seen": 30633485,
      "step": 1450,
      "time_per_iteration": 3.8545963764190674
    },
    {
      "auxiliary_loss_clip": 0.01189454,
      "auxiliary_loss_mlp": 0.01046726,
      "balance_loss_clip": 1.06303966,
      "balance_loss_mlp": 1.03648567,
      "epoch": 0.17447243431732098,
      "flos": 25666577504640.0,
      "grad_norm": 1.870516086207767,
      "language_loss": 0.77461231,
      "learning_rate": 3.7851497826400714e-06,
      "loss": 0.79697412,
      "num_input_tokens_seen": 30653625,
      "step": 1451,
      "time_per_iteration": 2.81807017326355
    },
    {
      "auxiliary_loss_clip": 0.01151294,
      "auxiliary_loss_mlp": 0.0104677,
      "balance_loss_clip": 1.06413758,
      "balance_loss_mlp": 1.03463399,
      "epoch": 0.17459267720796007,
      "flos": 36281950657920.0,
      "grad_norm": 2.365486537746293,
      "language_loss": 0.75986552,
      "learning_rate": 3.7847984095752034e-06,
      "loss": 0.78184617,
      "num_input_tokens_seen": 30677080,
      "step": 1452,
      "time_per_iteration": 3.062747001647949
    },
    {
      "auxiliary_loss_clip": 0.01223676,
      "auxiliary_loss_mlp": 0.01048238,
      "balance_loss_clip": 1.07227838,
      "balance_loss_mlp": 1.03798008,
      "epoch": 0.17471292009859918,
      "flos": 20011113959040.0,
      "grad_norm": 2.022793692754539,
      "language_loss": 0.80582052,
      "learning_rate": 3.784446765757836e-06,
      "loss": 0.82853961,
      "num_input_tokens_seen": 30695725,
      "step": 1453,
      "time_per_iteration": 3.108405351638794
    },
    {
      "auxiliary_loss_clip": 0.01195425,
      "auxiliary_loss_mlp": 0.01045783,
      "balance_loss_clip": 1.06506753,
      "balance_loss_mlp": 1.03531075,
      "epoch": 0.17483316298923826,
      "flos": 27819242559360.0,
      "grad_norm": 3.0787497978705343,
      "language_loss": 0.77512056,
      "learning_rate": 3.7840948512413133e-06,
      "loss": 0.79753262,
      "num_input_tokens_seen": 30713310,
      "step": 1454,
      "time_per_iteration": 2.7865989208221436
    },
    {
      "auxiliary_loss_clip": 0.0117014,
      "auxiliary_loss_mlp": 0.01041919,
      "balance_loss_clip": 1.06331122,
      "balance_loss_mlp": 1.03135681,
      "epoch": 0.17495340587987734,
      "flos": 44017934791680.0,
      "grad_norm": 2.3313409584450335,
      "language_loss": 0.78622097,
      "learning_rate": 3.7837426660790196e-06,
      "loss": 0.8083415,
      "num_input_tokens_seen": 30734725,
      "step": 1455,
      "time_per_iteration": 2.9888014793395996
    },
    {
      "auxiliary_loss_clip": 0.01194366,
      "auxiliary_loss_mlp": 0.01036002,
      "balance_loss_clip": 1.06558776,
      "balance_loss_mlp": 1.02554131,
      "epoch": 0.17507364877051645,
      "flos": 20885825957760.0,
      "grad_norm": 2.3178006423619952,
      "language_loss": 0.81801319,
      "learning_rate": 3.783390210324382e-06,
      "loss": 0.84031683,
      "num_input_tokens_seen": 30754450,
      "step": 1456,
      "time_per_iteration": 2.7797844409942627
    },
    {
      "auxiliary_loss_clip": 0.01180203,
      "auxiliary_loss_mlp": 0.00763907,
      "balance_loss_clip": 1.06219542,
      "balance_loss_mlp": 1.00060236,
      "epoch": 0.17519389166115554,
      "flos": 24717602136960.0,
      "grad_norm": 2.2660473098081186,
      "language_loss": 0.72325099,
      "learning_rate": 3.7830374840308676e-06,
      "loss": 0.74269205,
      "num_input_tokens_seen": 30774605,
      "step": 1457,
      "time_per_iteration": 2.8582448959350586
    },
    {
      "auxiliary_loss_clip": 0.01224597,
      "auxiliary_loss_mlp": 0.01041688,
      "balance_loss_clip": 1.07396972,
      "balance_loss_mlp": 1.03103662,
      "epoch": 0.17531413455179462,
      "flos": 23798144770560.0,
      "grad_norm": 2.6428284188942937,
      "language_loss": 0.82715583,
      "learning_rate": 3.7826844872519842e-06,
      "loss": 0.84981871,
      "num_input_tokens_seen": 30792460,
      "step": 1458,
      "time_per_iteration": 2.7658605575561523
    },
    {
      "auxiliary_loss_clip": 0.01196812,
      "auxiliary_loss_mlp": 0.01036485,
      "balance_loss_clip": 1.0710175,
      "balance_loss_mlp": 1.025792,
      "epoch": 0.1754343774424337,
      "flos": 24572379450240.0,
      "grad_norm": 2.1596364748689845,
      "language_loss": 0.72695076,
      "learning_rate": 3.782331220041282e-06,
      "loss": 0.74928373,
      "num_input_tokens_seen": 30812525,
      "step": 1459,
      "time_per_iteration": 2.80096697807312
    },
    {
      "auxiliary_loss_clip": 0.01200104,
      "auxiliary_loss_mlp": 0.01037779,
      "balance_loss_clip": 1.06462502,
      "balance_loss_mlp": 1.02694285,
      "epoch": 0.17555462033307281,
      "flos": 18114599767680.0,
      "grad_norm": 2.5949579823644116,
      "language_loss": 0.82486928,
      "learning_rate": 3.7819776824523504e-06,
      "loss": 0.84724808,
      "num_input_tokens_seen": 30830390,
      "step": 1460,
      "time_per_iteration": 2.69093656539917
    },
    {
      "auxiliary_loss_clip": 0.01223153,
      "auxiliary_loss_mlp": 0.01042285,
      "balance_loss_clip": 1.07279325,
      "balance_loss_mlp": 1.03219426,
      "epoch": 0.1756748632237119,
      "flos": 28366018364160.0,
      "grad_norm": 2.456712475516263,
      "language_loss": 0.83765042,
      "learning_rate": 3.7816238745388213e-06,
      "loss": 0.86030471,
      "num_input_tokens_seen": 30849935,
      "step": 1461,
      "time_per_iteration": 2.8269131183624268
    },
    {
      "auxiliary_loss_clip": 0.01204536,
      "auxiliary_loss_mlp": 0.01039615,
      "balance_loss_clip": 1.06407142,
      "balance_loss_mlp": 1.02994657,
      "epoch": 0.17579510611435098,
      "flos": 25732939881600.0,
      "grad_norm": 2.2634558680474144,
      "language_loss": 0.86909676,
      "learning_rate": 3.781269796354367e-06,
      "loss": 0.89153826,
      "num_input_tokens_seen": 30869555,
      "step": 1462,
      "time_per_iteration": 2.754300832748413
    },
    {
      "auxiliary_loss_clip": 0.01191239,
      "auxiliary_loss_mlp": 0.01044025,
      "balance_loss_clip": 1.07123303,
      "balance_loss_mlp": 1.03345108,
      "epoch": 0.1759153490049901,
      "flos": 18588081870720.0,
      "grad_norm": 3.298293461762424,
      "language_loss": 0.86056054,
      "learning_rate": 3.7809154479527006e-06,
      "loss": 0.88291311,
      "num_input_tokens_seen": 30888760,
      "step": 1463,
      "time_per_iteration": 2.69606876373291
    },
    {
      "auxiliary_loss_clip": 0.01192766,
      "auxiliary_loss_mlp": 0.01051737,
      "balance_loss_clip": 1.06645286,
      "balance_loss_mlp": 1.0413301,
      "epoch": 0.17603559189562917,
      "flos": 18619323724800.0,
      "grad_norm": 6.616122299324271,
      "language_loss": 0.83916414,
      "learning_rate": 3.780560829387577e-06,
      "loss": 0.86160922,
      "num_input_tokens_seen": 30907260,
      "step": 1464,
      "time_per_iteration": 2.712571859359741
    },
    {
      "auxiliary_loss_clip": 0.01107143,
      "auxiliary_loss_mlp": 0.01003983,
      "balance_loss_clip": 1.03477573,
      "balance_loss_mlp": 1.00152731,
      "epoch": 0.17615583478626826,
      "flos": 60530775373440.0,
      "grad_norm": 0.860354892196449,
      "language_loss": 0.57928407,
      "learning_rate": 3.7802059407127915e-06,
      "loss": 0.60039532,
      "num_input_tokens_seen": 30965810,
      "step": 1465,
      "time_per_iteration": 3.2358028888702393
    },
    {
      "auxiliary_loss_clip": 0.01208345,
      "auxiliary_loss_mlp": 0.01038645,
      "balance_loss_clip": 1.06719339,
      "balance_loss_mlp": 1.02923369,
      "epoch": 0.17627607767690734,
      "flos": 23616221362560.0,
      "grad_norm": 3.5598765426838552,
      "language_loss": 0.86482894,
      "learning_rate": 3.7798507819821797e-06,
      "loss": 0.88729882,
      "num_input_tokens_seen": 30982935,
      "step": 1466,
      "time_per_iteration": 2.723066568374634
    },
    {
      "auxiliary_loss_clip": 0.01214724,
      "auxiliary_loss_mlp": 0.01035896,
      "balance_loss_clip": 1.06926298,
      "balance_loss_mlp": 1.02613878,
      "epoch": 0.17639632056754645,
      "flos": 17639070589440.0,
      "grad_norm": 6.848878952654557,
      "language_loss": 0.78968871,
      "learning_rate": 3.7794953532496197e-06,
      "loss": 0.81219482,
      "num_input_tokens_seen": 30998840,
      "step": 1467,
      "time_per_iteration": 3.9168670177459717
    },
    {
      "auxiliary_loss_clip": 0.01095707,
      "auxiliary_loss_mlp": 0.01008397,
      "balance_loss_clip": 1.03482199,
      "balance_loss_mlp": 1.00577402,
      "epoch": 0.17651656345818553,
      "flos": 57932604910080.0,
      "grad_norm": 0.8813570077874934,
      "language_loss": 0.57910061,
      "learning_rate": 3.7791396545690295e-06,
      "loss": 0.60014164,
      "num_input_tokens_seen": 31060075,
      "step": 1468,
      "time_per_iteration": 3.286609411239624
    },
    {
      "auxiliary_loss_clip": 0.0119641,
      "auxiliary_loss_mlp": 0.01036574,
      "balance_loss_clip": 1.06681395,
      "balance_loss_mlp": 1.02645886,
      "epoch": 0.17663680634882462,
      "flos": 22929502170240.0,
      "grad_norm": 3.0885640082153683,
      "language_loss": 0.81003994,
      "learning_rate": 3.7787836859943685e-06,
      "loss": 0.8323698,
      "num_input_tokens_seen": 31078800,
      "step": 1469,
      "time_per_iteration": 3.7181646823883057
    },
    {
      "auxiliary_loss_clip": 0.01141635,
      "auxiliary_loss_mlp": 0.01036695,
      "balance_loss_clip": 1.05654359,
      "balance_loss_mlp": 1.02601397,
      "epoch": 0.17675704923946373,
      "flos": 22637979388800.0,
      "grad_norm": 2.58702831916529,
      "language_loss": 0.79118085,
      "learning_rate": 3.7784274475796363e-06,
      "loss": 0.8129642,
      "num_input_tokens_seen": 31097430,
      "step": 1470,
      "time_per_iteration": 2.8182785511016846
    },
    {
      "auxiliary_loss_clip": 0.01225026,
      "auxiliary_loss_mlp": 0.01033039,
      "balance_loss_clip": 1.07060933,
      "balance_loss_mlp": 1.02180386,
      "epoch": 0.1768772921301028,
      "flos": 27126525795840.0,
      "grad_norm": 34.39377402210615,
      "language_loss": 0.75654989,
      "learning_rate": 3.7780709393788745e-06,
      "loss": 0.77913058,
      "num_input_tokens_seen": 31117905,
      "step": 1471,
      "time_per_iteration": 2.776791572570801
    },
    {
      "auxiliary_loss_clip": 0.01167518,
      "auxiliary_loss_mlp": 0.01036257,
      "balance_loss_clip": 1.05646133,
      "balance_loss_mlp": 1.02625585,
      "epoch": 0.1769975350207419,
      "flos": 19172133014400.0,
      "grad_norm": 5.8099638665477595,
      "language_loss": 0.75072378,
      "learning_rate": 3.777714161446165e-06,
      "loss": 0.77276146,
      "num_input_tokens_seen": 31137610,
      "step": 1472,
      "time_per_iteration": 3.7656021118164062
    },
    {
      "auxiliary_loss_clip": 0.01208268,
      "auxiliary_loss_mlp": 0.01037011,
      "balance_loss_clip": 1.06983089,
      "balance_loss_mlp": 1.02631819,
      "epoch": 0.177117777911381,
      "flos": 36134932291200.0,
      "grad_norm": 2.907317921482434,
      "language_loss": 0.69295371,
      "learning_rate": 3.7773571138356304e-06,
      "loss": 0.71540648,
      "num_input_tokens_seen": 31157780,
      "step": 1473,
      "time_per_iteration": 2.8837852478027344
    },
    {
      "auxiliary_loss_clip": 0.01220103,
      "auxiliary_loss_mlp": 0.01047037,
      "balance_loss_clip": 1.07185793,
      "balance_loss_mlp": 1.0379889,
      "epoch": 0.17723802080202009,
      "flos": 22090593052800.0,
      "grad_norm": 2.635878424360158,
      "language_loss": 0.89430285,
      "learning_rate": 3.776999796601435e-06,
      "loss": 0.91697419,
      "num_input_tokens_seen": 31176540,
      "step": 1474,
      "time_per_iteration": 2.6593265533447266
    },
    {
      "auxiliary_loss_clip": 0.01218547,
      "auxiliary_loss_mlp": 0.01045428,
      "balance_loss_clip": 1.07172489,
      "balance_loss_mlp": 1.03514624,
      "epoch": 0.17735826369265917,
      "flos": 30222671437440.0,
      "grad_norm": 2.746437246842272,
      "language_loss": 0.72569335,
      "learning_rate": 3.776642209797783e-06,
      "loss": 0.7483331,
      "num_input_tokens_seen": 31198370,
      "step": 1475,
      "time_per_iteration": 2.7797091007232666
    },
    {
      "auxiliary_loss_clip": 0.01170465,
      "auxiliary_loss_mlp": 0.01041286,
      "balance_loss_clip": 1.06677938,
      "balance_loss_mlp": 1.03074825,
      "epoch": 0.17747850658329825,
      "flos": 21397588980480.0,
      "grad_norm": 2.6284116107553954,
      "language_loss": 0.77922678,
      "learning_rate": 3.7762843534789205e-06,
      "loss": 0.80134428,
      "num_input_tokens_seen": 31217120,
      "step": 1476,
      "time_per_iteration": 3.759772300720215
    },
    {
      "auxiliary_loss_clip": 0.01162102,
      "auxiliary_loss_mlp": 0.01044246,
      "balance_loss_clip": 1.06092429,
      "balance_loss_mlp": 1.03286755,
      "epoch": 0.17759874947393736,
      "flos": 16983341856000.0,
      "grad_norm": 2.971513209559404,
      "language_loss": 0.8838442,
      "learning_rate": 3.7759262276991343e-06,
      "loss": 0.90590769,
      "num_input_tokens_seen": 31234730,
      "step": 1477,
      "time_per_iteration": 2.751391887664795
    },
    {
      "auxiliary_loss_clip": 0.01226179,
      "auxiliary_loss_mlp": 0.00764479,
      "balance_loss_clip": 1.0699147,
      "balance_loss_mlp": 1.00057173,
      "epoch": 0.17771899236457644,
      "flos": 11546107390080.0,
      "grad_norm": 2.37894714641171,
      "language_loss": 0.80798686,
      "learning_rate": 3.7755678325127506e-06,
      "loss": 0.8278935,
      "num_input_tokens_seen": 31252410,
      "step": 1478,
      "time_per_iteration": 2.7026114463806152
    },
    {
      "auxiliary_loss_clip": 0.01188337,
      "auxiliary_loss_mlp": 0.01044018,
      "balance_loss_clip": 1.06289649,
      "balance_loss_mlp": 1.03275836,
      "epoch": 0.17783923525521553,
      "flos": 18807747494400.0,
      "grad_norm": 2.0172984366568465,
      "language_loss": 0.75457036,
      "learning_rate": 3.7752091679741393e-06,
      "loss": 0.77689385,
      "num_input_tokens_seen": 31270200,
      "step": 1479,
      "time_per_iteration": 2.746429443359375
    },
    {
      "auxiliary_loss_clip": 0.01180158,
      "auxiliary_loss_mlp": 0.01041029,
      "balance_loss_clip": 1.06607533,
      "balance_loss_mlp": 1.02966237,
      "epoch": 0.17795947814585464,
      "flos": 30408365773440.0,
      "grad_norm": 3.931125378616854,
      "language_loss": 0.7789886,
      "learning_rate": 3.774850234137708e-06,
      "loss": 0.80120045,
      "num_input_tokens_seen": 31287495,
      "step": 1480,
      "time_per_iteration": 2.868115186691284
    },
    {
      "auxiliary_loss_clip": 0.01178057,
      "auxiliary_loss_mlp": 0.01043932,
      "balance_loss_clip": 1.06568885,
      "balance_loss_mlp": 1.03368556,
      "epoch": 0.17807972103649372,
      "flos": 24389055411840.0,
      "grad_norm": 3.5107459025929755,
      "language_loss": 0.82966304,
      "learning_rate": 3.7744910310579076e-06,
      "loss": 0.85188287,
      "num_input_tokens_seen": 31306420,
      "step": 1481,
      "time_per_iteration": 2.7368032932281494
    },
    {
      "auxiliary_loss_clip": 0.01161484,
      "auxiliary_loss_mlp": 0.01038559,
      "balance_loss_clip": 1.05829549,
      "balance_loss_mlp": 1.02859354,
      "epoch": 0.1781999639271328,
      "flos": 20301559332480.0,
      "grad_norm": 1.8671541936599907,
      "language_loss": 0.85274297,
      "learning_rate": 3.774131558789229e-06,
      "loss": 0.8747434,
      "num_input_tokens_seen": 31325750,
      "step": 1482,
      "time_per_iteration": 2.881420612335205
    },
    {
      "auxiliary_loss_clip": 0.0119955,
      "auxiliary_loss_mlp": 0.01038696,
      "balance_loss_clip": 1.0678705,
      "balance_loss_mlp": 1.02769852,
      "epoch": 0.1783202068177719,
      "flos": 15924479806080.0,
      "grad_norm": 2.5593211261011994,
      "language_loss": 0.69486529,
      "learning_rate": 3.773771817386203e-06,
      "loss": 0.71724772,
      "num_input_tokens_seen": 31343080,
      "step": 1483,
      "time_per_iteration": 2.696402072906494
    },
    {
      "auxiliary_loss_clip": 0.01204358,
      "auxiliary_loss_mlp": 0.01035744,
      "balance_loss_clip": 1.06635118,
      "balance_loss_mlp": 1.02454472,
      "epoch": 0.178440449708411,
      "flos": 20631758083200.0,
      "grad_norm": 1.9641357724742057,
      "language_loss": 0.79220873,
      "learning_rate": 3.773411806903403e-06,
      "loss": 0.81460977,
      "num_input_tokens_seen": 31362160,
      "step": 1484,
      "time_per_iteration": 2.7413957118988037
    },
    {
      "auxiliary_loss_clip": 0.0117132,
      "auxiliary_loss_mlp": 0.01039132,
      "balance_loss_clip": 1.06014252,
      "balance_loss_mlp": 1.02889228,
      "epoch": 0.17856069259905008,
      "flos": 21686059105920.0,
      "grad_norm": 2.936914000320309,
      "language_loss": 0.94609523,
      "learning_rate": 3.7730515273954415e-06,
      "loss": 0.96819985,
      "num_input_tokens_seen": 31380770,
      "step": 1485,
      "time_per_iteration": 2.8054463863372803
    },
    {
      "auxiliary_loss_clip": 0.0115711,
      "auxiliary_loss_mlp": 0.01039589,
      "balance_loss_clip": 1.06350732,
      "balance_loss_mlp": 1.0292356,
      "epoch": 0.17868093548968916,
      "flos": 26572962320640.0,
      "grad_norm": 2.12881522749061,
      "language_loss": 0.8507095,
      "learning_rate": 3.772690978916973e-06,
      "loss": 0.87267649,
      "num_input_tokens_seen": 31400525,
      "step": 1486,
      "time_per_iteration": 2.828667402267456
    },
    {
      "auxiliary_loss_clip": 0.01174974,
      "auxiliary_loss_mlp": 0.00764736,
      "balance_loss_clip": 1.06243873,
      "balance_loss_mlp": 1.00050616,
      "epoch": 0.17880117838032827,
      "flos": 18581006891520.0,
      "grad_norm": 3.215101372457968,
      "language_loss": 0.86776465,
      "learning_rate": 3.772330161522693e-06,
      "loss": 0.88716173,
      "num_input_tokens_seen": 31418435,
      "step": 1487,
      "time_per_iteration": 2.7121241092681885
    },
    {
      "auxiliary_loss_clip": 0.01143532,
      "auxiliary_loss_mlp": 0.01043409,
      "balance_loss_clip": 1.06031561,
      "balance_loss_mlp": 1.03233445,
      "epoch": 0.17892142127096736,
      "flos": 26541217676160.0,
      "grad_norm": 3.7766948739303485,
      "language_loss": 0.79976666,
      "learning_rate": 3.7719690752673365e-06,
      "loss": 0.82163608,
      "num_input_tokens_seen": 31439230,
      "step": 1488,
      "time_per_iteration": 2.8587379455566406
    },
    {
      "auxiliary_loss_clip": 0.01199446,
      "auxiliary_loss_mlp": 0.0104269,
      "balance_loss_clip": 1.06654,
      "balance_loss_mlp": 1.03183007,
      "epoch": 0.17904166416160644,
      "flos": 23872623621120.0,
      "grad_norm": 7.162530095859367,
      "language_loss": 0.78245181,
      "learning_rate": 3.7716077202056796e-06,
      "loss": 0.80487317,
      "num_input_tokens_seen": 31457705,
      "step": 1489,
      "time_per_iteration": 2.7708613872528076
    },
    {
      "auxiliary_loss_clip": 0.01166036,
      "auxiliary_loss_mlp": 0.01046019,
      "balance_loss_clip": 1.05389929,
      "balance_loss_mlp": 1.03490257,
      "epoch": 0.17916190705224552,
      "flos": 19134426712320.0,
      "grad_norm": 10.403672317701332,
      "language_loss": 0.93457097,
      "learning_rate": 3.7712460963925404e-06,
      "loss": 0.9566915,
      "num_input_tokens_seen": 31473645,
      "step": 1490,
      "time_per_iteration": 2.7516543865203857
    },
    {
      "auxiliary_loss_clip": 0.01184452,
      "auxiliary_loss_mlp": 0.00764284,
      "balance_loss_clip": 1.0660547,
      "balance_loss_mlp": 1.00044084,
      "epoch": 0.17928214994288463,
      "flos": 25152120961920.0,
      "grad_norm": 14.233236108921323,
      "language_loss": 0.75333792,
      "learning_rate": 3.7708842038827775e-06,
      "loss": 0.77282524,
      "num_input_tokens_seen": 31492605,
      "step": 1491,
      "time_per_iteration": 2.7890894412994385
    },
    {
      "auxiliary_loss_clip": 0.01224343,
      "auxiliary_loss_mlp": 0.00763934,
      "balance_loss_clip": 1.07144034,
      "balance_loss_mlp": 1.00050366,
      "epoch": 0.17940239283352372,
      "flos": 22384629786240.0,
      "grad_norm": 1.7206541833324105,
      "language_loss": 0.85906458,
      "learning_rate": 3.770522042731288e-06,
      "loss": 0.87894726,
      "num_input_tokens_seen": 31514500,
      "step": 1492,
      "time_per_iteration": 2.6744883060455322
    },
    {
      "auxiliary_loss_clip": 0.0117328,
      "auxiliary_loss_mlp": 0.01045948,
      "balance_loss_clip": 1.06171072,
      "balance_loss_mlp": 1.03506398,
      "epoch": 0.1795226357241628,
      "flos": 23178685795200.0,
      "grad_norm": 2.340182745178562,
      "language_loss": 0.88062251,
      "learning_rate": 3.7701596129930122e-06,
      "loss": 0.90281487,
      "num_input_tokens_seen": 31533225,
      "step": 1493,
      "time_per_iteration": 2.823967456817627
    },
    {
      "auxiliary_loss_clip": 0.01154218,
      "auxiliary_loss_mlp": 0.00764853,
      "balance_loss_clip": 1.05861044,
      "balance_loss_mlp": 1.00059116,
      "epoch": 0.1796428786148019,
      "flos": 22090413484800.0,
      "grad_norm": 4.6834439497814495,
      "language_loss": 0.7330507,
      "learning_rate": 3.7697969147229315e-06,
      "loss": 0.75224143,
      "num_input_tokens_seen": 31551385,
      "step": 1494,
      "time_per_iteration": 4.047548532485962
    },
    {
      "auxiliary_loss_clip": 0.0123666,
      "auxiliary_loss_mlp": 0.01037847,
      "balance_loss_clip": 1.07333875,
      "balance_loss_mlp": 1.02704704,
      "epoch": 0.179763121505441,
      "flos": 21324618501120.0,
      "grad_norm": 2.8227682904875806,
      "language_loss": 0.85051662,
      "learning_rate": 3.7694339479760647e-06,
      "loss": 0.87326169,
      "num_input_tokens_seen": 31570415,
      "step": 1495,
      "time_per_iteration": 2.675382137298584
    },
    {
      "auxiliary_loss_clip": 0.01064762,
      "auxiliary_loss_mlp": 0.01004172,
      "balance_loss_clip": 1.03810906,
      "balance_loss_mlp": 1.00197887,
      "epoch": 0.17988336439608008,
      "flos": 68161864815360.0,
      "grad_norm": 0.7732311766553354,
      "language_loss": 0.57276648,
      "learning_rate": 3.769070712807476e-06,
      "loss": 0.59345573,
      "num_input_tokens_seen": 31632445,
      "step": 1496,
      "time_per_iteration": 4.739895582199097
    },
    {
      "auxiliary_loss_clip": 0.01214496,
      "auxiliary_loss_mlp": 0.01034719,
      "balance_loss_clip": 1.07152271,
      "balance_loss_mlp": 1.02424037,
      "epoch": 0.18000360728671919,
      "flos": 21945047143680.0,
      "grad_norm": 2.1645664945836245,
      "language_loss": 0.7901364,
      "learning_rate": 3.768707209272266e-06,
      "loss": 0.81262851,
      "num_input_tokens_seen": 31652575,
      "step": 1497,
      "time_per_iteration": 3.3039002418518066
    },
    {
      "auxiliary_loss_clip": 0.01183658,
      "auxiliary_loss_mlp": 0.01042892,
      "balance_loss_clip": 1.06311011,
      "balance_loss_mlp": 1.03259254,
      "epoch": 0.18012385017735827,
      "flos": 18986330937600.0,
      "grad_norm": 2.8923208566741763,
      "language_loss": 0.76697022,
      "learning_rate": 3.768343437425579e-06,
      "loss": 0.78923571,
      "num_input_tokens_seen": 31671145,
      "step": 1498,
      "time_per_iteration": 3.632354259490967
    },
    {
      "auxiliary_loss_clip": 0.01211396,
      "auxiliary_loss_mlp": 0.01041648,
      "balance_loss_clip": 1.0700748,
      "balance_loss_mlp": 1.03047824,
      "epoch": 0.18024409306799735,
      "flos": 19748103598080.0,
      "grad_norm": 3.5004223979139035,
      "language_loss": 0.85779655,
      "learning_rate": 3.7679793973225987e-06,
      "loss": 0.88032705,
      "num_input_tokens_seen": 31686955,
      "step": 1499,
      "time_per_iteration": 2.7396183013916016
    },
    {
      "auxiliary_loss_clip": 0.01071791,
      "auxiliary_loss_mlp": 0.0100362,
      "balance_loss_clip": 1.03083777,
      "balance_loss_mlp": 1.00092602,
      "epoch": 0.18036433595863643,
      "flos": 67227183060480.0,
      "grad_norm": 0.8479859430917048,
      "language_loss": 0.61518997,
      "learning_rate": 3.767615089018549e-06,
      "loss": 0.63594413,
      "num_input_tokens_seen": 31749300,
      "step": 1500,
      "time_per_iteration": 3.3739190101623535
    },
    {
      "auxiliary_loss_clip": 0.01230492,
      "auxiliary_loss_mlp": 0.01045178,
      "balance_loss_clip": 1.07427239,
      "balance_loss_mlp": 1.0339005,
      "epoch": 0.18048457884927555,
      "flos": 18181464935040.0,
      "grad_norm": 3.6739881913205132,
      "language_loss": 0.86052668,
      "learning_rate": 3.7672505125686966e-06,
      "loss": 0.88328338,
      "num_input_tokens_seen": 31765665,
      "step": 1501,
      "time_per_iteration": 2.820586919784546
    },
    {
      "auxiliary_loss_clip": 0.01192982,
      "auxiliary_loss_mlp": 0.01047899,
      "balance_loss_clip": 1.07076192,
      "balance_loss_mlp": 1.03674138,
      "epoch": 0.18060482173991463,
      "flos": 15813767111040.0,
      "grad_norm": 4.516804354189795,
      "language_loss": 0.83781981,
      "learning_rate": 3.7668856680283455e-06,
      "loss": 0.86022866,
      "num_input_tokens_seen": 31782690,
      "step": 1502,
      "time_per_iteration": 3.7545173168182373
    },
    {
      "auxiliary_loss_clip": 0.01216361,
      "auxiliary_loss_mlp": 0.01047002,
      "balance_loss_clip": 1.07384539,
      "balance_loss_mlp": 1.03618407,
      "epoch": 0.1807250646305537,
      "flos": 18587399512320.0,
      "grad_norm": 5.350741001974188,
      "language_loss": 0.82443321,
      "learning_rate": 3.7665205554528437e-06,
      "loss": 0.84706682,
      "num_input_tokens_seen": 31802045,
      "step": 1503,
      "time_per_iteration": 2.772160053253174
    },
    {
      "auxiliary_loss_clip": 0.01181219,
      "auxiliary_loss_mlp": 0.01046603,
      "balance_loss_clip": 1.06249285,
      "balance_loss_mlp": 1.03520083,
      "epoch": 0.18084530752119282,
      "flos": 23149131880320.0,
      "grad_norm": 2.1419286672565803,
      "language_loss": 0.74013579,
      "learning_rate": 3.7661551748975782e-06,
      "loss": 0.76241404,
      "num_input_tokens_seen": 31820220,
      "step": 1504,
      "time_per_iteration": 2.758787155151367
    },
    {
      "auxiliary_loss_clip": 0.01119972,
      "auxiliary_loss_mlp": 0.01007556,
      "balance_loss_clip": 1.03149652,
      "balance_loss_mlp": 1.00515962,
      "epoch": 0.1809655504118319,
      "flos": 59803153568640.0,
      "grad_norm": 0.8141882707209481,
      "language_loss": 0.60457623,
      "learning_rate": 3.7657895264179772e-06,
      "loss": 0.62585151,
      "num_input_tokens_seen": 31876195,
      "step": 1505,
      "time_per_iteration": 3.2105727195739746
    },
    {
      "auxiliary_loss_clip": 0.01155489,
      "auxiliary_loss_mlp": 0.01041796,
      "balance_loss_clip": 1.05730188,
      "balance_loss_mlp": 1.03028035,
      "epoch": 0.181085793302471,
      "flos": 44201941188480.0,
      "grad_norm": 1.9233880055110442,
      "language_loss": 0.74338531,
      "learning_rate": 3.765423610069509e-06,
      "loss": 0.76535815,
      "num_input_tokens_seen": 31901585,
      "step": 1506,
      "time_per_iteration": 2.947035074234009
    },
    {
      "auxiliary_loss_clip": 0.01203046,
      "auxiliary_loss_mlp": 0.01047076,
      "balance_loss_clip": 1.06854379,
      "balance_loss_mlp": 1.03621006,
      "epoch": 0.18120603619311007,
      "flos": 34898384638080.0,
      "grad_norm": 2.152373556316171,
      "language_loss": 0.72506475,
      "learning_rate": 3.765057425907683e-06,
      "loss": 0.74756598,
      "num_input_tokens_seen": 31923045,
      "step": 1507,
      "time_per_iteration": 2.8837015628814697
    },
    {
      "auxiliary_loss_clip": 0.01158669,
      "auxiliary_loss_mlp": 0.01052298,
      "balance_loss_clip": 1.05939281,
      "balance_loss_mlp": 1.04105639,
      "epoch": 0.18132627908374918,
      "flos": 21506757390720.0,
      "grad_norm": 2.674764655462687,
      "language_loss": 0.78515649,
      "learning_rate": 3.764690973988048e-06,
      "loss": 0.80726624,
      "num_input_tokens_seen": 31943385,
      "step": 1508,
      "time_per_iteration": 2.832873821258545
    },
    {
      "auxiliary_loss_clip": 0.01164411,
      "auxiliary_loss_mlp": 0.01045063,
      "balance_loss_clip": 1.06187844,
      "balance_loss_mlp": 1.03346395,
      "epoch": 0.18144652197438826,
      "flos": 29057693633280.0,
      "grad_norm": 7.294070982146748,
      "language_loss": 0.73633957,
      "learning_rate": 3.7643242543661967e-06,
      "loss": 0.7584343,
      "num_input_tokens_seen": 31966045,
      "step": 1509,
      "time_per_iteration": 2.843557834625244
    },
    {
      "auxiliary_loss_clip": 0.01075352,
      "auxiliary_loss_mlp": 0.01009227,
      "balance_loss_clip": 1.03250897,
      "balance_loss_mlp": 1.0069859,
      "epoch": 0.18156676486502735,
      "flos": 68675064382080.0,
      "grad_norm": 0.8276587200713054,
      "language_loss": 0.60504895,
      "learning_rate": 3.7639572670977573e-06,
      "loss": 0.62589473,
      "num_input_tokens_seen": 32021540,
      "step": 1510,
      "time_per_iteration": 3.2726285457611084
    },
    {
      "auxiliary_loss_clip": 0.01188689,
      "auxiliary_loss_mlp": 0.01042591,
      "balance_loss_clip": 1.06863725,
      "balance_loss_mlp": 1.03174913,
      "epoch": 0.18168700775566646,
      "flos": 26471515334400.0,
      "grad_norm": 2.2768759405346533,
      "language_loss": 0.76543152,
      "learning_rate": 3.7635900122384042e-06,
      "loss": 0.78774428,
      "num_input_tokens_seen": 32044535,
      "step": 1511,
      "time_per_iteration": 2.793851852416992
    },
    {
      "auxiliary_loss_clip": 0.01207726,
      "auxiliary_loss_mlp": 0.01050984,
      "balance_loss_clip": 1.0672431,
      "balance_loss_mlp": 1.03934896,
      "epoch": 0.18180725064630554,
      "flos": 15005668884480.0,
      "grad_norm": 2.6697813910448045,
      "language_loss": 0.86812317,
      "learning_rate": 3.7632224898438477e-06,
      "loss": 0.89071023,
      "num_input_tokens_seen": 32061010,
      "step": 1512,
      "time_per_iteration": 2.743767023086548
    },
    {
      "auxiliary_loss_clip": 0.01191676,
      "auxiliary_loss_mlp": 0.01032136,
      "balance_loss_clip": 1.06603026,
      "balance_loss_mlp": 1.02161539,
      "epoch": 0.18192749353694462,
      "flos": 19682387665920.0,
      "grad_norm": 2.068993647842336,
      "language_loss": 0.79142237,
      "learning_rate": 3.762854699969842e-06,
      "loss": 0.81366044,
      "num_input_tokens_seen": 32081520,
      "step": 1513,
      "time_per_iteration": 2.7657723426818848
    },
    {
      "auxiliary_loss_clip": 0.01232702,
      "auxiliary_loss_mlp": 0.01042424,
      "balance_loss_clip": 1.06929457,
      "balance_loss_mlp": 1.03220797,
      "epoch": 0.1820477364275837,
      "flos": 20702717400960.0,
      "grad_norm": 3.93031179251084,
      "language_loss": 0.73375756,
      "learning_rate": 3.762486642672179e-06,
      "loss": 0.75650883,
      "num_input_tokens_seen": 32098460,
      "step": 1514,
      "time_per_iteration": 2.7371387481689453
    },
    {
      "auxiliary_loss_clip": 0.01222835,
      "auxiliary_loss_mlp": 0.01048124,
      "balance_loss_clip": 1.0718745,
      "balance_loss_mlp": 1.0363462,
      "epoch": 0.18216797931822282,
      "flos": 17128708197120.0,
      "grad_norm": 4.167005448084708,
      "language_loss": 0.87194872,
      "learning_rate": 3.7621183180066946e-06,
      "loss": 0.89465827,
      "num_input_tokens_seen": 32116420,
      "step": 1515,
      "time_per_iteration": 2.649214267730713
    },
    {
      "auxiliary_loss_clip": 0.01171151,
      "auxiliary_loss_mlp": 0.01037236,
      "balance_loss_clip": 1.06397402,
      "balance_loss_mlp": 1.02591705,
      "epoch": 0.1822882222088619,
      "flos": 29242561956480.0,
      "grad_norm": 2.0067101243597274,
      "language_loss": 0.73462355,
      "learning_rate": 3.7617497260292625e-06,
      "loss": 0.75670737,
      "num_input_tokens_seen": 32138475,
      "step": 1516,
      "time_per_iteration": 2.8408825397491455
    },
    {
      "auxiliary_loss_clip": 0.01223192,
      "auxiliary_loss_mlp": 0.01042226,
      "balance_loss_clip": 1.069947,
      "balance_loss_mlp": 1.03013766,
      "epoch": 0.18240846509950098,
      "flos": 17702739446400.0,
      "grad_norm": 3.5175080699323757,
      "language_loss": 0.78722447,
      "learning_rate": 3.7613808667957967e-06,
      "loss": 0.80987865,
      "num_input_tokens_seen": 32151165,
      "step": 1517,
      "time_per_iteration": 2.6759912967681885
    },
    {
      "auxiliary_loss_clip": 0.01143522,
      "auxiliary_loss_mlp": 0.01039206,
      "balance_loss_clip": 1.05959487,
      "balance_loss_mlp": 1.02837563,
      "epoch": 0.1825287079901401,
      "flos": 14790025584000.0,
      "grad_norm": 2.3583530509612354,
      "language_loss": 0.90962529,
      "learning_rate": 3.7610117403622547e-06,
      "loss": 0.93145251,
      "num_input_tokens_seen": 32167725,
      "step": 1518,
      "time_per_iteration": 2.820289134979248
    },
    {
      "auxiliary_loss_clip": 0.01150831,
      "auxiliary_loss_mlp": 0.01040597,
      "balance_loss_clip": 1.05203152,
      "balance_loss_mlp": 1.029814,
      "epoch": 0.18264895088077918,
      "flos": 21946232292480.0,
      "grad_norm": 3.924624804788172,
      "language_loss": 0.90229911,
      "learning_rate": 3.7606423467846313e-06,
      "loss": 0.92421341,
      "num_input_tokens_seen": 32187330,
      "step": 1519,
      "time_per_iteration": 4.277153968811035
    },
    {
      "auxiliary_loss_clip": 0.01222533,
      "auxiliary_loss_mlp": 0.0104642,
      "balance_loss_clip": 1.07111704,
      "balance_loss_mlp": 1.03529179,
      "epoch": 0.18276919377141826,
      "flos": 20886759711360.0,
      "grad_norm": 2.629997867741618,
      "language_loss": 0.79762685,
      "learning_rate": 3.760272686118964e-06,
      "loss": 0.82031637,
      "num_input_tokens_seen": 32205550,
      "step": 1520,
      "time_per_iteration": 2.663816213607788
    },
    {
      "auxiliary_loss_clip": 0.01195461,
      "auxiliary_loss_mlp": 0.01037793,
      "balance_loss_clip": 1.06492257,
      "balance_loss_mlp": 1.02733815,
      "epoch": 0.18288943666205737,
      "flos": 21469877101440.0,
      "grad_norm": 3.244867619468247,
      "language_loss": 0.92666984,
      "learning_rate": 3.7599027584213297e-06,
      "loss": 0.94900244,
      "num_input_tokens_seen": 32224430,
      "step": 1521,
      "time_per_iteration": 2.7468068599700928
    },
    {
      "auxiliary_loss_clip": 0.01234015,
      "auxiliary_loss_mlp": 0.01042713,
      "balance_loss_clip": 1.07061768,
      "balance_loss_mlp": 1.03198457,
      "epoch": 0.18300967955269645,
      "flos": 21539363961600.0,
      "grad_norm": 2.4756791589166585,
      "language_loss": 0.78145599,
      "learning_rate": 3.7595325637478465e-06,
      "loss": 0.8042233,
      "num_input_tokens_seen": 32242455,
      "step": 1522,
      "time_per_iteration": 3.5984444618225098
    },
    {
      "auxiliary_loss_clip": 0.01219785,
      "auxiliary_loss_mlp": 0.00764542,
      "balance_loss_clip": 1.07320118,
      "balance_loss_mlp": 1.00062907,
      "epoch": 0.18312992244333554,
      "flos": 28876237102080.0,
      "grad_norm": 2.366732561076298,
      "language_loss": 0.81746036,
      "learning_rate": 3.7591621021546723e-06,
      "loss": 0.83730364,
      "num_input_tokens_seen": 32264450,
      "step": 1523,
      "time_per_iteration": 3.658748149871826
    },
    {
      "auxiliary_loss_clip": 0.01190961,
      "auxiliary_loss_mlp": 0.01047141,
      "balance_loss_clip": 1.06682003,
      "balance_loss_mlp": 1.03653693,
      "epoch": 0.18325016533397462,
      "flos": 20120102801280.0,
      "grad_norm": 16.466768340109827,
      "language_loss": 0.81499147,
      "learning_rate": 3.7587913736980062e-06,
      "loss": 0.83737248,
      "num_input_tokens_seen": 32284090,
      "step": 1524,
      "time_per_iteration": 2.766184091567993
    },
    {
      "auxiliary_loss_clip": 0.01201798,
      "auxiliary_loss_mlp": 0.00764064,
      "balance_loss_clip": 1.06796801,
      "balance_loss_mlp": 1.00058126,
      "epoch": 0.18337040822461373,
      "flos": 23329187781120.0,
      "grad_norm": 2.393000642649308,
      "language_loss": 0.84316081,
      "learning_rate": 3.7584203784340865e-06,
      "loss": 0.86281943,
      "num_input_tokens_seen": 32303260,
      "step": 1525,
      "time_per_iteration": 2.716798782348633
    },
    {
      "auxiliary_loss_clip": 0.01184368,
      "auxiliary_loss_mlp": 0.01044424,
      "balance_loss_clip": 1.06633306,
      "balance_loss_mlp": 1.03380871,
      "epoch": 0.1834906511152528,
      "flos": 25009555881600.0,
      "grad_norm": 3.0020678327573944,
      "language_loss": 0.85913497,
      "learning_rate": 3.7580491164191938e-06,
      "loss": 0.88142288,
      "num_input_tokens_seen": 32321570,
      "step": 1526,
      "time_per_iteration": 2.764993190765381
    },
    {
      "auxiliary_loss_clip": 0.0110396,
      "auxiliary_loss_mlp": 0.0102163,
      "balance_loss_clip": 1.03017616,
      "balance_loss_mlp": 1.01926947,
      "epoch": 0.1836108940058919,
      "flos": 67251493589760.0,
      "grad_norm": 0.7550696450546084,
      "language_loss": 0.61213976,
      "learning_rate": 3.757677587709648e-06,
      "loss": 0.63339567,
      "num_input_tokens_seen": 32384835,
      "step": 1527,
      "time_per_iteration": 3.413280725479126
    },
    {
      "auxiliary_loss_clip": 0.01154752,
      "auxiliary_loss_mlp": 0.01051854,
      "balance_loss_clip": 1.05657375,
      "balance_loss_mlp": 1.04055309,
      "epoch": 0.183731136896531,
      "flos": 25738721971200.0,
      "grad_norm": 9.25098107748228,
      "language_loss": 0.75516367,
      "learning_rate": 3.7573057923618095e-06,
      "loss": 0.77722979,
      "num_input_tokens_seen": 32404930,
      "step": 1528,
      "time_per_iteration": 3.9351816177368164
    },
    {
      "auxiliary_loss_clip": 0.01199778,
      "auxiliary_loss_mlp": 0.01048223,
      "balance_loss_clip": 1.06154561,
      "balance_loss_mlp": 1.03730941,
      "epoch": 0.1838513797871701,
      "flos": 20449403712000.0,
      "grad_norm": 3.2595635175657907,
      "language_loss": 0.74548936,
      "learning_rate": 3.7569337304320793e-06,
      "loss": 0.76796937,
      "num_input_tokens_seen": 32424515,
      "step": 1529,
      "time_per_iteration": 3.02176833152771
    },
    {
      "auxiliary_loss_clip": 0.01106108,
      "auxiliary_loss_mlp": 0.0101112,
      "balance_loss_clip": 1.03024149,
      "balance_loss_mlp": 1.00866401,
      "epoch": 0.18397162267780917,
      "flos": 68565141786240.0,
      "grad_norm": 0.851078704605112,
      "language_loss": 0.64406139,
      "learning_rate": 3.756561401976899e-06,
      "loss": 0.66523367,
      "num_input_tokens_seen": 32484220,
      "step": 1530,
      "time_per_iteration": 3.1816344261169434
    },
    {
      "auxiliary_loss_clip": 0.01238661,
      "auxiliary_loss_mlp": 0.0104363,
      "balance_loss_clip": 1.07277393,
      "balance_loss_mlp": 1.03241289,
      "epoch": 0.18409186556844825,
      "flos": 31941104976000.0,
      "grad_norm": 2.2001420083045087,
      "language_loss": 0.82375729,
      "learning_rate": 3.7561888070527514e-06,
      "loss": 0.84658021,
      "num_input_tokens_seen": 32506260,
      "step": 1531,
      "time_per_iteration": 2.7024080753326416
    },
    {
      "auxiliary_loss_clip": 0.0119072,
      "auxiliary_loss_mlp": 0.01034517,
      "balance_loss_clip": 1.06753922,
      "balance_loss_mlp": 1.02397346,
      "epoch": 0.18421210845908736,
      "flos": 20120533764480.0,
      "grad_norm": 2.6722476755996114,
      "language_loss": 0.79927605,
      "learning_rate": 3.7558159457161577e-06,
      "loss": 0.82152843,
      "num_input_tokens_seen": 32524225,
      "step": 1532,
      "time_per_iteration": 2.813209295272827
    },
    {
      "auxiliary_loss_clip": 0.01203427,
      "auxiliary_loss_mlp": 0.01043426,
      "balance_loss_clip": 1.06859684,
      "balance_loss_mlp": 1.03313231,
      "epoch": 0.18433235134972645,
      "flos": 23110491824640.0,
      "grad_norm": 2.433558187702981,
      "language_loss": 0.77800858,
      "learning_rate": 3.755442818023681e-06,
      "loss": 0.80047703,
      "num_input_tokens_seen": 32543850,
      "step": 1533,
      "time_per_iteration": 2.726533889770508
    },
    {
      "auxiliary_loss_clip": 0.01179927,
      "auxiliary_loss_mlp": 0.0103106,
      "balance_loss_clip": 1.06148911,
      "balance_loss_mlp": 1.01986659,
      "epoch": 0.18445259424036553,
      "flos": 18291351617280.0,
      "grad_norm": 2.480788180804267,
      "language_loss": 0.76158011,
      "learning_rate": 3.7550694240319246e-06,
      "loss": 0.78368998,
      "num_input_tokens_seen": 32561725,
      "step": 1534,
      "time_per_iteration": 2.667153835296631
    },
    {
      "auxiliary_loss_clip": 0.01157834,
      "auxiliary_loss_mlp": 0.01046906,
      "balance_loss_clip": 1.05741465,
      "balance_loss_mlp": 1.03620696,
      "epoch": 0.18457283713100464,
      "flos": 21324079797120.0,
      "grad_norm": 4.321915936162151,
      "language_loss": 0.76220757,
      "learning_rate": 3.7546957637975326e-06,
      "loss": 0.78425497,
      "num_input_tokens_seen": 32579135,
      "step": 1535,
      "time_per_iteration": 2.7928948402404785
    },
    {
      "auxiliary_loss_clip": 0.01202914,
      "auxiliary_loss_mlp": 0.01059588,
      "balance_loss_clip": 1.06770194,
      "balance_loss_mlp": 1.04945529,
      "epoch": 0.18469308002164372,
      "flos": 20375679047040.0,
      "grad_norm": 2.386452558437624,
      "language_loss": 0.73938894,
      "learning_rate": 3.7543218373771873e-06,
      "loss": 0.76201391,
      "num_input_tokens_seen": 32598460,
      "step": 1536,
      "time_per_iteration": 2.66678786277771
    },
    {
      "auxiliary_loss_clip": 0.01198239,
      "auxiliary_loss_mlp": 0.01044078,
      "balance_loss_clip": 1.06946492,
      "balance_loss_mlp": 1.03401709,
      "epoch": 0.1848133229122828,
      "flos": 26435892021120.0,
      "grad_norm": 1.4680018134269583,
      "language_loss": 0.78120685,
      "learning_rate": 3.753947644827615e-06,
      "loss": 0.80363005,
      "num_input_tokens_seen": 32621920,
      "step": 1537,
      "time_per_iteration": 2.8200814723968506
    },
    {
      "auxiliary_loss_clip": 0.01104896,
      "auxiliary_loss_mlp": 0.01022889,
      "balance_loss_clip": 1.02967596,
      "balance_loss_mlp": 1.02048099,
      "epoch": 0.1849335658029219,
      "flos": 70547447612160.0,
      "grad_norm": 0.9734672309813696,
      "language_loss": 0.57128942,
      "learning_rate": 3.753573186205579e-06,
      "loss": 0.59256721,
      "num_input_tokens_seen": 32690040,
      "step": 1538,
      "time_per_iteration": 3.3691868782043457
    },
    {
      "auxiliary_loss_clip": 0.01184669,
      "auxiliary_loss_mlp": 0.01041728,
      "balance_loss_clip": 1.06626654,
      "balance_loss_mlp": 1.03160071,
      "epoch": 0.185053808693561,
      "flos": 17384140788480.0,
      "grad_norm": 2.195564796440639,
      "language_loss": 0.78276813,
      "learning_rate": 3.753198461567885e-06,
      "loss": 0.80503207,
      "num_input_tokens_seen": 32707285,
      "step": 1539,
      "time_per_iteration": 2.7748122215270996
    },
    {
      "auxiliary_loss_clip": 0.01210443,
      "auxiliary_loss_mlp": 0.01041376,
      "balance_loss_clip": 1.06838,
      "balance_loss_mlp": 1.03108275,
      "epoch": 0.18517405158420008,
      "flos": 28986159697920.0,
      "grad_norm": 2.197119348777844,
      "language_loss": 0.9193275,
      "learning_rate": 3.7528234709713783e-06,
      "loss": 0.94184566,
      "num_input_tokens_seen": 32730030,
      "step": 1540,
      "time_per_iteration": 2.7231616973876953
    },
    {
      "auxiliary_loss_clip": 0.01193533,
      "auxiliary_loss_mlp": 0.01048005,
      "balance_loss_clip": 1.06351984,
      "balance_loss_mlp": 1.03787231,
      "epoch": 0.18529429447483917,
      "flos": 26794962328320.0,
      "grad_norm": 1.979114676693441,
      "language_loss": 0.84618968,
      "learning_rate": 3.7524482144729447e-06,
      "loss": 0.86860508,
      "num_input_tokens_seen": 32749485,
      "step": 1541,
      "time_per_iteration": 2.778031826019287
    },
    {
      "auxiliary_loss_clip": 0.01187451,
      "auxiliary_loss_mlp": 0.01037611,
      "balance_loss_clip": 1.06995308,
      "balance_loss_mlp": 1.02731705,
      "epoch": 0.18541453736547828,
      "flos": 13581595301760.0,
      "grad_norm": 14.435226524849295,
      "language_loss": 0.83921957,
      "learning_rate": 3.7520726921295106e-06,
      "loss": 0.8614701,
      "num_input_tokens_seen": 32766205,
      "step": 1542,
      "time_per_iteration": 2.688579559326172
    },
    {
      "auxiliary_loss_clip": 0.01169742,
      "auxiliary_loss_mlp": 0.01041202,
      "balance_loss_clip": 1.06274593,
      "balance_loss_mlp": 1.03094971,
      "epoch": 0.18553478025611736,
      "flos": 24025424077440.0,
      "grad_norm": 2.593844419565104,
      "language_loss": 0.71938753,
      "learning_rate": 3.751696903998042e-06,
      "loss": 0.74149698,
      "num_input_tokens_seen": 32784840,
      "step": 1543,
      "time_per_iteration": 2.780697822570801
    },
    {
      "auxiliary_loss_clip": 0.01201912,
      "auxiliary_loss_mlp": 0.01037982,
      "balance_loss_clip": 1.06615567,
      "balance_loss_mlp": 1.02770066,
      "epoch": 0.18565502314675644,
      "flos": 25885165720320.0,
      "grad_norm": 2.2768749380956628,
      "language_loss": 0.69851345,
      "learning_rate": 3.7513208501355456e-06,
      "loss": 0.7209124,
      "num_input_tokens_seen": 32805945,
      "step": 1544,
      "time_per_iteration": 2.777229070663452
    },
    {
      "auxiliary_loss_clip": 0.01228463,
      "auxiliary_loss_mlp": 0.01044377,
      "balance_loss_clip": 1.06870043,
      "balance_loss_mlp": 1.03419089,
      "epoch": 0.18577526603739553,
      "flos": 19610063631360.0,
      "grad_norm": 9.030281117493997,
      "language_loss": 0.84027982,
      "learning_rate": 3.750944530599069e-06,
      "loss": 0.86300826,
      "num_input_tokens_seen": 32825515,
      "step": 1545,
      "time_per_iteration": 2.6110939979553223
    },
    {
      "auxiliary_loss_clip": 0.01176779,
      "auxiliary_loss_mlp": 0.01037498,
      "balance_loss_clip": 1.06362271,
      "balance_loss_mlp": 1.02653658,
      "epoch": 0.18589550892803464,
      "flos": 18474891137280.0,
      "grad_norm": 2.513899007587522,
      "language_loss": 0.80867779,
      "learning_rate": 3.7505679454456992e-06,
      "loss": 0.83082062,
      "num_input_tokens_seen": 32842125,
      "step": 1546,
      "time_per_iteration": 3.6683645248413086
    },
    {
      "auxiliary_loss_clip": 0.01194389,
      "auxiliary_loss_mlp": 0.01032713,
      "balance_loss_clip": 1.0612582,
      "balance_loss_mlp": 1.02150166,
      "epoch": 0.18601575181867372,
      "flos": 23549966726400.0,
      "grad_norm": 4.270596923842119,
      "language_loss": 0.69511819,
      "learning_rate": 3.750191094732564e-06,
      "loss": 0.71738917,
      "num_input_tokens_seen": 32862990,
      "step": 1547,
      "time_per_iteration": 2.736151695251465
    },
    {
      "auxiliary_loss_clip": 0.01218838,
      "auxiliary_loss_mlp": 0.01037848,
      "balance_loss_clip": 1.06953406,
      "balance_loss_mlp": 1.02791786,
      "epoch": 0.1861359947093128,
      "flos": 26360192108160.0,
      "grad_norm": 2.0960807268012913,
      "language_loss": 0.75415152,
      "learning_rate": 3.7498139785168313e-06,
      "loss": 0.77671838,
      "num_input_tokens_seen": 32883595,
      "step": 1548,
      "time_per_iteration": 3.6097538471221924
    },
    {
      "auxiliary_loss_clip": 0.01172075,
      "auxiliary_loss_mlp": 0.01035739,
      "balance_loss_clip": 1.06180823,
      "balance_loss_mlp": 1.0253315,
      "epoch": 0.1862562375999519,
      "flos": 23331198942720.0,
      "grad_norm": 2.2715103991773073,
      "language_loss": 0.77805567,
      "learning_rate": 3.749436596855709e-06,
      "loss": 0.80013388,
      "num_input_tokens_seen": 32902895,
      "step": 1549,
      "time_per_iteration": 3.688610553741455
    },
    {
      "auxiliary_loss_clip": 0.01213897,
      "auxiliary_loss_mlp": 0.01033194,
      "balance_loss_clip": 1.06850493,
      "balance_loss_mlp": 1.02290595,
      "epoch": 0.186376480490591,
      "flos": 16648222942080.0,
      "grad_norm": 2.4996129785260264,
      "language_loss": 0.90461284,
      "learning_rate": 3.749058949806446e-06,
      "loss": 0.92708373,
      "num_input_tokens_seen": 32919620,
      "step": 1550,
      "time_per_iteration": 2.6540327072143555
    },
    {
      "auxiliary_loss_clip": 0.01215985,
      "auxiliary_loss_mlp": 0.01037792,
      "balance_loss_clip": 1.06733632,
      "balance_loss_mlp": 1.02798653,
      "epoch": 0.18649672338123008,
      "flos": 21468656039040.0,
      "grad_norm": 2.168767524935626,
      "language_loss": 0.84189105,
      "learning_rate": 3.748681037426331e-06,
      "loss": 0.86442876,
      "num_input_tokens_seen": 32938830,
      "step": 1551,
      "time_per_iteration": 2.70281982421875
    },
    {
      "auxiliary_loss_clip": 0.01209287,
      "auxiliary_loss_mlp": 0.01041434,
      "balance_loss_clip": 1.06663084,
      "balance_loss_mlp": 1.03012717,
      "epoch": 0.1866169662718692,
      "flos": 12312728386560.0,
      "grad_norm": 2.723265373636597,
      "language_loss": 0.92145586,
      "learning_rate": 3.7483028597726936e-06,
      "loss": 0.94396311,
      "num_input_tokens_seen": 32955600,
      "step": 1552,
      "time_per_iteration": 2.687465190887451
    },
    {
      "auxiliary_loss_clip": 0.01165809,
      "auxiliary_loss_mlp": 0.01039914,
      "balance_loss_clip": 1.05657423,
      "balance_loss_mlp": 1.0287801,
      "epoch": 0.18673720916250827,
      "flos": 23581280407680.0,
      "grad_norm": 2.0951887219998917,
      "language_loss": 0.62285352,
      "learning_rate": 3.7479244169029017e-06,
      "loss": 0.64491069,
      "num_input_tokens_seen": 32975390,
      "step": 1553,
      "time_per_iteration": 2.7663156986236572
    },
    {
      "auxiliary_loss_clip": 0.01211186,
      "auxiliary_loss_mlp": 0.01036907,
      "balance_loss_clip": 1.06546855,
      "balance_loss_mlp": 1.02623188,
      "epoch": 0.18685745205314735,
      "flos": 19718370115200.0,
      "grad_norm": 3.319162313919606,
      "language_loss": 0.73185605,
      "learning_rate": 3.7475457088743658e-06,
      "loss": 0.75433695,
      "num_input_tokens_seen": 32992640,
      "step": 1554,
      "time_per_iteration": 3.6521413326263428
    },
    {
      "auxiliary_loss_clip": 0.01199313,
      "auxiliary_loss_mlp": 0.01042112,
      "balance_loss_clip": 1.06873202,
      "balance_loss_mlp": 1.03140068,
      "epoch": 0.18697769494378644,
      "flos": 34204123589760.0,
      "grad_norm": 9.686991638267482,
      "language_loss": 0.74719656,
      "learning_rate": 3.7471667357445348e-06,
      "loss": 0.76961076,
      "num_input_tokens_seen": 33012470,
      "step": 1555,
      "time_per_iteration": 2.799145460128784
    },
    {
      "auxiliary_loss_clip": 0.01198069,
      "auxiliary_loss_mlp": 0.01047256,
      "balance_loss_clip": 1.06744456,
      "balance_loss_mlp": 1.03610396,
      "epoch": 0.18709793783442555,
      "flos": 34241327101440.0,
      "grad_norm": 3.446434914293949,
      "language_loss": 0.7254625,
      "learning_rate": 3.7467874975709e-06,
      "loss": 0.74791574,
      "num_input_tokens_seen": 33033275,
      "step": 1556,
      "time_per_iteration": 2.884984254837036
    },
    {
      "auxiliary_loss_clip": 0.01233497,
      "auxiliary_loss_mlp": 0.0104502,
      "balance_loss_clip": 1.07227778,
      "balance_loss_mlp": 1.03448832,
      "epoch": 0.18721818072506463,
      "flos": 40734550529280.0,
      "grad_norm": 2.835465235601631,
      "language_loss": 0.78019649,
      "learning_rate": 3.7464079944109904e-06,
      "loss": 0.80298167,
      "num_input_tokens_seen": 33055135,
      "step": 1557,
      "time_per_iteration": 2.8196377754211426
    },
    {
      "auxiliary_loss_clip": 0.01204803,
      "auxiliary_loss_mlp": 0.01040644,
      "balance_loss_clip": 1.06179452,
      "balance_loss_mlp": 1.03009391,
      "epoch": 0.18733842361570371,
      "flos": 22157386392960.0,
      "grad_norm": 2.1882423376876328,
      "language_loss": 0.77332836,
      "learning_rate": 3.746028226322376e-06,
      "loss": 0.7957828,
      "num_input_tokens_seen": 33071015,
      "step": 1558,
      "time_per_iteration": 2.7185258865356445
    },
    {
      "auxiliary_loss_clip": 0.01229107,
      "auxiliary_loss_mlp": 0.0103509,
      "balance_loss_clip": 1.07001626,
      "balance_loss_mlp": 1.02542853,
      "epoch": 0.18745866650634282,
      "flos": 18914940656640.0,
      "grad_norm": 3.807969539648895,
      "language_loss": 0.75426203,
      "learning_rate": 3.745648193362669e-06,
      "loss": 0.77690399,
      "num_input_tokens_seen": 33090370,
      "step": 1559,
      "time_per_iteration": 2.717054843902588
    },
    {
      "auxiliary_loss_clip": 0.01186067,
      "auxiliary_loss_mlp": 0.01031915,
      "balance_loss_clip": 1.06133115,
      "balance_loss_mlp": 1.0216397,
      "epoch": 0.1875789093969819,
      "flos": 19314626267520.0,
      "grad_norm": 2.9121846651405003,
      "language_loss": 0.72551167,
      "learning_rate": 3.745267895589518e-06,
      "loss": 0.74769145,
      "num_input_tokens_seen": 33108910,
      "step": 1560,
      "time_per_iteration": 2.736285924911499
    },
    {
      "auxiliary_loss_clip": 0.01203186,
      "auxiliary_loss_mlp": 0.01036843,
      "balance_loss_clip": 1.06766486,
      "balance_loss_mlp": 1.02570868,
      "epoch": 0.187699152287621,
      "flos": 17018965169280.0,
      "grad_norm": 2.9934233226502003,
      "language_loss": 0.82313728,
      "learning_rate": 3.7448873330606154e-06,
      "loss": 0.84553754,
      "num_input_tokens_seen": 33126680,
      "step": 1561,
      "time_per_iteration": 2.733468770980835
    },
    {
      "auxiliary_loss_clip": 0.01206761,
      "auxiliary_loss_mlp": 0.01035655,
      "balance_loss_clip": 1.06630075,
      "balance_loss_mlp": 1.02544498,
      "epoch": 0.18781939517826007,
      "flos": 22346384780160.0,
      "grad_norm": 2.46808752805559,
      "language_loss": 0.87515295,
      "learning_rate": 3.7445065058336914e-06,
      "loss": 0.89757717,
      "num_input_tokens_seen": 33145550,
      "step": 1562,
      "time_per_iteration": 2.7025508880615234
    },
    {
      "auxiliary_loss_clip": 0.01226936,
      "auxiliary_loss_mlp": 0.01039889,
      "balance_loss_clip": 1.06713247,
      "balance_loss_mlp": 1.0300957,
      "epoch": 0.18793963806889918,
      "flos": 14611478054400.0,
      "grad_norm": 2.0744629126072467,
      "language_loss": 0.86521703,
      "learning_rate": 3.7441254139665176e-06,
      "loss": 0.88788527,
      "num_input_tokens_seen": 33161735,
      "step": 1563,
      "time_per_iteration": 2.6679444313049316
    },
    {
      "auxiliary_loss_clip": 0.01203897,
      "auxiliary_loss_mlp": 0.01042732,
      "balance_loss_clip": 1.07039857,
      "balance_loss_mlp": 1.03283143,
      "epoch": 0.18805988095953827,
      "flos": 17457075354240.0,
      "grad_norm": 1.9309562239241602,
      "language_loss": 0.82576096,
      "learning_rate": 3.743744057516905e-06,
      "loss": 0.84822726,
      "num_input_tokens_seen": 33179795,
      "step": 1564,
      "time_per_iteration": 2.6663074493408203
    },
    {
      "auxiliary_loss_clip": 0.01201798,
      "auxiliary_loss_mlp": 0.01045017,
      "balance_loss_clip": 1.06632721,
      "balance_loss_mlp": 1.03464544,
      "epoch": 0.18818012385017735,
      "flos": 15043877976960.0,
      "grad_norm": 5.328630566636633,
      "language_loss": 0.8771168,
      "learning_rate": 3.743362436542706e-06,
      "loss": 0.89958489,
      "num_input_tokens_seen": 33194485,
      "step": 1565,
      "time_per_iteration": 2.7098875045776367
    },
    {
      "auxiliary_loss_clip": 0.01218674,
      "auxiliary_loss_mlp": 0.01039291,
      "balance_loss_clip": 1.06706572,
      "balance_loss_mlp": 1.02935481,
      "epoch": 0.18830036674081646,
      "flos": 47551975136640.0,
      "grad_norm": 2.6280070882875908,
      "language_loss": 0.76737505,
      "learning_rate": 3.7429805511018115e-06,
      "loss": 0.78995466,
      "num_input_tokens_seen": 33216145,
      "step": 1566,
      "time_per_iteration": 2.883535385131836
    },
    {
      "auxiliary_loss_clip": 0.01128963,
      "auxiliary_loss_mlp": 0.01041361,
      "balance_loss_clip": 1.05340362,
      "balance_loss_mlp": 1.03054881,
      "epoch": 0.18842060963145554,
      "flos": 30044626698240.0,
      "grad_norm": 4.134239307150833,
      "language_loss": 0.78144634,
      "learning_rate": 3.7425984012521524e-06,
      "loss": 0.80314958,
      "num_input_tokens_seen": 33236345,
      "step": 1567,
      "time_per_iteration": 3.0614733695983887
    },
    {
      "auxiliary_loss_clip": 0.01122879,
      "auxiliary_loss_mlp": 0.01011469,
      "balance_loss_clip": 1.02830029,
      "balance_loss_mlp": 1.00914431,
      "epoch": 0.18854085252209463,
      "flos": 70318372625280.0,
      "grad_norm": 0.7632162953833662,
      "language_loss": 0.60456979,
      "learning_rate": 3.7422159870517025e-06,
      "loss": 0.62591326,
      "num_input_tokens_seen": 33301600,
      "step": 1568,
      "time_per_iteration": 3.5153648853302
    },
    {
      "auxiliary_loss_clip": 0.01212846,
      "auxiliary_loss_mlp": 0.01038953,
      "balance_loss_clip": 1.07023501,
      "balance_loss_mlp": 1.02791429,
      "epoch": 0.1886610954127337,
      "flos": 21289318410240.0,
      "grad_norm": 1.8757462309634017,
      "language_loss": 0.78943712,
      "learning_rate": 3.7418333085584717e-06,
      "loss": 0.81195509,
      "num_input_tokens_seen": 33322785,
      "step": 1569,
      "time_per_iteration": 2.728398323059082
    },
    {
      "auxiliary_loss_clip": 0.01159469,
      "auxiliary_loss_mlp": 0.01043318,
      "balance_loss_clip": 1.05844855,
      "balance_loss_mlp": 1.03282189,
      "epoch": 0.18878133830337282,
      "flos": 17266819991040.0,
      "grad_norm": 19.655204002761078,
      "language_loss": 0.90432823,
      "learning_rate": 3.7414503658305128e-06,
      "loss": 0.92635608,
      "num_input_tokens_seen": 33340020,
      "step": 1570,
      "time_per_iteration": 2.8661108016967773
    },
    {
      "auxiliary_loss_clip": 0.01189561,
      "auxiliary_loss_mlp": 0.0103581,
      "balance_loss_clip": 1.06562161,
      "balance_loss_mlp": 1.02542663,
      "epoch": 0.1889015811940119,
      "flos": 25775207210880.0,
      "grad_norm": 3.0596170441860027,
      "language_loss": 0.77651596,
      "learning_rate": 3.7410671589259185e-06,
      "loss": 0.79876971,
      "num_input_tokens_seen": 33358620,
      "step": 1571,
      "time_per_iteration": 3.7382380962371826
    },
    {
      "auxiliary_loss_clip": 0.01128237,
      "auxiliary_loss_mlp": 0.00764194,
      "balance_loss_clip": 1.05361354,
      "balance_loss_mlp": 1.00055289,
      "epoch": 0.18902182408465099,
      "flos": 21032197879680.0,
      "grad_norm": 1.9980106214923417,
      "language_loss": 0.79910493,
      "learning_rate": 3.7406836879028205e-06,
      "loss": 0.81802922,
      "num_input_tokens_seen": 33378845,
      "step": 1572,
      "time_per_iteration": 2.8440375328063965
    },
    {
      "auxiliary_loss_clip": 0.01183143,
      "auxiliary_loss_mlp": 0.01038554,
      "balance_loss_clip": 1.06304622,
      "balance_loss_mlp": 1.02852297,
      "epoch": 0.1891420669752901,
      "flos": 22272121411200.0,
      "grad_norm": 10.613169819170427,
      "language_loss": 0.76324397,
      "learning_rate": 3.7402999528193907e-06,
      "loss": 0.78546095,
      "num_input_tokens_seen": 33398345,
      "step": 1573,
      "time_per_iteration": 2.7848212718963623
    },
    {
      "auxiliary_loss_clip": 0.01188647,
      "auxiliary_loss_mlp": 0.01041162,
      "balance_loss_clip": 1.0621419,
      "balance_loss_mlp": 1.03057599,
      "epoch": 0.18926230986592918,
      "flos": 22017802141440.0,
      "grad_norm": 2.660248602318878,
      "language_loss": 0.85289264,
      "learning_rate": 3.739915953733842e-06,
      "loss": 0.87519068,
      "num_input_tokens_seen": 33416390,
      "step": 1574,
      "time_per_iteration": 3.956892967224121
    },
    {
      "auxiliary_loss_clip": 0.01200028,
      "auxiliary_loss_mlp": 0.00764544,
      "balance_loss_clip": 1.06547177,
      "balance_loss_mlp": 1.00055075,
      "epoch": 0.18938255275656826,
      "flos": 24462672336000.0,
      "grad_norm": 2.299484889199137,
      "language_loss": 0.81863225,
      "learning_rate": 3.7395316907044264e-06,
      "loss": 0.83827806,
      "num_input_tokens_seen": 33437175,
      "step": 1575,
      "time_per_iteration": 2.768782377243042
    },
    {
      "auxiliary_loss_clip": 0.01212534,
      "auxiliary_loss_mlp": 0.01035785,
      "balance_loss_clip": 1.06242228,
      "balance_loss_mlp": 1.02466249,
      "epoch": 0.18950279564720737,
      "flos": 24427049022720.0,
      "grad_norm": 2.4105522750505495,
      "language_loss": 0.79611266,
      "learning_rate": 3.7391471637894364e-06,
      "loss": 0.81859583,
      "num_input_tokens_seen": 33459440,
      "step": 1576,
      "time_per_iteration": 3.650709629058838
    },
    {
      "auxiliary_loss_clip": 0.01212607,
      "auxiliary_loss_mlp": 0.00764688,
      "balance_loss_clip": 1.06712937,
      "balance_loss_mlp": 1.00071824,
      "epoch": 0.18962303853784646,
      "flos": 19756291898880.0,
      "grad_norm": 1.973617137241623,
      "language_loss": 0.84758091,
      "learning_rate": 3.738762373047205e-06,
      "loss": 0.86735386,
      "num_input_tokens_seen": 33479360,
      "step": 1577,
      "time_per_iteration": 2.7091174125671387
    },
    {
      "auxiliary_loss_clip": 0.01189706,
      "auxiliary_loss_mlp": 0.01042458,
      "balance_loss_clip": 1.06302154,
      "balance_loss_mlp": 1.03103137,
      "epoch": 0.18974328142848554,
      "flos": 21032054225280.0,
      "grad_norm": 2.000685680486301,
      "language_loss": 0.83684623,
      "learning_rate": 3.738377318536103e-06,
      "loss": 0.85916787,
      "num_input_tokens_seen": 33499245,
      "step": 1578,
      "time_per_iteration": 2.729329824447632
    },
    {
      "auxiliary_loss_clip": 0.01103659,
      "auxiliary_loss_mlp": 0.01041127,
      "balance_loss_clip": 1.04907966,
      "balance_loss_mlp": 1.03173292,
      "epoch": 0.18986352431912462,
      "flos": 12966122736000.0,
      "grad_norm": 2.600533967849279,
      "language_loss": 0.71587002,
      "learning_rate": 3.7379920003145447e-06,
      "loss": 0.73731786,
      "num_input_tokens_seen": 33513520,
      "step": 1579,
      "time_per_iteration": 3.8344929218292236
    },
    {
      "auxiliary_loss_clip": 0.01237457,
      "auxiliary_loss_mlp": 0.01045442,
      "balance_loss_clip": 1.07335079,
      "balance_loss_mlp": 1.03424263,
      "epoch": 0.18998376720976373,
      "flos": 23767908497280.0,
      "grad_norm": 6.680269981126659,
      "language_loss": 0.83614707,
      "learning_rate": 3.7376064184409817e-06,
      "loss": 0.85897601,
      "num_input_tokens_seen": 33533100,
      "step": 1580,
      "time_per_iteration": 2.6280570030212402
    },
    {
      "auxiliary_loss_clip": 0.01190332,
      "auxiliary_loss_mlp": 0.01042647,
      "balance_loss_clip": 1.06876159,
      "balance_loss_mlp": 1.03204393,
      "epoch": 0.19010401010040281,
      "flos": 22966023323520.0,
      "grad_norm": 1.7775477590801856,
      "language_loss": 0.86875486,
      "learning_rate": 3.7372205729739063e-06,
      "loss": 0.89108467,
      "num_input_tokens_seen": 33554915,
      "step": 1581,
      "time_per_iteration": 2.8114376068115234
    },
    {
      "auxiliary_loss_clip": 0.01218019,
      "auxiliary_loss_mlp": 0.01042184,
      "balance_loss_clip": 1.0719974,
      "balance_loss_mlp": 1.03138924,
      "epoch": 0.1902242529910419,
      "flos": 19135647774720.0,
      "grad_norm": 2.9070860688220304,
      "language_loss": 0.71890414,
      "learning_rate": 3.7368344639718514e-06,
      "loss": 0.74150622,
      "num_input_tokens_seen": 33572850,
      "step": 1582,
      "time_per_iteration": 2.677760124206543
    },
    {
      "auxiliary_loss_clip": 0.01199663,
      "auxiliary_loss_mlp": 0.00764138,
      "balance_loss_clip": 1.06695509,
      "balance_loss_mlp": 1.00079191,
      "epoch": 0.190344495881681,
      "flos": 25483935824640.0,
      "grad_norm": 1.9503950082399892,
      "language_loss": 0.80535614,
      "learning_rate": 3.7364480914933895e-06,
      "loss": 0.82499421,
      "num_input_tokens_seen": 33593090,
      "step": 1583,
      "time_per_iteration": 2.784029722213745
    },
    {
      "auxiliary_loss_clip": 0.01137746,
      "auxiliary_loss_mlp": 0.01036944,
      "balance_loss_clip": 1.05398762,
      "balance_loss_mlp": 1.02582777,
      "epoch": 0.1904647387723201,
      "flos": 26792843425920.0,
      "grad_norm": 3.6382096076523545,
      "language_loss": 0.81563199,
      "learning_rate": 3.7360614555971325e-06,
      "loss": 0.83737892,
      "num_input_tokens_seen": 33612745,
      "step": 1584,
      "time_per_iteration": 2.8154819011688232
    },
    {
      "auxiliary_loss_clip": 0.01219632,
      "auxiliary_loss_mlp": 0.01036688,
      "balance_loss_clip": 1.06951737,
      "balance_loss_mlp": 1.02608991,
      "epoch": 0.19058498166295917,
      "flos": 23987753688960.0,
      "grad_norm": 6.2274443454958215,
      "language_loss": 0.85100597,
      "learning_rate": 3.735674556341733e-06,
      "loss": 0.87356913,
      "num_input_tokens_seen": 33632360,
      "step": 1585,
      "time_per_iteration": 2.7161545753479004
    },
    {
      "auxiliary_loss_clip": 0.01172262,
      "auxiliary_loss_mlp": 0.01049907,
      "balance_loss_clip": 1.05949938,
      "balance_loss_mlp": 1.03932679,
      "epoch": 0.19070522455359826,
      "flos": 28293299280000.0,
      "grad_norm": 2.920689971321681,
      "language_loss": 0.82733244,
      "learning_rate": 3.7352873937858835e-06,
      "loss": 0.84955406,
      "num_input_tokens_seen": 33653895,
      "step": 1586,
      "time_per_iteration": 2.8398115634918213
    },
    {
      "auxiliary_loss_clip": 0.01217229,
      "auxiliary_loss_mlp": 0.01039892,
      "balance_loss_clip": 1.06738544,
      "balance_loss_mlp": 1.02955008,
      "epoch": 0.19082546744423737,
      "flos": 25660220797440.0,
      "grad_norm": 4.488736040018577,
      "language_loss": 0.72521466,
      "learning_rate": 3.734899967988316e-06,
      "loss": 0.74778593,
      "num_input_tokens_seen": 33672075,
      "step": 1587,
      "time_per_iteration": 2.6700735092163086
    },
    {
      "auxiliary_loss_clip": 0.01213831,
      "auxiliary_loss_mlp": 0.01034833,
      "balance_loss_clip": 1.06593716,
      "balance_loss_mlp": 1.02499795,
      "epoch": 0.19094571033487645,
      "flos": 19719483436800.0,
      "grad_norm": 2.8030314679040798,
      "language_loss": 0.83756375,
      "learning_rate": 3.7345122790078026e-06,
      "loss": 0.86005044,
      "num_input_tokens_seen": 33689640,
      "step": 1588,
      "time_per_iteration": 2.69130802154541
    },
    {
      "auxiliary_loss_clip": 0.01197302,
      "auxiliary_loss_mlp": 0.01035293,
      "balance_loss_clip": 1.06351757,
      "balance_loss_mlp": 1.02436757,
      "epoch": 0.19106595322551553,
      "flos": 21616320850560.0,
      "grad_norm": 4.645751087451185,
      "language_loss": 0.92927825,
      "learning_rate": 3.7341243269031556e-06,
      "loss": 0.95160419,
      "num_input_tokens_seen": 33708630,
      "step": 1589,
      "time_per_iteration": 2.6927311420440674
    },
    {
      "auxiliary_loss_clip": 0.01214198,
      "auxiliary_loss_mlp": 0.01035077,
      "balance_loss_clip": 1.0713594,
      "balance_loss_mlp": 1.02500343,
      "epoch": 0.19118619611615464,
      "flos": 29896890059520.0,
      "grad_norm": 6.078904344088924,
      "language_loss": 0.77575076,
      "learning_rate": 3.7337361117332275e-06,
      "loss": 0.79824352,
      "num_input_tokens_seen": 33730370,
      "step": 1590,
      "time_per_iteration": 2.8374009132385254
    },
    {
      "auxiliary_loss_clip": 0.01213357,
      "auxiliary_loss_mlp": 0.0076449,
      "balance_loss_clip": 1.06998491,
      "balance_loss_mlp": 1.00079381,
      "epoch": 0.19130643900679373,
      "flos": 17273428093440.0,
      "grad_norm": 2.810660103279861,
      "language_loss": 0.76890016,
      "learning_rate": 3.7333476335569087e-06,
      "loss": 0.78867865,
      "num_input_tokens_seen": 33748370,
      "step": 1591,
      "time_per_iteration": 2.6794497966766357
    },
    {
      "auxiliary_loss_clip": 0.01171005,
      "auxiliary_loss_mlp": 0.01034907,
      "balance_loss_clip": 1.06605029,
      "balance_loss_mlp": 1.0238328,
      "epoch": 0.1914266818974328,
      "flos": 24826339584000.0,
      "grad_norm": 3.893789756332813,
      "language_loss": 0.66715229,
      "learning_rate": 3.7329588924331325e-06,
      "loss": 0.68921149,
      "num_input_tokens_seen": 33769575,
      "step": 1592,
      "time_per_iteration": 2.7934978008270264
    },
    {
      "auxiliary_loss_clip": 0.01143055,
      "auxiliary_loss_mlp": 0.01037902,
      "balance_loss_clip": 1.05493736,
      "balance_loss_mlp": 1.02672064,
      "epoch": 0.1915469247880719,
      "flos": 18952467390720.0,
      "grad_norm": 2.0152773142874194,
      "language_loss": 0.82326239,
      "learning_rate": 3.732569888420871e-06,
      "loss": 0.84507191,
      "num_input_tokens_seen": 33789110,
      "step": 1593,
      "time_per_iteration": 2.795663833618164
    },
    {
      "auxiliary_loss_clip": 0.01154387,
      "auxiliary_loss_mlp": 0.01036112,
      "balance_loss_clip": 1.05607247,
      "balance_loss_mlp": 1.02624154,
      "epoch": 0.191667167678711,
      "flos": 21032952065280.0,
      "grad_norm": 3.4210283599744598,
      "language_loss": 0.8335644,
      "learning_rate": 3.732180621579134e-06,
      "loss": 0.85546935,
      "num_input_tokens_seen": 33808325,
      "step": 1594,
      "time_per_iteration": 2.7783772945404053
    },
    {
      "auxiliary_loss_clip": 0.01171454,
      "auxiliary_loss_mlp": 0.01040227,
      "balance_loss_clip": 1.05990911,
      "balance_loss_mlp": 1.02971315,
      "epoch": 0.1917874105693501,
      "flos": 34237663914240.0,
      "grad_norm": 2.217370056577203,
      "language_loss": 0.81018126,
      "learning_rate": 3.7317910919669745e-06,
      "loss": 0.83229804,
      "num_input_tokens_seen": 33829520,
      "step": 1595,
      "time_per_iteration": 2.859001874923706
    },
    {
      "auxiliary_loss_clip": 0.01159017,
      "auxiliary_loss_mlp": 0.01039874,
      "balance_loss_clip": 1.06184626,
      "balance_loss_mlp": 1.02958047,
      "epoch": 0.19190765345998917,
      "flos": 23550613171200.0,
      "grad_norm": 2.3784722877008937,
      "language_loss": 0.76177251,
      "learning_rate": 3.7314012996434826e-06,
      "loss": 0.7837615,
      "num_input_tokens_seen": 33848250,
      "step": 1596,
      "time_per_iteration": 2.8309946060180664
    },
    {
      "auxiliary_loss_clip": 0.01212997,
      "auxiliary_loss_mlp": 0.01039562,
      "balance_loss_clip": 1.06442475,
      "balance_loss_mlp": 1.0295608,
      "epoch": 0.19202789635062828,
      "flos": 19861330245120.0,
      "grad_norm": 3.174461494590396,
      "language_loss": 0.81401443,
      "learning_rate": 3.7310112446677907e-06,
      "loss": 0.8365401,
      "num_input_tokens_seen": 33866160,
      "step": 1597,
      "time_per_iteration": 3.551875114440918
    },
    {
      "auxiliary_loss_clip": 0.01154215,
      "auxiliary_loss_mlp": 0.01040407,
      "balance_loss_clip": 1.06416106,
      "balance_loss_mlp": 1.02921319,
      "epoch": 0.19214813924126736,
      "flos": 20922957642240.0,
      "grad_norm": 3.879509151364046,
      "language_loss": 0.69397378,
      "learning_rate": 3.7306209270990695e-06,
      "loss": 0.71591997,
      "num_input_tokens_seen": 33884165,
      "step": 1598,
      "time_per_iteration": 2.813650131225586
    },
    {
      "auxiliary_loss_clip": 0.01188306,
      "auxiliary_loss_mlp": 0.01045752,
      "balance_loss_clip": 1.06291497,
      "balance_loss_mlp": 1.03603601,
      "epoch": 0.19226838213190645,
      "flos": 26359725231360.0,
      "grad_norm": 2.2121209727760007,
      "language_loss": 0.86436892,
      "learning_rate": 3.7302303469965292e-06,
      "loss": 0.88670945,
      "num_input_tokens_seen": 33903705,
      "step": 1599,
      "time_per_iteration": 2.9302289485931396
    },
    {
      "auxiliary_loss_clip": 0.01185861,
      "auxiliary_loss_mlp": 0.01040739,
      "balance_loss_clip": 1.06387138,
      "balance_loss_mlp": 1.03035593,
      "epoch": 0.19238862502254553,
      "flos": 20850525866880.0,
      "grad_norm": 20.066791034641827,
      "language_loss": 0.70962656,
      "learning_rate": 3.7298395044194206e-06,
      "loss": 0.73189259,
      "num_input_tokens_seen": 33922515,
      "step": 1600,
      "time_per_iteration": 3.7142326831817627
    },
    {
      "auxiliary_loss_clip": 0.01154182,
      "auxiliary_loss_mlp": 0.01037431,
      "balance_loss_clip": 1.05914068,
      "balance_loss_mlp": 1.02713692,
      "epoch": 0.19250886791318464,
      "flos": 21726063878400.0,
      "grad_norm": 2.0020651293819647,
      "language_loss": 0.94472915,
      "learning_rate": 3.7294483994270356e-06,
      "loss": 0.9666453,
      "num_input_tokens_seen": 33940840,
      "step": 1601,
      "time_per_iteration": 2.765004873275757
    },
    {
      "auxiliary_loss_clip": 0.0117049,
      "auxiliary_loss_mlp": 0.01042643,
      "balance_loss_clip": 1.06049502,
      "balance_loss_mlp": 1.03243303,
      "epoch": 0.19262911080382372,
      "flos": 23367827836800.0,
      "grad_norm": 3.0713720812151157,
      "language_loss": 0.77645802,
      "learning_rate": 3.7290570320787033e-06,
      "loss": 0.79858935,
      "num_input_tokens_seen": 33960420,
      "step": 1602,
      "time_per_iteration": 3.6492292881011963
    },
    {
      "auxiliary_loss_clip": 0.01207959,
      "auxiliary_loss_mlp": 0.01036125,
      "balance_loss_clip": 1.06512177,
      "balance_loss_mlp": 1.02549791,
      "epoch": 0.1927493536944628,
      "flos": 21943502858880.0,
      "grad_norm": 2.5000474791512226,
      "language_loss": 0.71525776,
      "learning_rate": 3.728665402433793e-06,
      "loss": 0.73769855,
      "num_input_tokens_seen": 33978990,
      "step": 1603,
      "time_per_iteration": 2.6863365173339844
    },
    {
      "auxiliary_loss_clip": 0.01188485,
      "auxiliary_loss_mlp": 0.01035659,
      "balance_loss_clip": 1.06467903,
      "balance_loss_mlp": 1.02579427,
      "epoch": 0.19286959658510192,
      "flos": 16545590807040.0,
      "grad_norm": 2.724151490274102,
      "language_loss": 0.85400468,
      "learning_rate": 3.7282735105517164e-06,
      "loss": 0.87624621,
      "num_input_tokens_seen": 33997115,
      "step": 1604,
      "time_per_iteration": 2.686596632003784
    },
    {
      "auxiliary_loss_clip": 0.01149977,
      "auxiliary_loss_mlp": 0.01038169,
      "balance_loss_clip": 1.05760562,
      "balance_loss_mlp": 1.02702856,
      "epoch": 0.192989839475741,
      "flos": 21616967295360.0,
      "grad_norm": 4.000874439856232,
      "language_loss": 0.67947972,
      "learning_rate": 3.727881356491922e-06,
      "loss": 0.70136112,
      "num_input_tokens_seen": 34015525,
      "step": 1605,
      "time_per_iteration": 3.825294256210327
    },
    {
      "auxiliary_loss_clip": 0.01166857,
      "auxiliary_loss_mlp": 0.01041368,
      "balance_loss_clip": 1.05827308,
      "balance_loss_mlp": 1.03162885,
      "epoch": 0.19311008236638008,
      "flos": 19281516906240.0,
      "grad_norm": 2.1864471671415906,
      "language_loss": 0.75910473,
      "learning_rate": 3.7274889403139002e-06,
      "loss": 0.78118694,
      "num_input_tokens_seen": 34033150,
      "step": 1606,
      "time_per_iteration": 2.718564748764038
    },
    {
      "auxiliary_loss_clip": 0.0119931,
      "auxiliary_loss_mlp": 0.01027812,
      "balance_loss_clip": 1.06297529,
      "balance_loss_mlp": 1.01802444,
      "epoch": 0.1932303252570192,
      "flos": 28652369587200.0,
      "grad_norm": 5.677607818734286,
      "language_loss": 0.78651536,
      "learning_rate": 3.727096262077179e-06,
      "loss": 0.80878663,
      "num_input_tokens_seen": 34052145,
      "step": 1607,
      "time_per_iteration": 2.7198143005371094
    },
    {
      "auxiliary_loss_clip": 0.01164816,
      "auxiliary_loss_mlp": 0.0103855,
      "balance_loss_clip": 1.05746174,
      "balance_loss_mlp": 1.02900743,
      "epoch": 0.19335056814765827,
      "flos": 18368990864640.0,
      "grad_norm": 1.853999941343885,
      "language_loss": 0.84911823,
      "learning_rate": 3.7267033218413285e-06,
      "loss": 0.87115186,
      "num_input_tokens_seen": 34069940,
      "step": 1608,
      "time_per_iteration": 2.6469805240631104
    },
    {
      "auxiliary_loss_clip": 0.01157304,
      "auxiliary_loss_mlp": 0.01032107,
      "balance_loss_clip": 1.05840909,
      "balance_loss_mlp": 1.0213306,
      "epoch": 0.19347081103829736,
      "flos": 13260877741440.0,
      "grad_norm": 2.3652596127885936,
      "language_loss": 0.8140372,
      "learning_rate": 3.726310119665957e-06,
      "loss": 0.8359313,
      "num_input_tokens_seen": 34086275,
      "step": 1609,
      "time_per_iteration": 2.7312533855438232
    },
    {
      "auxiliary_loss_clip": 0.01197716,
      "auxiliary_loss_mlp": 0.01035378,
      "balance_loss_clip": 1.06351721,
      "balance_loss_mlp": 1.02474427,
      "epoch": 0.19359105392893644,
      "flos": 20300122788480.0,
      "grad_norm": 2.152497070982658,
      "language_loss": 0.85201502,
      "learning_rate": 3.725916655610713e-06,
      "loss": 0.8743459,
      "num_input_tokens_seen": 34105605,
      "step": 1610,
      "time_per_iteration": 2.769015312194824
    },
    {
      "auxiliary_loss_clip": 0.01216718,
      "auxiliary_loss_mlp": 0.0103911,
      "balance_loss_clip": 1.06413245,
      "balance_loss_mlp": 1.02835774,
      "epoch": 0.19371129681957555,
      "flos": 20484596062080.0,
      "grad_norm": 3.563617710094631,
      "language_loss": 0.75473428,
      "learning_rate": 3.725522929735284e-06,
      "loss": 0.77729261,
      "num_input_tokens_seen": 34122540,
      "step": 1611,
      "time_per_iteration": 2.6248741149902344
    },
    {
      "auxiliary_loss_clip": 0.0113274,
      "auxiliary_loss_mlp": 0.01032638,
      "balance_loss_clip": 1.0499258,
      "balance_loss_mlp": 1.02264822,
      "epoch": 0.19383153971021463,
      "flos": 30445497457920.0,
      "grad_norm": 4.973669967685178,
      "language_loss": 0.7466315,
      "learning_rate": 3.725128942099399e-06,
      "loss": 0.76828533,
      "num_input_tokens_seen": 34142940,
      "step": 1612,
      "time_per_iteration": 2.9974257946014404
    },
    {
      "auxiliary_loss_clip": 0.01191453,
      "auxiliary_loss_mlp": 0.01030756,
      "balance_loss_clip": 1.06370425,
      "balance_loss_mlp": 1.02056372,
      "epoch": 0.19395178260085372,
      "flos": 24569937325440.0,
      "grad_norm": 2.165370752425255,
      "language_loss": 0.7990073,
      "learning_rate": 3.7247346927628245e-06,
      "loss": 0.82122946,
      "num_input_tokens_seen": 34162875,
      "step": 1613,
      "time_per_iteration": 2.7141153812408447
    },
    {
      "auxiliary_loss_clip": 0.01177377,
      "auxiliary_loss_mlp": 0.01038805,
      "balance_loss_clip": 1.06302762,
      "balance_loss_mlp": 1.02785611,
      "epoch": 0.19407202549149283,
      "flos": 28950608211840.0,
      "grad_norm": 7.651940977551442,
      "language_loss": 0.7922703,
      "learning_rate": 3.7243401817853694e-06,
      "loss": 0.81443214,
      "num_input_tokens_seen": 34183565,
      "step": 1614,
      "time_per_iteration": 2.8930468559265137
    },
    {
      "auxiliary_loss_clip": 0.01180416,
      "auxiliary_loss_mlp": 0.01035965,
      "balance_loss_clip": 1.05799007,
      "balance_loss_mlp": 1.02557039,
      "epoch": 0.1941922683821319,
      "flos": 18004497603840.0,
      "grad_norm": 2.7211800880798194,
      "language_loss": 0.71739519,
      "learning_rate": 3.723945409226879e-06,
      "loss": 0.73955899,
      "num_input_tokens_seen": 34202055,
      "step": 1615,
      "time_per_iteration": 2.72456431388855
    },
    {
      "auxiliary_loss_clip": 0.01216598,
      "auxiliary_loss_mlp": 0.01044133,
      "balance_loss_clip": 1.06582105,
      "balance_loss_mlp": 1.03420305,
      "epoch": 0.194312511272771,
      "flos": 9720337034880.0,
      "grad_norm": 3.1810142469635214,
      "language_loss": 0.79972249,
      "learning_rate": 3.723550375147241e-06,
      "loss": 0.82232988,
      "num_input_tokens_seen": 34216830,
      "step": 1616,
      "time_per_iteration": 2.6420938968658447
    },
    {
      "auxiliary_loss_clip": 0.01205045,
      "auxiliary_loss_mlp": 0.01037835,
      "balance_loss_clip": 1.06775534,
      "balance_loss_mlp": 1.0275768,
      "epoch": 0.19443275416341008,
      "flos": 27016208150400.0,
      "grad_norm": 2.4450349750248823,
      "language_loss": 0.80053115,
      "learning_rate": 3.7231550796063816e-06,
      "loss": 0.82295996,
      "num_input_tokens_seen": 34236840,
      "step": 1617,
      "time_per_iteration": 2.699620246887207
    },
    {
      "auxiliary_loss_clip": 0.01142195,
      "auxiliary_loss_mlp": 0.01050693,
      "balance_loss_clip": 1.05831671,
      "balance_loss_mlp": 1.03907585,
      "epoch": 0.1945529970540492,
      "flos": 15846625077120.0,
      "grad_norm": 1.935432057356931,
      "language_loss": 0.65054059,
      "learning_rate": 3.722759522664266e-06,
      "loss": 0.6724695,
      "num_input_tokens_seen": 34254140,
      "step": 1618,
      "time_per_iteration": 2.795779228210449
    },
    {
      "auxiliary_loss_clip": 0.01173714,
      "auxiliary_loss_mlp": 0.01036315,
      "balance_loss_clip": 1.06356812,
      "balance_loss_mlp": 1.02666521,
      "epoch": 0.19467323994468827,
      "flos": 19314985403520.0,
      "grad_norm": 5.000194610637167,
      "language_loss": 0.81814921,
      "learning_rate": 3.7223637043809016e-06,
      "loss": 0.84024948,
      "num_input_tokens_seen": 34273120,
      "step": 1619,
      "time_per_iteration": 2.760413646697998
    },
    {
      "auxiliary_loss_clip": 0.01165726,
      "auxiliary_loss_mlp": 0.01040469,
      "balance_loss_clip": 1.06511819,
      "balance_loss_mlp": 1.03042579,
      "epoch": 0.19479348283532735,
      "flos": 24133227770880.0,
      "grad_norm": 3.3722527004589957,
      "language_loss": 0.85988808,
      "learning_rate": 3.7219676248163322e-06,
      "loss": 0.88195002,
      "num_input_tokens_seen": 34290285,
      "step": 1620,
      "time_per_iteration": 2.7873167991638184
    },
    {
      "auxiliary_loss_clip": 0.01217771,
      "auxiliary_loss_mlp": 0.01036827,
      "balance_loss_clip": 1.06813419,
      "balance_loss_mlp": 1.02665234,
      "epoch": 0.19491372572596646,
      "flos": 25775638174080.0,
      "grad_norm": 2.5496611340370987,
      "language_loss": 0.93306071,
      "learning_rate": 3.721571284030643e-06,
      "loss": 0.9556067,
      "num_input_tokens_seen": 34310095,
      "step": 1621,
      "time_per_iteration": 2.757869243621826
    },
    {
      "auxiliary_loss_clip": 0.01151748,
      "auxiliary_loss_mlp": 0.01035745,
      "balance_loss_clip": 1.05365717,
      "balance_loss_mlp": 1.0262078,
      "epoch": 0.19503396861660555,
      "flos": 19645220067840.0,
      "grad_norm": 2.2377850666198125,
      "language_loss": 0.7886759,
      "learning_rate": 3.7211746820839587e-06,
      "loss": 0.81055087,
      "num_input_tokens_seen": 34327190,
      "step": 1622,
      "time_per_iteration": 2.7751758098602295
    },
    {
      "auxiliary_loss_clip": 0.01157966,
      "auxiliary_loss_mlp": 0.01036932,
      "balance_loss_clip": 1.0601846,
      "balance_loss_mlp": 1.02686512,
      "epoch": 0.19515421150724463,
      "flos": 21033023892480.0,
      "grad_norm": 4.266202147314309,
      "language_loss": 0.80517578,
      "learning_rate": 3.7207778190364437e-06,
      "loss": 0.82712471,
      "num_input_tokens_seen": 34345615,
      "step": 1623,
      "time_per_iteration": 3.7458837032318115
    },
    {
      "auxiliary_loss_clip": 0.01211506,
      "auxiliary_loss_mlp": 0.01037529,
      "balance_loss_clip": 1.0702796,
      "balance_loss_mlp": 1.027426,
      "epoch": 0.1952744543978837,
      "flos": 32961255143040.0,
      "grad_norm": 2.4877199529557985,
      "language_loss": 0.74029243,
      "learning_rate": 3.720380694948302e-06,
      "loss": 0.76278281,
      "num_input_tokens_seen": 34368500,
      "step": 1624,
      "time_per_iteration": 2.7665083408355713
    },
    {
      "auxiliary_loss_clip": 0.01078354,
      "auxiliary_loss_mlp": 0.01009238,
      "balance_loss_clip": 1.0280937,
      "balance_loss_mlp": 1.00687742,
      "epoch": 0.19539469728852282,
      "flos": 64044312030720.0,
      "grad_norm": 1.0462482227433396,
      "language_loss": 0.7114706,
      "learning_rate": 3.719983309879777e-06,
      "loss": 0.73234653,
      "num_input_tokens_seen": 34428280,
      "step": 1625,
      "time_per_iteration": 3.302661657333374
    },
    {
      "auxiliary_loss_clip": 0.01183999,
      "auxiliary_loss_mlp": 0.01034965,
      "balance_loss_clip": 1.06574881,
      "balance_loss_mlp": 1.02468872,
      "epoch": 0.1955149401791619,
      "flos": 13370908078080.0,
      "grad_norm": 1.9352701206649048,
      "language_loss": 0.77668893,
      "learning_rate": 3.719585663891151e-06,
      "loss": 0.79887861,
      "num_input_tokens_seen": 34445815,
      "step": 1626,
      "time_per_iteration": 3.7562167644500732
    },
    {
      "auxiliary_loss_clip": 0.01220038,
      "auxiliary_loss_mlp": 0.01042093,
      "balance_loss_clip": 1.06788445,
      "balance_loss_mlp": 1.03156722,
      "epoch": 0.195635183069801,
      "flos": 18728887184640.0,
      "grad_norm": 3.5800539971047023,
      "language_loss": 0.78992885,
      "learning_rate": 3.719187757042747e-06,
      "loss": 0.81255019,
      "num_input_tokens_seen": 34463635,
      "step": 1627,
      "time_per_iteration": 3.56905460357666
    },
    {
      "auxiliary_loss_clip": 0.01051073,
      "auxiliary_loss_mlp": 0.01004335,
      "balance_loss_clip": 1.02201998,
      "balance_loss_mlp": 1.00227249,
      "epoch": 0.1957554259604401,
      "flos": 69313952615040.0,
      "grad_norm": 0.7243280845627563,
      "language_loss": 0.54869699,
      "learning_rate": 3.7187895893949275e-06,
      "loss": 0.56925106,
      "num_input_tokens_seen": 34530105,
      "step": 1628,
      "time_per_iteration": 3.4516868591308594
    },
    {
      "auxiliary_loss_clip": 0.01210849,
      "auxiliary_loss_mlp": 0.01039707,
      "balance_loss_clip": 1.06597519,
      "balance_loss_mlp": 1.02920485,
      "epoch": 0.19587566885107918,
      "flos": 21069257736960.0,
      "grad_norm": 4.382664603328261,
      "language_loss": 0.75430393,
      "learning_rate": 3.7183911610080937e-06,
      "loss": 0.77680957,
      "num_input_tokens_seen": 34546970,
      "step": 1629,
      "time_per_iteration": 2.7010858058929443
    },
    {
      "auxiliary_loss_clip": 0.01166869,
      "auxiliary_loss_mlp": 0.01041501,
      "balance_loss_clip": 1.05586004,
      "balance_loss_mlp": 1.03100443,
      "epoch": 0.19599591174171827,
      "flos": 22194661731840.0,
      "grad_norm": 5.8382871111934005,
      "language_loss": 0.75349778,
      "learning_rate": 3.7179924719426872e-06,
      "loss": 0.77558148,
      "num_input_tokens_seen": 34564865,
      "step": 1630,
      "time_per_iteration": 2.725857734680176
    },
    {
      "auxiliary_loss_clip": 0.01215297,
      "auxiliary_loss_mlp": 0.01041075,
      "balance_loss_clip": 1.06808019,
      "balance_loss_mlp": 1.03123975,
      "epoch": 0.19611615463235738,
      "flos": 23768375374080.0,
      "grad_norm": 2.578120648699412,
      "language_loss": 0.76130056,
      "learning_rate": 3.7175935222591885e-06,
      "loss": 0.78386426,
      "num_input_tokens_seen": 34584165,
      "step": 1631,
      "time_per_iteration": 3.7346410751342773
    },
    {
      "auxiliary_loss_clip": 0.01177506,
      "auxiliary_loss_mlp": 0.00763846,
      "balance_loss_clip": 1.06454158,
      "balance_loss_mlp": 1.00125527,
      "epoch": 0.19623639752299646,
      "flos": 28618218731520.0,
      "grad_norm": 2.019534876531059,
      "language_loss": 0.7412194,
      "learning_rate": 3.717194312018118e-06,
      "loss": 0.76063287,
      "num_input_tokens_seen": 34603150,
      "step": 1632,
      "time_per_iteration": 2.793797016143799
    },
    {
      "auxiliary_loss_clip": 0.01225758,
      "auxiliary_loss_mlp": 0.01039895,
      "balance_loss_clip": 1.06607366,
      "balance_loss_mlp": 1.02993512,
      "epoch": 0.19635664041363554,
      "flos": 21032700670080.0,
      "grad_norm": 2.802266554806584,
      "language_loss": 0.76148522,
      "learning_rate": 3.716794841280036e-06,
      "loss": 0.78414178,
      "num_input_tokens_seen": 34621855,
      "step": 1633,
      "time_per_iteration": 2.6366240978240967
    },
    {
      "auxiliary_loss_clip": 0.0118596,
      "auxiliary_loss_mlp": 0.01038592,
      "balance_loss_clip": 1.06275272,
      "balance_loss_mlp": 1.02758312,
      "epoch": 0.19647688330427462,
      "flos": 18879748306560.0,
      "grad_norm": 2.6504258569379493,
      "language_loss": 0.77834743,
      "learning_rate": 3.7163951101055407e-06,
      "loss": 0.80059302,
      "num_input_tokens_seen": 34639915,
      "step": 1634,
      "time_per_iteration": 2.6852622032165527
    },
    {
      "auxiliary_loss_clip": 0.01160261,
      "auxiliary_loss_mlp": 0.0104941,
      "balance_loss_clip": 1.06358027,
      "balance_loss_mlp": 1.03847861,
      "epoch": 0.19659712619491373,
      "flos": 24242503921920.0,
      "grad_norm": 1.9007264514403959,
      "language_loss": 0.78919435,
      "learning_rate": 3.715995118555273e-06,
      "loss": 0.81129104,
      "num_input_tokens_seen": 34659890,
      "step": 1635,
      "time_per_iteration": 2.810671806335449
    },
    {
      "auxiliary_loss_clip": 0.01183757,
      "auxiliary_loss_mlp": 0.01042809,
      "balance_loss_clip": 1.06352866,
      "balance_loss_mlp": 1.03238428,
      "epoch": 0.19671736908555282,
      "flos": 24717422568960.0,
      "grad_norm": 2.581585570906675,
      "language_loss": 0.85998094,
      "learning_rate": 3.71559486668991e-06,
      "loss": 0.88224661,
      "num_input_tokens_seen": 34678750,
      "step": 1636,
      "time_per_iteration": 2.748655080795288
    },
    {
      "auxiliary_loss_clip": 0.01118114,
      "auxiliary_loss_mlp": 0.01042473,
      "balance_loss_clip": 1.05251503,
      "balance_loss_mlp": 1.03217304,
      "epoch": 0.1968376119761919,
      "flos": 23842279607040.0,
      "grad_norm": 2.3090671186722305,
      "language_loss": 0.77131343,
      "learning_rate": 3.715194354570169e-06,
      "loss": 0.79291934,
      "num_input_tokens_seen": 34698755,
      "step": 1637,
      "time_per_iteration": 2.8371026515960693
    },
    {
      "auxiliary_loss_clip": 0.01126051,
      "auxiliary_loss_mlp": 0.00763619,
      "balance_loss_clip": 1.05822444,
      "balance_loss_mlp": 1.00096321,
      "epoch": 0.196957854866831,
      "flos": 18113917409280.0,
      "grad_norm": 4.07983802231,
      "language_loss": 0.83520693,
      "learning_rate": 3.714793582256809e-06,
      "loss": 0.85410368,
      "num_input_tokens_seen": 34715820,
      "step": 1638,
      "time_per_iteration": 2.809871196746826
    },
    {
      "auxiliary_loss_clip": 0.01181043,
      "auxiliary_loss_mlp": 0.01050504,
      "balance_loss_clip": 1.06191051,
      "balance_loss_mlp": 1.0409019,
      "epoch": 0.1970780977574701,
      "flos": 21653129312640.0,
      "grad_norm": 2.8621249794457824,
      "language_loss": 0.84957558,
      "learning_rate": 3.7143925498106253e-06,
      "loss": 0.87189102,
      "num_input_tokens_seen": 34734360,
      "step": 1639,
      "time_per_iteration": 2.7583084106445312
    },
    {
      "auxiliary_loss_clip": 0.01208167,
      "auxiliary_loss_mlp": 0.01032938,
      "balance_loss_clip": 1.06353235,
      "balance_loss_mlp": 1.0231328,
      "epoch": 0.19719834064810918,
      "flos": 20811813984000.0,
      "grad_norm": 4.125362022299753,
      "language_loss": 0.79254925,
      "learning_rate": 3.7139912572924558e-06,
      "loss": 0.8149603,
      "num_input_tokens_seen": 34753390,
      "step": 1640,
      "time_per_iteration": 2.6902873516082764
    },
    {
      "auxiliary_loss_clip": 0.01200037,
      "auxiliary_loss_mlp": 0.01041738,
      "balance_loss_clip": 1.06396604,
      "balance_loss_mlp": 1.03134906,
      "epoch": 0.19731858353874826,
      "flos": 23434800744960.0,
      "grad_norm": 3.2406053229815135,
      "language_loss": 0.80637145,
      "learning_rate": 3.7135897047631744e-06,
      "loss": 0.82878917,
      "num_input_tokens_seen": 34771275,
      "step": 1641,
      "time_per_iteration": 2.754904270172119
    },
    {
      "auxiliary_loss_clip": 0.0119712,
      "auxiliary_loss_mlp": 0.01033837,
      "balance_loss_clip": 1.06135488,
      "balance_loss_mlp": 1.02392459,
      "epoch": 0.19743882642938737,
      "flos": 23988184652160.0,
      "grad_norm": 2.40672096247338,
      "language_loss": 0.76469523,
      "learning_rate": 3.713187892283698e-06,
      "loss": 0.78700477,
      "num_input_tokens_seen": 34790885,
      "step": 1642,
      "time_per_iteration": 2.7349131107330322
    },
    {
      "auxiliary_loss_clip": 0.01227348,
      "auxiliary_loss_mlp": 0.00763662,
      "balance_loss_clip": 1.0693388,
      "balance_loss_mlp": 1.00093162,
      "epoch": 0.19755906932002645,
      "flos": 15004340081280.0,
      "grad_norm": 3.6193854707706197,
      "language_loss": 0.87377059,
      "learning_rate": 3.71278581991498e-06,
      "loss": 0.89368069,
      "num_input_tokens_seen": 34806745,
      "step": 1643,
      "time_per_iteration": 2.6533992290496826
    },
    {
      "auxiliary_loss_clip": 0.01151858,
      "auxiliary_loss_mlp": 0.01039776,
      "balance_loss_clip": 1.05772281,
      "balance_loss_mlp": 1.03020978,
      "epoch": 0.19767931221066554,
      "flos": 19494466686720.0,
      "grad_norm": 8.665000883602849,
      "language_loss": 0.78815222,
      "learning_rate": 3.712383487718015e-06,
      "loss": 0.81006855,
      "num_input_tokens_seen": 34824985,
      "step": 1644,
      "time_per_iteration": 2.8189706802368164
    },
    {
      "auxiliary_loss_clip": 0.01207037,
      "auxiliary_loss_mlp": 0.01039358,
      "balance_loss_clip": 1.0692997,
      "balance_loss_mlp": 1.02920699,
      "epoch": 0.19779955510130465,
      "flos": 25737895958400.0,
      "grad_norm": 3.3216141816039353,
      "language_loss": 0.87102002,
      "learning_rate": 3.7119808957538365e-06,
      "loss": 0.89348394,
      "num_input_tokens_seen": 34843980,
      "step": 1645,
      "time_per_iteration": 2.7681427001953125
    },
    {
      "auxiliary_loss_clip": 0.01148552,
      "auxiliary_loss_mlp": 0.01041765,
      "balance_loss_clip": 1.05968046,
      "balance_loss_mlp": 1.03201342,
      "epoch": 0.19791979799194373,
      "flos": 20777699041920.0,
      "grad_norm": 3.1403883648495423,
      "language_loss": 0.79473627,
      "learning_rate": 3.711578044083517e-06,
      "loss": 0.81663942,
      "num_input_tokens_seen": 34860780,
      "step": 1646,
      "time_per_iteration": 2.7774789333343506
    },
    {
      "auxiliary_loss_clip": 0.01186695,
      "auxiliary_loss_mlp": 0.01036139,
      "balance_loss_clip": 1.06193876,
      "balance_loss_mlp": 1.02614319,
      "epoch": 0.1980400408825828,
      "flos": 25589010084480.0,
      "grad_norm": 2.651141242872653,
      "language_loss": 0.74096054,
      "learning_rate": 3.7111749327681698e-06,
      "loss": 0.76318884,
      "num_input_tokens_seen": 34880815,
      "step": 1647,
      "time_per_iteration": 2.7036097049713135
    },
    {
      "auxiliary_loss_clip": 0.01175338,
      "auxiliary_loss_mlp": 0.01035106,
      "balance_loss_clip": 1.06295609,
      "balance_loss_mlp": 1.0251044,
      "epoch": 0.1981602837732219,
      "flos": 23513840622720.0,
      "grad_norm": 2.3983268201065755,
      "language_loss": 0.86447328,
      "learning_rate": 3.7107715618689455e-06,
      "loss": 0.88657773,
      "num_input_tokens_seen": 34899790,
      "step": 1648,
      "time_per_iteration": 2.797194480895996
    },
    {
      "auxiliary_loss_clip": 0.01198057,
      "auxiliary_loss_mlp": 0.01038246,
      "balance_loss_clip": 1.06443667,
      "balance_loss_mlp": 1.02897167,
      "epoch": 0.198280526663861,
      "flos": 23185365724800.0,
      "grad_norm": 1.6482021079910205,
      "language_loss": 0.83544135,
      "learning_rate": 3.710367931447035e-06,
      "loss": 0.85780442,
      "num_input_tokens_seen": 34921570,
      "step": 1649,
      "time_per_iteration": 3.691058874130249
    },
    {
      "auxiliary_loss_clip": 0.01209327,
      "auxiliary_loss_mlp": 0.01043736,
      "balance_loss_clip": 1.0684166,
      "balance_loss_mlp": 1.03359747,
      "epoch": 0.1984007695545001,
      "flos": 21689470897920.0,
      "grad_norm": 3.1387135286614285,
      "language_loss": 0.8649804,
      "learning_rate": 3.70996404156367e-06,
      "loss": 0.88751101,
      "num_input_tokens_seen": 34941205,
      "step": 1650,
      "time_per_iteration": 2.715348482131958
    },
    {
      "auxiliary_loss_clip": 0.01199439,
      "auxiliary_loss_mlp": 0.01034444,
      "balance_loss_clip": 1.06500006,
      "balance_loss_mlp": 1.02453744,
      "epoch": 0.19852101244513917,
      "flos": 36064008887040.0,
      "grad_norm": 2.311685013969429,
      "language_loss": 0.73108715,
      "learning_rate": 3.7095598922801187e-06,
      "loss": 0.75342602,
      "num_input_tokens_seen": 34963280,
      "step": 1651,
      "time_per_iteration": 2.7897682189941406
    },
    {
      "auxiliary_loss_clip": 0.01173301,
      "auxiliary_loss_mlp": 0.01038913,
      "balance_loss_clip": 1.05880642,
      "balance_loss_mlp": 1.02946579,
      "epoch": 0.19864125533577828,
      "flos": 23105894883840.0,
      "grad_norm": 3.10997234612352,
      "language_loss": 0.76340854,
      "learning_rate": 3.7091554836576914e-06,
      "loss": 0.78553069,
      "num_input_tokens_seen": 34979955,
      "step": 1652,
      "time_per_iteration": 3.669168472290039
    },
    {
      "auxiliary_loss_clip": 0.01142072,
      "auxiliary_loss_mlp": 0.01040118,
      "balance_loss_clip": 1.05370784,
      "balance_loss_mlp": 1.03077841,
      "epoch": 0.19876149822641737,
      "flos": 24608505553920.0,
      "grad_norm": 2.0180714128116297,
      "language_loss": 0.82328165,
      "learning_rate": 3.708750815757736e-06,
      "loss": 0.84510356,
      "num_input_tokens_seen": 35000725,
      "step": 1653,
      "time_per_iteration": 3.7994468212127686
    },
    {
      "auxiliary_loss_clip": 0.01135239,
      "auxiliary_loss_mlp": 0.01040595,
      "balance_loss_clip": 1.05160725,
      "balance_loss_mlp": 1.03003287,
      "epoch": 0.19888174111705645,
      "flos": 32196645308160.0,
      "grad_norm": 3.679611450273783,
      "language_loss": 0.73001987,
      "learning_rate": 3.7083458886416407e-06,
      "loss": 0.75177824,
      "num_input_tokens_seen": 35019920,
      "step": 1654,
      "time_per_iteration": 2.8677492141723633
    },
    {
      "auxiliary_loss_clip": 0.01168218,
      "auxiliary_loss_mlp": 0.01037902,
      "balance_loss_clip": 1.05773735,
      "balance_loss_mlp": 1.02814484,
      "epoch": 0.19900198400769553,
      "flos": 24608469640320.0,
      "grad_norm": 6.282608582455174,
      "language_loss": 0.87920564,
      "learning_rate": 3.707940702370832e-06,
      "loss": 0.90126681,
      "num_input_tokens_seen": 35040765,
      "step": 1655,
      "time_per_iteration": 2.8345465660095215
    },
    {
      "auxiliary_loss_clip": 0.01087646,
      "auxiliary_loss_mlp": 0.00756512,
      "balance_loss_clip": 1.02660918,
      "balance_loss_mlp": 1.00210595,
      "epoch": 0.19912222689833464,
      "flos": 67915805673600.0,
      "grad_norm": 0.7771952781555648,
      "language_loss": 0.58183634,
      "learning_rate": 3.707535257006777e-06,
      "loss": 0.6002779,
      "num_input_tokens_seen": 35106390,
      "step": 1656,
      "time_per_iteration": 4.20514178276062
    },
    {
      "auxiliary_loss_clip": 0.01186217,
      "auxiliary_loss_mlp": 0.01036555,
      "balance_loss_clip": 1.06122732,
      "balance_loss_mlp": 1.02659535,
      "epoch": 0.19924246978897373,
      "flos": 15742340916480.0,
      "grad_norm": 21.40730918369674,
      "language_loss": 0.88574129,
      "learning_rate": 3.707129552610981e-06,
      "loss": 0.907969,
      "num_input_tokens_seen": 35125040,
      "step": 1657,
      "time_per_iteration": 2.68113112449646
    },
    {
      "auxiliary_loss_clip": 0.01195371,
      "auxiliary_loss_mlp": 0.01038849,
      "balance_loss_clip": 1.0637666,
      "balance_loss_mlp": 1.02791715,
      "epoch": 0.1993627126796128,
      "flos": 17566566986880.0,
      "grad_norm": 2.9996283875539134,
      "language_loss": 0.73605645,
      "learning_rate": 3.70672358924499e-06,
      "loss": 0.75839865,
      "num_input_tokens_seen": 35144280,
      "step": 1658,
      "time_per_iteration": 2.6877496242523193
    },
    {
      "auxiliary_loss_clip": 0.01173092,
      "auxiliary_loss_mlp": 0.01043453,
      "balance_loss_clip": 1.06344867,
      "balance_loss_mlp": 1.03250933,
      "epoch": 0.19948295557025192,
      "flos": 40843826680320.0,
      "grad_norm": 2.8278237800794983,
      "language_loss": 0.78986335,
      "learning_rate": 3.706317366970386e-06,
      "loss": 0.81202877,
      "num_input_tokens_seen": 35165280,
      "step": 1659,
      "time_per_iteration": 2.920008897781372
    },
    {
      "auxiliary_loss_clip": 0.01168716,
      "auxiliary_loss_mlp": 0.0104248,
      "balance_loss_clip": 1.0585202,
      "balance_loss_mlp": 1.03180432,
      "epoch": 0.199603198460891,
      "flos": 25082418620160.0,
      "grad_norm": 2.0287809751414443,
      "language_loss": 0.83302653,
      "learning_rate": 3.705910885848795e-06,
      "loss": 0.85513842,
      "num_input_tokens_seen": 35183655,
      "step": 1660,
      "time_per_iteration": 2.776064395904541
    },
    {
      "auxiliary_loss_clip": 0.01194177,
      "auxiliary_loss_mlp": 0.01039434,
      "balance_loss_clip": 1.06765711,
      "balance_loss_mlp": 1.02975452,
      "epoch": 0.19972344135153008,
      "flos": 20084120352000.0,
      "grad_norm": 2.897648080857401,
      "language_loss": 0.84473717,
      "learning_rate": 3.705504145941879e-06,
      "loss": 0.8670733,
      "num_input_tokens_seen": 35201825,
      "step": 1661,
      "time_per_iteration": 2.7034943103790283
    },
    {
      "auxiliary_loss_clip": 0.01167769,
      "auxiliary_loss_mlp": 0.01032395,
      "balance_loss_clip": 1.0543704,
      "balance_loss_mlp": 1.02285886,
      "epoch": 0.1998436842421692,
      "flos": 23727472761600.0,
      "grad_norm": 3.2004519209146105,
      "language_loss": 0.78589225,
      "learning_rate": 3.7050971473113403e-06,
      "loss": 0.80789387,
      "num_input_tokens_seen": 35221600,
      "step": 1662,
      "time_per_iteration": 2.7363874912261963
    },
    {
      "auxiliary_loss_clip": 0.01227066,
      "auxiliary_loss_mlp": 0.01038842,
      "balance_loss_clip": 1.07014287,
      "balance_loss_mlp": 1.02910876,
      "epoch": 0.19996392713280828,
      "flos": 36102361633920.0,
      "grad_norm": 3.391721715625202,
      "language_loss": 0.79637456,
      "learning_rate": 3.7046898900189196e-06,
      "loss": 0.81903368,
      "num_input_tokens_seen": 35245935,
      "step": 1663,
      "time_per_iteration": 2.7770020961761475
    },
    {
      "auxiliary_loss_clip": 0.01198986,
      "auxiliary_loss_mlp": 0.01035783,
      "balance_loss_clip": 1.06397152,
      "balance_loss_mlp": 1.02566159,
      "epoch": 0.20008417002344736,
      "flos": 23657662679040.0,
      "grad_norm": 2.149490585395519,
      "language_loss": 0.83469492,
      "learning_rate": 3.704282374126398e-06,
      "loss": 0.85704255,
      "num_input_tokens_seen": 35265615,
      "step": 1664,
      "time_per_iteration": 2.716355800628662
    },
    {
      "auxiliary_loss_clip": 0.01217897,
      "auxiliary_loss_mlp": 0.01048079,
      "balance_loss_clip": 1.07077909,
      "balance_loss_mlp": 1.03725469,
      "epoch": 0.20020441291408644,
      "flos": 21872076664320.0,
      "grad_norm": 1.9751186294005936,
      "language_loss": 0.87618893,
      "learning_rate": 3.7038745996955954e-06,
      "loss": 0.89884877,
      "num_input_tokens_seen": 35284960,
      "step": 1665,
      "time_per_iteration": 2.904573678970337
    },
    {
      "auxiliary_loss_clip": 0.01236773,
      "auxiliary_loss_mlp": 0.01043131,
      "balance_loss_clip": 1.07285547,
      "balance_loss_mlp": 1.03149629,
      "epoch": 0.20032465580472555,
      "flos": 23179691376000.0,
      "grad_norm": 5.900544460311403,
      "language_loss": 0.72504479,
      "learning_rate": 3.703466566788371e-06,
      "loss": 0.74784386,
      "num_input_tokens_seen": 35304090,
      "step": 1666,
      "time_per_iteration": 2.81367826461792
    },
    {
      "auxiliary_loss_clip": 0.01188799,
      "auxiliary_loss_mlp": 0.01037107,
      "balance_loss_clip": 1.06467342,
      "balance_loss_mlp": 1.02689075,
      "epoch": 0.20044489869536464,
      "flos": 23873521461120.0,
      "grad_norm": 2.844652352783047,
      "language_loss": 0.74839091,
      "learning_rate": 3.703058275466622e-06,
      "loss": 0.77064997,
      "num_input_tokens_seen": 35323325,
      "step": 1667,
      "time_per_iteration": 2.873039722442627
    },
    {
      "auxiliary_loss_clip": 0.01204154,
      "auxiliary_loss_mlp": 0.01036803,
      "balance_loss_clip": 1.0661025,
      "balance_loss_mlp": 1.02693868,
      "epoch": 0.20056514158600372,
      "flos": 21945226711680.0,
      "grad_norm": 3.8470775392213508,
      "language_loss": 0.77504647,
      "learning_rate": 3.7026497257922877e-06,
      "loss": 0.79745603,
      "num_input_tokens_seen": 35343635,
      "step": 1668,
      "time_per_iteration": 2.779733180999756
    },
    {
      "auxiliary_loss_clip": 0.01169511,
      "auxiliary_loss_mlp": 0.01035435,
      "balance_loss_clip": 1.05864596,
      "balance_loss_mlp": 1.02542138,
      "epoch": 0.20068538447664283,
      "flos": 23879159896320.0,
      "grad_norm": 5.620169962165064,
      "language_loss": 0.85214782,
      "learning_rate": 3.7022409178273436e-06,
      "loss": 0.8741973,
      "num_input_tokens_seen": 35364615,
      "step": 1669,
      "time_per_iteration": 2.867647647857666
    },
    {
      "auxiliary_loss_clip": 0.01168719,
      "auxiliary_loss_mlp": 0.01036518,
      "balance_loss_clip": 1.06063306,
      "balance_loss_mlp": 1.02648091,
      "epoch": 0.2008056273672819,
      "flos": 18442823270400.0,
      "grad_norm": 8.694439101071538,
      "language_loss": 0.78539276,
      "learning_rate": 3.7018318516338054e-06,
      "loss": 0.80744517,
      "num_input_tokens_seen": 35383775,
      "step": 1670,
      "time_per_iteration": 2.8148257732391357
    },
    {
      "auxiliary_loss_clip": 0.01151012,
      "auxiliary_loss_mlp": 0.01035882,
      "balance_loss_clip": 1.05749941,
      "balance_loss_mlp": 1.02554631,
      "epoch": 0.200925870257921,
      "flos": 23659530186240.0,
      "grad_norm": 3.318328833232391,
      "language_loss": 0.81768674,
      "learning_rate": 3.7014225272737284e-06,
      "loss": 0.83955574,
      "num_input_tokens_seen": 35403000,
      "step": 1671,
      "time_per_iteration": 2.8785758018493652
    },
    {
      "auxiliary_loss_clip": 0.01226503,
      "auxiliary_loss_mlp": 0.01040417,
      "balance_loss_clip": 1.06923485,
      "balance_loss_mlp": 1.03006959,
      "epoch": 0.20104611314856008,
      "flos": 16217115909120.0,
      "grad_norm": 2.9632056463047807,
      "language_loss": 0.74248147,
      "learning_rate": 3.701012944809207e-06,
      "loss": 0.76515067,
      "num_input_tokens_seen": 35420115,
      "step": 1672,
      "time_per_iteration": 2.7111799716949463
    },
    {
      "auxiliary_loss_clip": 0.01199188,
      "auxiliary_loss_mlp": 0.01034612,
      "balance_loss_clip": 1.06804729,
      "balance_loss_mlp": 1.02480149,
      "epoch": 0.2011663560391992,
      "flos": 21397373498880.0,
      "grad_norm": 2.834184026075779,
      "language_loss": 0.79224825,
      "learning_rate": 3.700603104302374e-06,
      "loss": 0.81458628,
      "num_input_tokens_seen": 35439925,
      "step": 1673,
      "time_per_iteration": 2.835474967956543
    },
    {
      "auxiliary_loss_clip": 0.01113331,
      "auxiliary_loss_mlp": 0.01002309,
      "balance_loss_clip": 1.02932131,
      "balance_loss_mlp": 0.99999613,
      "epoch": 0.20128659892983827,
      "flos": 62229459409920.0,
      "grad_norm": 0.9069289648872848,
      "language_loss": 0.55865467,
      "learning_rate": 3.7001930058154027e-06,
      "loss": 0.5798111,
      "num_input_tokens_seen": 35504885,
      "step": 1674,
      "time_per_iteration": 3.358030319213867
    },
    {
      "auxiliary_loss_clip": 0.01218903,
      "auxiliary_loss_mlp": 0.00765025,
      "balance_loss_clip": 1.06741989,
      "balance_loss_mlp": 1.00117767,
      "epoch": 0.20140684182047736,
      "flos": 28438737448320.0,
      "grad_norm": 3.1793982785841086,
      "language_loss": 0.7978797,
      "learning_rate": 3.6997826494105037e-06,
      "loss": 0.81771898,
      "num_input_tokens_seen": 35525330,
      "step": 1675,
      "time_per_iteration": 3.750825881958008
    },
    {
      "auxiliary_loss_clip": 0.01167452,
      "auxiliary_loss_mlp": 0.01042203,
      "balance_loss_clip": 1.05701566,
      "balance_loss_mlp": 1.03167713,
      "epoch": 0.20152708471111647,
      "flos": 28074064619520.0,
      "grad_norm": 21.541192552994325,
      "language_loss": 0.68929195,
      "learning_rate": 3.6993720351499286e-06,
      "loss": 0.71138853,
      "num_input_tokens_seen": 35546455,
      "step": 1676,
      "time_per_iteration": 2.8947677612304688
    },
    {
      "auxiliary_loss_clip": 0.01199714,
      "auxiliary_loss_mlp": 0.01030359,
      "balance_loss_clip": 1.06727695,
      "balance_loss_mlp": 1.02073288,
      "epoch": 0.20164732760175555,
      "flos": 23549751244800.0,
      "grad_norm": 4.186143603199412,
      "language_loss": 0.76833642,
      "learning_rate": 3.6989611630959666e-06,
      "loss": 0.7906372,
      "num_input_tokens_seen": 35565010,
      "step": 1677,
      "time_per_iteration": 2.846686363220215
    },
    {
      "auxiliary_loss_clip": 0.01075393,
      "auxiliary_loss_mlp": 0.01003285,
      "balance_loss_clip": 1.02674901,
      "balance_loss_mlp": 1.00090086,
      "epoch": 0.20176757049239463,
      "flos": 71100616037760.0,
      "grad_norm": 0.6798714854424499,
      "language_loss": 0.58203697,
      "learning_rate": 3.6985500333109474e-06,
      "loss": 0.60282373,
      "num_input_tokens_seen": 35633340,
      "step": 1678,
      "time_per_iteration": 5.8262939453125
    },
    {
      "auxiliary_loss_clip": 0.01215803,
      "auxiliary_loss_mlp": 0.01039766,
      "balance_loss_clip": 1.07135701,
      "balance_loss_mlp": 1.02968121,
      "epoch": 0.20188781338303372,
      "flos": 21430159637760.0,
      "grad_norm": 2.5713531296100114,
      "language_loss": 0.76567239,
      "learning_rate": 3.6981386458572385e-06,
      "loss": 0.78822809,
      "num_input_tokens_seen": 35651315,
      "step": 1679,
      "time_per_iteration": 2.878119945526123
    },
    {
      "auxiliary_loss_clip": 0.01231893,
      "auxiliary_loss_mlp": 0.01044648,
      "balance_loss_clip": 1.07090354,
      "balance_loss_mlp": 1.03371024,
      "epoch": 0.20200805627367283,
      "flos": 11546215130880.0,
      "grad_norm": 2.5563815085438946,
      "language_loss": 0.7640357,
      "learning_rate": 3.6977270007972468e-06,
      "loss": 0.7868011,
      "num_input_tokens_seen": 35668850,
      "step": 1680,
      "time_per_iteration": 2.7189536094665527
    },
    {
      "auxiliary_loss_clip": 0.01166036,
      "auxiliary_loss_mlp": 0.01043948,
      "balance_loss_clip": 1.06227016,
      "balance_loss_mlp": 1.03401756,
      "epoch": 0.2021282991643119,
      "flos": 28545391906560.0,
      "grad_norm": 9.997213841418969,
      "language_loss": 0.71969479,
      "learning_rate": 3.6973150981934196e-06,
      "loss": 0.74179459,
      "num_input_tokens_seen": 35690080,
      "step": 1681,
      "time_per_iteration": 2.9664390087127686
    },
    {
      "auxiliary_loss_clip": 0.01193872,
      "auxiliary_loss_mlp": 0.01043036,
      "balance_loss_clip": 1.06347322,
      "balance_loss_mlp": 1.03194904,
      "epoch": 0.202248542054951,
      "flos": 17923446564480.0,
      "grad_norm": 2.806235028251643,
      "language_loss": 0.84219766,
      "learning_rate": 3.6969029381082415e-06,
      "loss": 0.86456668,
      "num_input_tokens_seen": 35706075,
      "step": 1682,
      "time_per_iteration": 3.659203290939331
    },
    {
      "auxiliary_loss_clip": 0.01170182,
      "auxiliary_loss_mlp": 0.01035436,
      "balance_loss_clip": 1.05942011,
      "balance_loss_mlp": 1.02532697,
      "epoch": 0.2023687849455901,
      "flos": 19864634296320.0,
      "grad_norm": 2.420369362003721,
      "language_loss": 0.79482251,
      "learning_rate": 3.696490520604237e-06,
      "loss": 0.81687868,
      "num_input_tokens_seen": 35724765,
      "step": 1683,
      "time_per_iteration": 2.771146059036255
    },
    {
      "auxiliary_loss_clip": 0.0116773,
      "auxiliary_loss_mlp": 0.00763156,
      "balance_loss_clip": 1.05808771,
      "balance_loss_mlp": 1.00083089,
      "epoch": 0.20248902783622919,
      "flos": 22564721600640.0,
      "grad_norm": 2.4043435400970834,
      "language_loss": 0.80877924,
      "learning_rate": 3.696077845743968e-06,
      "loss": 0.82808805,
      "num_input_tokens_seen": 35744355,
      "step": 1684,
      "time_per_iteration": 2.8792638778686523
    },
    {
      "auxiliary_loss_clip": 0.01176083,
      "auxiliary_loss_mlp": 0.0103586,
      "balance_loss_clip": 1.0612545,
      "balance_loss_mlp": 1.02505398,
      "epoch": 0.20260927072686827,
      "flos": 22709728805760.0,
      "grad_norm": 3.2895195910245785,
      "language_loss": 0.7301563,
      "learning_rate": 3.69566491359004e-06,
      "loss": 0.75227576,
      "num_input_tokens_seen": 35761000,
      "step": 1685,
      "time_per_iteration": 2.8304836750030518
    },
    {
      "auxiliary_loss_clip": 0.01212349,
      "auxiliary_loss_mlp": 0.00764316,
      "balance_loss_clip": 1.06996465,
      "balance_loss_mlp": 1.00086415,
      "epoch": 0.20272951361750738,
      "flos": 51023998650240.0,
      "grad_norm": 3.4675357441458368,
      "language_loss": 0.6942004,
      "learning_rate": 3.695251724205092e-06,
      "loss": 0.71396708,
      "num_input_tokens_seen": 35785360,
      "step": 1686,
      "time_per_iteration": 3.0090134143829346
    },
    {
      "auxiliary_loss_clip": 0.01212776,
      "auxiliary_loss_mlp": 0.01047348,
      "balance_loss_clip": 1.06768131,
      "balance_loss_mlp": 1.03689933,
      "epoch": 0.20284975650814646,
      "flos": 26578133879040.0,
      "grad_norm": 2.0751432953602924,
      "language_loss": 0.85936058,
      "learning_rate": 3.6948382776518054e-06,
      "loss": 0.88196182,
      "num_input_tokens_seen": 35806065,
      "step": 1687,
      "time_per_iteration": 2.789914131164551
    },
    {
      "auxiliary_loss_clip": 0.01165899,
      "auxiliary_loss_mlp": 0.0104456,
      "balance_loss_clip": 1.06318736,
      "balance_loss_mlp": 1.03438568,
      "epoch": 0.20296999939878554,
      "flos": 16034222833920.0,
      "grad_norm": 2.311609180944968,
      "language_loss": 0.79583967,
      "learning_rate": 3.6944245739929e-06,
      "loss": 0.81794429,
      "num_input_tokens_seen": 35822225,
      "step": 1688,
      "time_per_iteration": 2.7892465591430664
    },
    {
      "auxiliary_loss_clip": 0.01208534,
      "auxiliary_loss_mlp": 0.0105773,
      "balance_loss_clip": 1.06724811,
      "balance_loss_mlp": 1.04669762,
      "epoch": 0.20309024228942463,
      "flos": 19203374868480.0,
      "grad_norm": 2.8843965573981367,
      "language_loss": 0.7123214,
      "learning_rate": 3.6940106132911332e-06,
      "loss": 0.73498404,
      "num_input_tokens_seen": 35839410,
      "step": 1689,
      "time_per_iteration": 2.715245246887207
    },
    {
      "auxiliary_loss_clip": 0.01194666,
      "auxiliary_loss_mlp": 0.01041759,
      "balance_loss_clip": 1.06395578,
      "balance_loss_mlp": 1.03145909,
      "epoch": 0.20321048518006374,
      "flos": 22821087945600.0,
      "grad_norm": 2.0865486673633593,
      "language_loss": 0.88637042,
      "learning_rate": 3.6935963956093037e-06,
      "loss": 0.90873468,
      "num_input_tokens_seen": 35859495,
      "step": 1690,
      "time_per_iteration": 2.835359811782837
    },
    {
      "auxiliary_loss_clip": 0.0119548,
      "auxiliary_loss_mlp": 0.00763685,
      "balance_loss_clip": 1.07012081,
      "balance_loss_mlp": 1.00090599,
      "epoch": 0.20333072807070282,
      "flos": 19096397187840.0,
      "grad_norm": 1.846048795726126,
      "language_loss": 0.69071966,
      "learning_rate": 3.6931819210102474e-06,
      "loss": 0.71031129,
      "num_input_tokens_seen": 35878890,
      "step": 1691,
      "time_per_iteration": 2.770801544189453
    },
    {
      "auxiliary_loss_clip": 0.01201473,
      "auxiliary_loss_mlp": 0.00764248,
      "balance_loss_clip": 1.06841755,
      "balance_loss_mlp": 1.00093246,
      "epoch": 0.2034509709613419,
      "flos": 18180962144640.0,
      "grad_norm": 1.9277716723487728,
      "language_loss": 0.84224027,
      "learning_rate": 3.6927671895568402e-06,
      "loss": 0.86189747,
      "num_input_tokens_seen": 35897950,
      "step": 1692,
      "time_per_iteration": 2.7896950244903564
    },
    {
      "auxiliary_loss_clip": 0.01183488,
      "auxiliary_loss_mlp": 0.01033002,
      "balance_loss_clip": 1.06903362,
      "balance_loss_mlp": 1.02241039,
      "epoch": 0.20357121385198101,
      "flos": 22923899648640.0,
      "grad_norm": 2.5757645110828427,
      "language_loss": 0.86607766,
      "learning_rate": 3.692352201311996e-06,
      "loss": 0.8882426,
      "num_input_tokens_seen": 35916800,
      "step": 1693,
      "time_per_iteration": 2.8671326637268066
    },
    {
      "auxiliary_loss_clip": 0.01212589,
      "auxiliary_loss_mlp": 0.01043084,
      "balance_loss_clip": 1.07048404,
      "balance_loss_mlp": 1.03242111,
      "epoch": 0.2036914567426201,
      "flos": 20922131629440.0,
      "grad_norm": 2.458089821022394,
      "language_loss": 0.76386452,
      "learning_rate": 3.6919369563386687e-06,
      "loss": 0.78642124,
      "num_input_tokens_seen": 35936600,
      "step": 1694,
      "time_per_iteration": 2.7505948543548584
    },
    {
      "auxiliary_loss_clip": 0.01184818,
      "auxiliary_loss_mlp": 0.01041329,
      "balance_loss_clip": 1.06376266,
      "balance_loss_mlp": 1.03188169,
      "epoch": 0.20381169963325918,
      "flos": 15519155760000.0,
      "grad_norm": 4.295344847637873,
      "language_loss": 0.79270339,
      "learning_rate": 3.69152145469985e-06,
      "loss": 0.81496489,
      "num_input_tokens_seen": 35953645,
      "step": 1695,
      "time_per_iteration": 2.7820708751678467
    },
    {
      "auxiliary_loss_clip": 0.01179897,
      "auxiliary_loss_mlp": 0.01041539,
      "balance_loss_clip": 1.06160426,
      "balance_loss_mlp": 1.03045869,
      "epoch": 0.20393194252389826,
      "flos": 28833143760000.0,
      "grad_norm": 2.5792242815408604,
      "language_loss": 0.82496834,
      "learning_rate": 3.691105696458572e-06,
      "loss": 0.84718269,
      "num_input_tokens_seen": 35970940,
      "step": 1696,
      "time_per_iteration": 2.850456953048706
    },
    {
      "auxiliary_loss_clip": 0.0120232,
      "auxiliary_loss_mlp": 0.01037476,
      "balance_loss_clip": 1.07267714,
      "balance_loss_mlp": 1.02731967,
      "epoch": 0.20405218541453737,
      "flos": 22488554810880.0,
      "grad_norm": 4.006193237428477,
      "language_loss": 0.6831665,
      "learning_rate": 3.690689681677904e-06,
      "loss": 0.70556444,
      "num_input_tokens_seen": 35989410,
      "step": 1697,
      "time_per_iteration": 2.75911021232605
    },
    {
      "auxiliary_loss_clip": 0.01214202,
      "auxiliary_loss_mlp": 0.0103951,
      "balance_loss_clip": 1.07102418,
      "balance_loss_mlp": 1.02997291,
      "epoch": 0.20417242830517646,
      "flos": 25374408278400.0,
      "grad_norm": 2.405621634771458,
      "language_loss": 0.88739055,
      "learning_rate": 3.690273410420956e-06,
      "loss": 0.90992773,
      "num_input_tokens_seen": 36009175,
      "step": 1698,
      "time_per_iteration": 2.809511661529541
    },
    {
      "auxiliary_loss_clip": 0.01201586,
      "auxiliary_loss_mlp": 0.01031792,
      "balance_loss_clip": 1.06575811,
      "balance_loss_mlp": 1.02090776,
      "epoch": 0.20429267119581554,
      "flos": 14793078240000.0,
      "grad_norm": 2.788374309757807,
      "language_loss": 0.76863587,
      "learning_rate": 3.689856882750875e-06,
      "loss": 0.79096961,
      "num_input_tokens_seen": 36024375,
      "step": 1699,
      "time_per_iteration": 2.7455050945281982
    },
    {
      "auxiliary_loss_clip": 0.01192421,
      "auxiliary_loss_mlp": 0.0103869,
      "balance_loss_clip": 1.06584454,
      "balance_loss_mlp": 1.02884305,
      "epoch": 0.20441291408645465,
      "flos": 17781851151360.0,
      "grad_norm": 2.269642539771912,
      "language_loss": 0.78750765,
      "learning_rate": 3.6894400987308486e-06,
      "loss": 0.80981869,
      "num_input_tokens_seen": 36041895,
      "step": 1700,
      "time_per_iteration": 2.7243340015411377
    },
    {
      "auxiliary_loss_clip": 0.01197116,
      "auxiliary_loss_mlp": 0.01034032,
      "balance_loss_clip": 1.06491435,
      "balance_loss_mlp": 1.02340412,
      "epoch": 0.20453315697709373,
      "flos": 16435668211200.0,
      "grad_norm": 2.452871435249145,
      "language_loss": 0.84688556,
      "learning_rate": 3.6890230584241024e-06,
      "loss": 0.86919707,
      "num_input_tokens_seen": 36058825,
      "step": 1701,
      "time_per_iteration": 3.6490039825439453
    },
    {
      "auxiliary_loss_clip": 0.01078501,
      "auxiliary_loss_mlp": 0.01020249,
      "balance_loss_clip": 1.02418423,
      "balance_loss_mlp": 1.01773381,
      "epoch": 0.20465339986773282,
      "flos": 66713085653760.0,
      "grad_norm": 1.0948931368383246,
      "language_loss": 0.66285509,
      "learning_rate": 3.6886057618939016e-06,
      "loss": 0.6838426,
      "num_input_tokens_seen": 36121645,
      "step": 1702,
      "time_per_iteration": 3.362197160720825
    },
    {
      "auxiliary_loss_clip": 0.01213406,
      "auxiliary_loss_mlp": 0.01037183,
      "balance_loss_clip": 1.06994486,
      "balance_loss_mlp": 1.02651358,
      "epoch": 0.2047736427583719,
      "flos": 41974114924800.0,
      "grad_norm": 2.752872188469523,
      "language_loss": 0.69877326,
      "learning_rate": 3.6881882092035492e-06,
      "loss": 0.72127914,
      "num_input_tokens_seen": 36143030,
      "step": 1703,
      "time_per_iteration": 4.882632732391357
    },
    {
      "auxiliary_loss_clip": 0.01108921,
      "auxiliary_loss_mlp": 0.01013877,
      "balance_loss_clip": 1.0253135,
      "balance_loss_mlp": 1.01142144,
      "epoch": 0.204893885649011,
      "flos": 69940878641280.0,
      "grad_norm": 0.9949269504213445,
      "language_loss": 0.61154819,
      "learning_rate": 3.6877704004163873e-06,
      "loss": 0.63277614,
      "num_input_tokens_seen": 36203435,
      "step": 1704,
      "time_per_iteration": 3.408586025238037
    },
    {
      "auxiliary_loss_clip": 0.01171262,
      "auxiliary_loss_mlp": 0.01042026,
      "balance_loss_clip": 1.06258774,
      "balance_loss_mlp": 1.03088593,
      "epoch": 0.2050141285396501,
      "flos": 22200012858240.0,
      "grad_norm": 2.2638589374343816,
      "language_loss": 0.77817702,
      "learning_rate": 3.6873523355957984e-06,
      "loss": 0.80030996,
      "num_input_tokens_seen": 36222435,
      "step": 1705,
      "time_per_iteration": 2.8672077655792236
    },
    {
      "auxiliary_loss_clip": 0.01119693,
      "auxiliary_loss_mlp": 0.01001626,
      "balance_loss_clip": 1.02698648,
      "balance_loss_mlp": 0.99939644,
      "epoch": 0.20513437143028918,
      "flos": 46283721730560.0,
      "grad_norm": 1.0196395097398225,
      "language_loss": 0.64112079,
      "learning_rate": 3.686934014805201e-06,
      "loss": 0.66233397,
      "num_input_tokens_seen": 36273065,
      "step": 1706,
      "time_per_iteration": 3.142138719558716
    },
    {
      "auxiliary_loss_clip": 0.01145971,
      "auxiliary_loss_mlp": 0.01040776,
      "balance_loss_clip": 1.05918622,
      "balance_loss_mlp": 1.03046441,
      "epoch": 0.20525461432092829,
      "flos": 21904324099200.0,
      "grad_norm": 3.7177322842902623,
      "language_loss": 0.81024563,
      "learning_rate": 3.6865154381080552e-06,
      "loss": 0.83211315,
      "num_input_tokens_seen": 36293750,
      "step": 1707,
      "time_per_iteration": 2.9147608280181885
    },
    {
      "auxiliary_loss_clip": 0.01199713,
      "auxiliary_loss_mlp": 0.0103525,
      "balance_loss_clip": 1.06423473,
      "balance_loss_mlp": 1.02502775,
      "epoch": 0.20537485721156737,
      "flos": 21214264942080.0,
      "grad_norm": 2.7957110223514436,
      "language_loss": 0.8237322,
      "learning_rate": 3.6860966055678585e-06,
      "loss": 0.84608185,
      "num_input_tokens_seen": 36310105,
      "step": 1708,
      "time_per_iteration": 3.81831955909729
    },
    {
      "auxiliary_loss_clip": 0.0121948,
      "auxiliary_loss_mlp": 0.01046021,
      "balance_loss_clip": 1.06990004,
      "balance_loss_mlp": 1.03457069,
      "epoch": 0.20549510010220645,
      "flos": 20191205773440.0,
      "grad_norm": 2.138360165916907,
      "language_loss": 0.86473477,
      "learning_rate": 3.685677517248147e-06,
      "loss": 0.88738978,
      "num_input_tokens_seen": 36328995,
      "step": 1709,
      "time_per_iteration": 2.897312641143799
    },
    {
      "auxiliary_loss_clip": 0.01212588,
      "auxiliary_loss_mlp": 0.01050246,
      "balance_loss_clip": 1.06962824,
      "balance_loss_mlp": 1.0394516,
      "epoch": 0.20561534299284553,
      "flos": 17016702612480.0,
      "grad_norm": 4.871406472553274,
      "language_loss": 0.8015092,
      "learning_rate": 3.6852581732124967e-06,
      "loss": 0.82413757,
      "num_input_tokens_seen": 36346340,
      "step": 1710,
      "time_per_iteration": 2.845710515975952
    },
    {
      "auxiliary_loss_clip": 0.01218248,
      "auxiliary_loss_mlp": 0.00764441,
      "balance_loss_clip": 1.07189679,
      "balance_loss_mlp": 1.0006789,
      "epoch": 0.20573558588348465,
      "flos": 22890467064960.0,
      "grad_norm": 3.1828102437067676,
      "language_loss": 0.76529592,
      "learning_rate": 3.6848385735245213e-06,
      "loss": 0.78512275,
      "num_input_tokens_seen": 36365430,
      "step": 1711,
      "time_per_iteration": 2.804446220397949
    },
    {
      "auxiliary_loss_clip": 0.01206636,
      "auxiliary_loss_mlp": 0.01035308,
      "balance_loss_clip": 1.06472158,
      "balance_loss_mlp": 1.02506244,
      "epoch": 0.20585582877412373,
      "flos": 24643123286400.0,
      "grad_norm": 1.9874458832630932,
      "language_loss": 0.86420321,
      "learning_rate": 3.6844187182478734e-06,
      "loss": 0.88662267,
      "num_input_tokens_seen": 36386285,
      "step": 1712,
      "time_per_iteration": 2.723360300064087
    },
    {
      "auxiliary_loss_clip": 0.01198933,
      "auxiliary_loss_mlp": 0.01039602,
      "balance_loss_clip": 1.06627274,
      "balance_loss_mlp": 1.02898002,
      "epoch": 0.2059760716647628,
      "flos": 24206952435840.0,
      "grad_norm": 2.1617789642030902,
      "language_loss": 0.75213099,
      "learning_rate": 3.683998607446246e-06,
      "loss": 0.77451628,
      "num_input_tokens_seen": 36404935,
      "step": 1713,
      "time_per_iteration": 2.7975711822509766
    },
    {
      "auxiliary_loss_clip": 0.01217642,
      "auxiliary_loss_mlp": 0.0105008,
      "balance_loss_clip": 1.06942499,
      "balance_loss_mlp": 1.03941655,
      "epoch": 0.20609631455540192,
      "flos": 20229522606720.0,
      "grad_norm": 2.2792716374607496,
      "language_loss": 0.74944073,
      "learning_rate": 3.6835782411833686e-06,
      "loss": 0.77211797,
      "num_input_tokens_seen": 36424455,
      "step": 1714,
      "time_per_iteration": 2.742173910140991
    },
    {
      "auxiliary_loss_clip": 0.01210076,
      "auxiliary_loss_mlp": 0.01050412,
      "balance_loss_clip": 1.06804359,
      "balance_loss_mlp": 1.03944468,
      "epoch": 0.206216557446041,
      "flos": 19864957518720.0,
      "grad_norm": 2.098401505413505,
      "language_loss": 0.7439245,
      "learning_rate": 3.68315761952301e-06,
      "loss": 0.76652932,
      "num_input_tokens_seen": 36441685,
      "step": 1715,
      "time_per_iteration": 2.784900188446045
    },
    {
      "auxiliary_loss_clip": 0.01176519,
      "auxiliary_loss_mlp": 0.01047747,
      "balance_loss_clip": 1.06200218,
      "balance_loss_mlp": 1.0367676,
      "epoch": 0.2063368003366801,
      "flos": 24096311568000.0,
      "grad_norm": 3.3368244887416765,
      "language_loss": 0.82988101,
      "learning_rate": 3.6827367425289797e-06,
      "loss": 0.85212362,
      "num_input_tokens_seen": 36461460,
      "step": 1716,
      "time_per_iteration": 2.8437511920928955
    },
    {
      "auxiliary_loss_clip": 0.01175202,
      "auxiliary_loss_mlp": 0.01047827,
      "balance_loss_clip": 1.06228328,
      "balance_loss_mlp": 1.03681755,
      "epoch": 0.2064570432273192,
      "flos": 20340163474560.0,
      "grad_norm": 3.590024461892872,
      "language_loss": 0.72455859,
      "learning_rate": 3.6823156102651225e-06,
      "loss": 0.74678886,
      "num_input_tokens_seen": 36479615,
      "step": 1717,
      "time_per_iteration": 2.801696300506592
    },
    {
      "auxiliary_loss_clip": 0.01138542,
      "auxiliary_loss_mlp": 0.01041448,
      "balance_loss_clip": 1.05792439,
      "balance_loss_mlp": 1.03023076,
      "epoch": 0.20657728611795828,
      "flos": 20520363029760.0,
      "grad_norm": 10.949354346126254,
      "language_loss": 0.70975351,
      "learning_rate": 3.6818942227953257e-06,
      "loss": 0.73155338,
      "num_input_tokens_seen": 36500160,
      "step": 1718,
      "time_per_iteration": 2.9641189575195312
    },
    {
      "auxiliary_loss_clip": 0.01161367,
      "auxiliary_loss_mlp": 0.00764674,
      "balance_loss_clip": 1.05851841,
      "balance_loss_mlp": 1.00047517,
      "epoch": 0.20669752900859736,
      "flos": 21799285752960.0,
      "grad_norm": 2.761090413986386,
      "language_loss": 0.68874896,
      "learning_rate": 3.681472580183512e-06,
      "loss": 0.70800936,
      "num_input_tokens_seen": 36518810,
      "step": 1719,
      "time_per_iteration": 2.8646998405456543
    },
    {
      "auxiliary_loss_clip": 0.01153066,
      "auxiliary_loss_mlp": 0.01041503,
      "balance_loss_clip": 1.05981922,
      "balance_loss_mlp": 1.03097701,
      "epoch": 0.20681777189923645,
      "flos": 15122020014720.0,
      "grad_norm": 2.384973533399359,
      "language_loss": 0.8666988,
      "learning_rate": 3.6810506824936455e-06,
      "loss": 0.88864446,
      "num_input_tokens_seen": 36536890,
      "step": 1720,
      "time_per_iteration": 2.9469969272613525
    },
    {
      "auxiliary_loss_clip": 0.01087537,
      "auxiliary_loss_mlp": 0.01021051,
      "balance_loss_clip": 1.02968049,
      "balance_loss_mlp": 1.01863098,
      "epoch": 0.20693801478987556,
      "flos": 56481021509760.0,
      "grad_norm": 1.05404300549343,
      "language_loss": 0.62554306,
      "learning_rate": 3.680628529789726e-06,
      "loss": 0.64662892,
      "num_input_tokens_seen": 36589300,
      "step": 1721,
      "time_per_iteration": 3.2169697284698486
    },
    {
      "auxiliary_loss_clip": 0.01208546,
      "auxiliary_loss_mlp": 0.01042731,
      "balance_loss_clip": 1.07034588,
      "balance_loss_mlp": 1.0311681,
      "epoch": 0.20705825768051464,
      "flos": 21614201948160.0,
      "grad_norm": 2.172751806370793,
      "language_loss": 0.864254,
      "learning_rate": 3.680206122135796e-06,
      "loss": 0.88676679,
      "num_input_tokens_seen": 36609905,
      "step": 1722,
      "time_per_iteration": 2.8336706161499023
    },
    {
      "auxiliary_loss_clip": 0.01205692,
      "auxiliary_loss_mlp": 0.01044569,
      "balance_loss_clip": 1.07187819,
      "balance_loss_mlp": 1.03449011,
      "epoch": 0.20717850057115372,
      "flos": 25848895962240.0,
      "grad_norm": 2.7318963493981467,
      "language_loss": 0.78353941,
      "learning_rate": 3.6797834595959323e-06,
      "loss": 0.80604208,
      "num_input_tokens_seen": 36629805,
      "step": 1723,
      "time_per_iteration": 2.8661069869995117
    },
    {
      "auxiliary_loss_clip": 0.01231166,
      "auxiliary_loss_mlp": 0.01038163,
      "balance_loss_clip": 1.06948233,
      "balance_loss_mlp": 1.02683246,
      "epoch": 0.20729874346179283,
      "flos": 29130807767040.0,
      "grad_norm": 5.92232143557035,
      "language_loss": 0.7756952,
      "learning_rate": 3.679360542234254e-06,
      "loss": 0.79838848,
      "num_input_tokens_seen": 36649150,
      "step": 1724,
      "time_per_iteration": 2.7758708000183105
    },
    {
      "auxiliary_loss_clip": 0.01156251,
      "auxiliary_loss_mlp": 0.00764155,
      "balance_loss_clip": 1.06018746,
      "balance_loss_mlp": 1.00038505,
      "epoch": 0.20741898635243192,
      "flos": 29023363209600.0,
      "grad_norm": 3.19075946524994,
      "language_loss": 0.72256231,
      "learning_rate": 3.678937370114916e-06,
      "loss": 0.74176639,
      "num_input_tokens_seen": 36668955,
      "step": 1725,
      "time_per_iteration": 3.2160348892211914
    },
    {
      "auxiliary_loss_clip": 0.01182955,
      "auxiliary_loss_mlp": 0.01030302,
      "balance_loss_clip": 1.06902337,
      "balance_loss_mlp": 1.02047896,
      "epoch": 0.207539229243071,
      "flos": 15559447841280.0,
      "grad_norm": 2.227276426225569,
      "language_loss": 0.78442502,
      "learning_rate": 3.678513943302114e-06,
      "loss": 0.80655754,
      "num_input_tokens_seen": 36685730,
      "step": 1726,
      "time_per_iteration": 2.802962064743042
    },
    {
      "auxiliary_loss_clip": 0.01183126,
      "auxiliary_loss_mlp": 0.01038347,
      "balance_loss_clip": 1.07121718,
      "balance_loss_mlp": 1.02735019,
      "epoch": 0.20765947213371008,
      "flos": 20521081301760.0,
      "grad_norm": 2.280561554096551,
      "language_loss": 0.85091406,
      "learning_rate": 3.678090261860082e-06,
      "loss": 0.87312883,
      "num_input_tokens_seen": 36705460,
      "step": 1727,
      "time_per_iteration": 3.722201108932495
    },
    {
      "auxiliary_loss_clip": 0.01211767,
      "auxiliary_loss_mlp": 0.01044582,
      "balance_loss_clip": 1.0684495,
      "balance_loss_mlp": 1.03431845,
      "epoch": 0.2077797150243492,
      "flos": 19354415558400.0,
      "grad_norm": 2.978703442364855,
      "language_loss": 0.77655554,
      "learning_rate": 3.6776663258530906e-06,
      "loss": 0.799119,
      "num_input_tokens_seen": 36724110,
      "step": 1728,
      "time_per_iteration": 2.796360969543457
    },
    {
      "auxiliary_loss_clip": 0.01147267,
      "auxiliary_loss_mlp": 0.01035044,
      "balance_loss_clip": 1.05763984,
      "balance_loss_mlp": 1.02475011,
      "epoch": 0.20789995791498828,
      "flos": 21829952989440.0,
      "grad_norm": 2.3649608563521003,
      "language_loss": 0.71639001,
      "learning_rate": 3.6772421353454516e-06,
      "loss": 0.73821306,
      "num_input_tokens_seen": 36742705,
      "step": 1729,
      "time_per_iteration": 4.7521584033966064
    },
    {
      "auxiliary_loss_clip": 0.01183924,
      "auxiliary_loss_mlp": 0.00764112,
      "balance_loss_clip": 1.06521559,
      "balance_loss_mlp": 1.00056052,
      "epoch": 0.20802020080562736,
      "flos": 23148844571520.0,
      "grad_norm": 2.5612502484414064,
      "language_loss": 0.8834281,
      "learning_rate": 3.6768176904015153e-06,
      "loss": 0.90290844,
      "num_input_tokens_seen": 36762510,
      "step": 1730,
      "time_per_iteration": 2.8505938053131104
    },
    {
      "auxiliary_loss_clip": 0.01212409,
      "auxiliary_loss_mlp": 0.01041491,
      "balance_loss_clip": 1.06703401,
      "balance_loss_mlp": 1.03045201,
      "epoch": 0.20814044369626647,
      "flos": 23072677781760.0,
      "grad_norm": 2.789957606986402,
      "language_loss": 0.6046015,
      "learning_rate": 3.6763929910856674e-06,
      "loss": 0.62714052,
      "num_input_tokens_seen": 36780960,
      "step": 1731,
      "time_per_iteration": 2.733569622039795
    },
    {
      "auxiliary_loss_clip": 0.01143041,
      "auxiliary_loss_mlp": 0.01036259,
      "balance_loss_clip": 1.05383086,
      "balance_loss_mlp": 1.02650142,
      "epoch": 0.20826068658690555,
      "flos": 19608016556160.0,
      "grad_norm": 2.5785025603821903,
      "language_loss": 0.77648181,
      "learning_rate": 3.6759680374623365e-06,
      "loss": 0.79827482,
      "num_input_tokens_seen": 36798875,
      "step": 1732,
      "time_per_iteration": 2.7615251541137695
    },
    {
      "auxiliary_loss_clip": 0.01180892,
      "auxiliary_loss_mlp": 0.01038055,
      "balance_loss_clip": 1.06566572,
      "balance_loss_mlp": 1.02786255,
      "epoch": 0.20838092947754464,
      "flos": 25374049142400.0,
      "grad_norm": 2.315289861821465,
      "language_loss": 0.75211191,
      "learning_rate": 3.675542829595986e-06,
      "loss": 0.77430141,
      "num_input_tokens_seen": 36818540,
      "step": 1733,
      "time_per_iteration": 2.8580429553985596
    },
    {
      "auxiliary_loss_clip": 0.01151514,
      "auxiliary_loss_mlp": 0.01043749,
      "balance_loss_clip": 1.05838752,
      "balance_loss_mlp": 1.03301382,
      "epoch": 0.20850117236818372,
      "flos": 24061729749120.0,
      "grad_norm": 2.165263514575013,
      "language_loss": 0.79290789,
      "learning_rate": 3.6751173675511213e-06,
      "loss": 0.81486052,
      "num_input_tokens_seen": 36840585,
      "step": 1734,
      "time_per_iteration": 3.7504799365997314
    },
    {
      "auxiliary_loss_clip": 0.01194121,
      "auxiliary_loss_mlp": 0.01042278,
      "balance_loss_clip": 1.0633781,
      "balance_loss_mlp": 1.03128731,
      "epoch": 0.20862141525882283,
      "flos": 20077799558400.0,
      "grad_norm": 3.9705658171862597,
      "language_loss": 0.87334836,
      "learning_rate": 3.674691651392283e-06,
      "loss": 0.89571238,
      "num_input_tokens_seen": 36858255,
      "step": 1735,
      "time_per_iteration": 2.761723041534424
    },
    {
      "auxiliary_loss_clip": 0.0118648,
      "auxiliary_loss_mlp": 0.01041903,
      "balance_loss_clip": 1.06620085,
      "balance_loss_mlp": 1.03103709,
      "epoch": 0.2087416581494619,
      "flos": 39015183237120.0,
      "grad_norm": 3.44719436476558,
      "language_loss": 0.75431919,
      "learning_rate": 3.674265681184053e-06,
      "loss": 0.77660298,
      "num_input_tokens_seen": 36881515,
      "step": 1736,
      "time_per_iteration": 2.97029972076416
    },
    {
      "auxiliary_loss_clip": 0.01215274,
      "auxiliary_loss_mlp": 0.01038313,
      "balance_loss_clip": 1.06955791,
      "balance_loss_mlp": 1.02759588,
      "epoch": 0.208861901040101,
      "flos": 26101994169600.0,
      "grad_norm": 1.7629468298815443,
      "language_loss": 0.8654418,
      "learning_rate": 3.6738394569910504e-06,
      "loss": 0.88797772,
      "num_input_tokens_seen": 36902055,
      "step": 1737,
      "time_per_iteration": 2.761791229248047
    },
    {
      "auxiliary_loss_clip": 0.01215605,
      "auxiliary_loss_mlp": 0.01044629,
      "balance_loss_clip": 1.06896722,
      "balance_loss_mlp": 1.03353071,
      "epoch": 0.2089821439307401,
      "flos": 28398732675840.0,
      "grad_norm": 2.957673600328311,
      "language_loss": 0.82674062,
      "learning_rate": 3.6734129788779333e-06,
      "loss": 0.84934294,
      "num_input_tokens_seen": 36921230,
      "step": 1738,
      "time_per_iteration": 2.776869058609009
    },
    {
      "auxiliary_loss_clip": 0.01196434,
      "auxiliary_loss_mlp": 0.01042037,
      "balance_loss_clip": 1.06635904,
      "balance_loss_mlp": 1.0313735,
      "epoch": 0.2091023868213792,
      "flos": 21069616872960.0,
      "grad_norm": 1.9028182619949525,
      "language_loss": 0.90412283,
      "learning_rate": 3.6729862469093976e-06,
      "loss": 0.92650753,
      "num_input_tokens_seen": 36940325,
      "step": 1739,
      "time_per_iteration": 2.8069279193878174
    },
    {
      "auxiliary_loss_clip": 0.01196991,
      "auxiliary_loss_mlp": 0.01040091,
      "balance_loss_clip": 1.06721175,
      "balance_loss_mlp": 1.02942753,
      "epoch": 0.20922262971201827,
      "flos": 22455481363200.0,
      "grad_norm": 2.45606122250745,
      "language_loss": 0.83130097,
      "learning_rate": 3.6725592611501782e-06,
      "loss": 0.85367179,
      "num_input_tokens_seen": 36959000,
      "step": 1740,
      "time_per_iteration": 2.7991249561309814
    },
    {
      "auxiliary_loss_clip": 0.01212407,
      "auxiliary_loss_mlp": 0.01039542,
      "balance_loss_clip": 1.0684576,
      "balance_loss_mlp": 1.02954006,
      "epoch": 0.20934287260265738,
      "flos": 27852244179840.0,
      "grad_norm": 2.8194515778737257,
      "language_loss": 0.76823437,
      "learning_rate": 3.6721320216650496e-06,
      "loss": 0.79075384,
      "num_input_tokens_seen": 36979615,
      "step": 1741,
      "time_per_iteration": 2.777174711227417
    },
    {
      "auxiliary_loss_clip": 0.01180314,
      "auxiliary_loss_mlp": 0.01035695,
      "balance_loss_clip": 1.05637467,
      "balance_loss_mlp": 1.02553856,
      "epoch": 0.20946311549329646,
      "flos": 16435309075200.0,
      "grad_norm": 2.0811999969333885,
      "language_loss": 0.83442765,
      "learning_rate": 3.6717045285188215e-06,
      "loss": 0.85658777,
      "num_input_tokens_seen": 36997310,
      "step": 1742,
      "time_per_iteration": 2.7328169345855713
    },
    {
      "auxiliary_loss_clip": 0.01229749,
      "auxiliary_loss_mlp": 0.01048391,
      "balance_loss_clip": 1.06896162,
      "balance_loss_mlp": 1.03741765,
      "epoch": 0.20958335838393555,
      "flos": 22492720788480.0,
      "grad_norm": 3.74369593354061,
      "language_loss": 0.86627328,
      "learning_rate": 3.671276781776346e-06,
      "loss": 0.88905472,
      "num_input_tokens_seen": 37015965,
      "step": 1743,
      "time_per_iteration": 2.6990737915039062
    },
    {
      "auxiliary_loss_clip": 0.0116058,
      "auxiliary_loss_mlp": 0.01042522,
      "balance_loss_clip": 1.06053889,
      "balance_loss_mlp": 1.03271675,
      "epoch": 0.20970360127457463,
      "flos": 25224768218880.0,
      "grad_norm": 3.035100193840881,
      "language_loss": 0.6690889,
      "learning_rate": 3.6708487815025128e-06,
      "loss": 0.69111991,
      "num_input_tokens_seen": 37036545,
      "step": 1744,
      "time_per_iteration": 2.8330419063568115
    },
    {
      "auxiliary_loss_clip": 0.01162208,
      "auxiliary_loss_mlp": 0.01035869,
      "balance_loss_clip": 1.06136215,
      "balance_loss_mlp": 1.02477705,
      "epoch": 0.20982384416521374,
      "flos": 18479164855680.0,
      "grad_norm": 3.058550646429852,
      "language_loss": 0.7419616,
      "learning_rate": 3.6704205277622463e-06,
      "loss": 0.76394236,
      "num_input_tokens_seen": 37054985,
      "step": 1745,
      "time_per_iteration": 2.886451244354248
    },
    {
      "auxiliary_loss_clip": 0.01195927,
      "auxiliary_loss_mlp": 0.01037135,
      "balance_loss_clip": 1.0684309,
      "balance_loss_mlp": 1.02741313,
      "epoch": 0.20994408705585282,
      "flos": 25373546352000.0,
      "grad_norm": 2.782071689656215,
      "language_loss": 0.80439484,
      "learning_rate": 3.6699920206205146e-06,
      "loss": 0.82672542,
      "num_input_tokens_seen": 37075725,
      "step": 1746,
      "time_per_iteration": 2.7797555923461914
    },
    {
      "auxiliary_loss_clip": 0.01197932,
      "auxiliary_loss_mlp": 0.01030387,
      "balance_loss_clip": 1.06583953,
      "balance_loss_mlp": 1.02083206,
      "epoch": 0.2100643299464919,
      "flos": 21320955313920.0,
      "grad_norm": 1.9427789655355043,
      "language_loss": 0.81730008,
      "learning_rate": 3.669563260142321e-06,
      "loss": 0.83958328,
      "num_input_tokens_seen": 37094615,
      "step": 1747,
      "time_per_iteration": 2.814218759536743
    },
    {
      "auxiliary_loss_clip": 0.01129629,
      "auxiliary_loss_mlp": 0.01042323,
      "balance_loss_clip": 1.05472302,
      "balance_loss_mlp": 1.03167117,
      "epoch": 0.21018457283713102,
      "flos": 19354379644800.0,
      "grad_norm": 2.8755282480384485,
      "language_loss": 0.84220695,
      "learning_rate": 3.6691342463927083e-06,
      "loss": 0.86392653,
      "num_input_tokens_seen": 37113610,
      "step": 1748,
      "time_per_iteration": 2.891777992248535
    },
    {
      "auxiliary_loss_clip": 0.01216307,
      "auxiliary_loss_mlp": 0.01042037,
      "balance_loss_clip": 1.0707109,
      "balance_loss_mlp": 1.03090262,
      "epoch": 0.2103048157277701,
      "flos": 28330035914880.0,
      "grad_norm": 1.7902634867106768,
      "language_loss": 0.81615674,
      "learning_rate": 3.668704979436758e-06,
      "loss": 0.83874023,
      "num_input_tokens_seen": 37133705,
      "step": 1749,
      "time_per_iteration": 2.80043888092041
    },
    {
      "auxiliary_loss_clip": 0.01231711,
      "auxiliary_loss_mlp": 0.01039912,
      "balance_loss_clip": 1.07089114,
      "balance_loss_mlp": 1.02908802,
      "epoch": 0.21042505861840918,
      "flos": 17457290835840.0,
      "grad_norm": 3.413870284535619,
      "language_loss": 0.78745973,
      "learning_rate": 3.668275459339588e-06,
      "loss": 0.8101759,
      "num_input_tokens_seen": 37152185,
      "step": 1750,
      "time_per_iteration": 2.774514675140381
    },
    {
      "auxiliary_loss_clip": 0.01216465,
      "auxiliary_loss_mlp": 0.00764307,
      "balance_loss_clip": 1.0700748,
      "balance_loss_mlp": 1.00040734,
      "epoch": 0.21054530150904827,
      "flos": 14209817195520.0,
      "grad_norm": 3.1866861919387426,
      "language_loss": 0.8050918,
      "learning_rate": 3.667845686166358e-06,
      "loss": 0.82489955,
      "num_input_tokens_seen": 37169110,
      "step": 1751,
      "time_per_iteration": 2.750264883041382
    },
    {
      "auxiliary_loss_clip": 0.01166523,
      "auxiliary_loss_mlp": 0.01048665,
      "balance_loss_clip": 1.06292021,
      "balance_loss_mlp": 1.03739381,
      "epoch": 0.21066554439968738,
      "flos": 18618210403200.0,
      "grad_norm": 4.392599819631459,
      "language_loss": 0.86116153,
      "learning_rate": 3.6674156599822634e-06,
      "loss": 0.8833133,
      "num_input_tokens_seen": 37184905,
      "step": 1752,
      "time_per_iteration": 2.8422999382019043
    },
    {
      "auxiliary_loss_clip": 0.01171291,
      "auxiliary_loss_mlp": 0.01044772,
      "balance_loss_clip": 1.05974388,
      "balance_loss_mlp": 1.03294611,
      "epoch": 0.21078578729032646,
      "flos": 23658883741440.0,
      "grad_norm": 2.7816756942566094,
      "language_loss": 0.81795537,
      "learning_rate": 3.666985380852539e-06,
      "loss": 0.84011602,
      "num_input_tokens_seen": 37203910,
      "step": 1753,
      "time_per_iteration": 3.6934666633605957
    },
    {
      "auxiliary_loss_clip": 0.012038,
      "auxiliary_loss_mlp": 0.01046331,
      "balance_loss_clip": 1.06802154,
      "balance_loss_mlp": 1.03533435,
      "epoch": 0.21090603018096554,
      "flos": 29346379240320.0,
      "grad_norm": 2.7487205201508242,
      "language_loss": 0.74746561,
      "learning_rate": 3.6665548488424576e-06,
      "loss": 0.76996696,
      "num_input_tokens_seen": 37222670,
      "step": 1754,
      "time_per_iteration": 2.888841390609741
    },
    {
      "auxiliary_loss_clip": 0.01236844,
      "auxiliary_loss_mlp": 0.0104628,
      "balance_loss_clip": 1.07276237,
      "balance_loss_mlp": 1.03520584,
      "epoch": 0.21102627307160465,
      "flos": 23261245205760.0,
      "grad_norm": 1.8808665732534404,
      "language_loss": 0.87772179,
      "learning_rate": 3.6661240640173307e-06,
      "loss": 0.90055305,
      "num_input_tokens_seen": 37244140,
      "step": 1755,
      "time_per_iteration": 4.696909666061401
    },
    {
      "auxiliary_loss_clip": 0.01048025,
      "auxiliary_loss_mlp": 0.01006359,
      "balance_loss_clip": 1.02646887,
      "balance_loss_mlp": 1.00393915,
      "epoch": 0.21114651596224374,
      "flos": 54633454577280.0,
      "grad_norm": 0.8917751224056961,
      "language_loss": 0.5787456,
      "learning_rate": 3.6656930264425085e-06,
      "loss": 0.59928948,
      "num_input_tokens_seen": 37308185,
      "step": 1756,
      "time_per_iteration": 3.4572699069976807
    },
    {
      "auxiliary_loss_clip": 0.0115284,
      "auxiliary_loss_mlp": 0.01035384,
      "balance_loss_clip": 1.05910778,
      "balance_loss_mlp": 1.02492344,
      "epoch": 0.21126675885288282,
      "flos": 21543314457600.0,
      "grad_norm": 2.6406711766417588,
      "language_loss": 0.75306475,
      "learning_rate": 3.665261736183378e-06,
      "loss": 0.77494705,
      "num_input_tokens_seen": 37328220,
      "step": 1757,
      "time_per_iteration": 2.8167412281036377
    },
    {
      "auxiliary_loss_clip": 0.01169861,
      "auxiliary_loss_mlp": 0.01044537,
      "balance_loss_clip": 1.06057596,
      "balance_loss_mlp": 1.03311729,
      "epoch": 0.2113870017435219,
      "flos": 10961876678400.0,
      "grad_norm": 29.217641763162494,
      "language_loss": 0.88604021,
      "learning_rate": 3.664830193305366e-06,
      "loss": 0.90818417,
      "num_input_tokens_seen": 37345995,
      "step": 1758,
      "time_per_iteration": 2.8709328174591064
    },
    {
      "auxiliary_loss_clip": 0.01219398,
      "auxiliary_loss_mlp": 0.01042444,
      "balance_loss_clip": 1.06700766,
      "balance_loss_mlp": 1.03070807,
      "epoch": 0.211507244634161,
      "flos": 16653825463680.0,
      "grad_norm": 2.3966677146064708,
      "language_loss": 0.76698178,
      "learning_rate": 3.6643983978739373e-06,
      "loss": 0.78960019,
      "num_input_tokens_seen": 37362610,
      "step": 1759,
      "time_per_iteration": 2.679253101348877
    },
    {
      "auxiliary_loss_clip": 0.01147726,
      "auxiliary_loss_mlp": 0.01048892,
      "balance_loss_clip": 1.06250978,
      "balance_loss_mlp": 1.03795469,
      "epoch": 0.2116274875248001,
      "flos": 20954091755520.0,
      "grad_norm": 1.9892604655370936,
      "language_loss": 0.82164198,
      "learning_rate": 3.663966349954596e-06,
      "loss": 0.84360814,
      "num_input_tokens_seen": 37382790,
      "step": 1760,
      "time_per_iteration": 3.8437323570251465
    },
    {
      "auxiliary_loss_clip": 0.01120588,
      "auxiliary_loss_mlp": 0.01006418,
      "balance_loss_clip": 1.02882481,
      "balance_loss_mlp": 1.00433147,
      "epoch": 0.21174773041543918,
      "flos": 68196949424640.0,
      "grad_norm": 0.7912492105881802,
      "language_loss": 0.59658456,
      "learning_rate": 3.6635340496128816e-06,
      "loss": 0.6178546,
      "num_input_tokens_seen": 37439720,
      "step": 1761,
      "time_per_iteration": 3.2316761016845703
    },
    {
      "auxiliary_loss_clip": 0.01209003,
      "auxiliary_loss_mlp": 0.0103972,
      "balance_loss_clip": 1.06709158,
      "balance_loss_mlp": 1.02983141,
      "epoch": 0.2118679733060783,
      "flos": 20668315150080.0,
      "grad_norm": 2.1277766982600097,
      "language_loss": 0.92970937,
      "learning_rate": 3.6631014969143747e-06,
      "loss": 0.9521966,
      "num_input_tokens_seen": 37459410,
      "step": 1762,
      "time_per_iteration": 2.719568967819214
    },
    {
      "auxiliary_loss_clip": 0.01232867,
      "auxiliary_loss_mlp": 0.0104167,
      "balance_loss_clip": 1.07257318,
      "balance_loss_mlp": 1.02983892,
      "epoch": 0.21198821619671737,
      "flos": 23223431162880.0,
      "grad_norm": 2.7126883468389686,
      "language_loss": 0.89168751,
      "learning_rate": 3.662668691924693e-06,
      "loss": 0.91443288,
      "num_input_tokens_seen": 37480460,
      "step": 1763,
      "time_per_iteration": 2.7518486976623535
    },
    {
      "auxiliary_loss_clip": 0.0118904,
      "auxiliary_loss_mlp": 0.01042852,
      "balance_loss_clip": 1.06728959,
      "balance_loss_mlp": 1.031533,
      "epoch": 0.21210845908735645,
      "flos": 24498547044480.0,
      "grad_norm": 3.1426407463089636,
      "language_loss": 0.71166289,
      "learning_rate": 3.6622356347094927e-06,
      "loss": 0.73398179,
      "num_input_tokens_seen": 37502025,
      "step": 1764,
      "time_per_iteration": 2.847273826599121
    },
    {
      "auxiliary_loss_clip": 0.01196795,
      "auxiliary_loss_mlp": 0.01043595,
      "balance_loss_clip": 1.06624794,
      "balance_loss_mlp": 1.03233612,
      "epoch": 0.21222870197799554,
      "flos": 27089789160960.0,
      "grad_norm": 2.2134286056373824,
      "language_loss": 0.79125339,
      "learning_rate": 3.6618023253344684e-06,
      "loss": 0.81365728,
      "num_input_tokens_seen": 37520885,
      "step": 1765,
      "time_per_iteration": 2.8256189823150635
    },
    {
      "auxiliary_loss_clip": 0.01217203,
      "auxiliary_loss_mlp": 0.01051589,
      "balance_loss_clip": 1.06892395,
      "balance_loss_mlp": 1.03951287,
      "epoch": 0.21234894486863465,
      "flos": 16873850223360.0,
      "grad_norm": 1.8833943252317493,
      "language_loss": 0.8347609,
      "learning_rate": 3.6613687638653527e-06,
      "loss": 0.85744882,
      "num_input_tokens_seen": 37539055,
      "step": 1766,
      "time_per_iteration": 2.7365493774414062
    },
    {
      "auxiliary_loss_clip": 0.01218569,
      "auxiliary_loss_mlp": 0.01051873,
      "balance_loss_clip": 1.06801486,
      "balance_loss_mlp": 1.04135895,
      "epoch": 0.21246918775927373,
      "flos": 23474949171840.0,
      "grad_norm": 2.3410794003693085,
      "language_loss": 0.78001577,
      "learning_rate": 3.660934950367916e-06,
      "loss": 0.80272019,
      "num_input_tokens_seen": 37558300,
      "step": 1767,
      "time_per_iteration": 2.722208023071289
    },
    {
      "auxiliary_loss_clip": 0.01170871,
      "auxiliary_loss_mlp": 0.01054918,
      "balance_loss_clip": 1.06413484,
      "balance_loss_mlp": 1.04436195,
      "epoch": 0.21258943064991281,
      "flos": 22382295402240.0,
      "grad_norm": 2.539228615284188,
      "language_loss": 0.8373484,
      "learning_rate": 3.660500884907968e-06,
      "loss": 0.85960627,
      "num_input_tokens_seen": 37579040,
      "step": 1768,
      "time_per_iteration": 2.836787223815918
    },
    {
      "auxiliary_loss_clip": 0.01062972,
      "auxiliary_loss_mlp": 0.01016676,
      "balance_loss_clip": 1.02666211,
      "balance_loss_mlp": 1.01450634,
      "epoch": 0.21270967354055192,
      "flos": 59440168679040.0,
      "grad_norm": 0.82519497084691,
      "language_loss": 0.60001165,
      "learning_rate": 3.660066567551356e-06,
      "loss": 0.62080812,
      "num_input_tokens_seen": 37639185,
      "step": 1769,
      "time_per_iteration": 3.3073577880859375
    },
    {
      "auxiliary_loss_clip": 0.01192947,
      "auxiliary_loss_mlp": 0.01043483,
      "balance_loss_clip": 1.06348586,
      "balance_loss_mlp": 1.03281963,
      "epoch": 0.212829916431191,
      "flos": 21544032729600.0,
      "grad_norm": 2.8658128032559964,
      "language_loss": 0.84590936,
      "learning_rate": 3.6596319983639657e-06,
      "loss": 0.86827362,
      "num_input_tokens_seen": 37657765,
      "step": 1770,
      "time_per_iteration": 2.7936089038848877
    },
    {
      "auxiliary_loss_clip": 0.01231058,
      "auxiliary_loss_mlp": 0.01039262,
      "balance_loss_clip": 1.07266498,
      "balance_loss_mlp": 1.02826488,
      "epoch": 0.2129501593218301,
      "flos": 28987739896320.0,
      "grad_norm": 1.8267614481656549,
      "language_loss": 0.86189222,
      "learning_rate": 3.6591971774117214e-06,
      "loss": 0.88459545,
      "num_input_tokens_seen": 37680740,
      "step": 1771,
      "time_per_iteration": 2.785959243774414
    },
    {
      "auxiliary_loss_clip": 0.01204784,
      "auxiliary_loss_mlp": 0.00764982,
      "balance_loss_clip": 1.07052743,
      "balance_loss_mlp": 1.00039458,
      "epoch": 0.2130704022124692,
      "flos": 18806993308800.0,
      "grad_norm": 3.888111630739148,
      "language_loss": 0.80768913,
      "learning_rate": 3.6587621047605833e-06,
      "loss": 0.82738674,
      "num_input_tokens_seen": 37697910,
      "step": 1772,
      "time_per_iteration": 2.7697653770446777
    },
    {
      "auxiliary_loss_clip": 0.0115166,
      "auxiliary_loss_mlp": 0.01042424,
      "balance_loss_clip": 1.06327844,
      "balance_loss_mlp": 1.03233874,
      "epoch": 0.21319064510310828,
      "flos": 13918150759680.0,
      "grad_norm": 12.503767878426931,
      "language_loss": 0.86428225,
      "learning_rate": 3.6583267804765542e-06,
      "loss": 0.88622308,
      "num_input_tokens_seen": 37712245,
      "step": 1773,
      "time_per_iteration": 2.809265375137329
    },
    {
      "auxiliary_loss_clip": 0.01161559,
      "auxiliary_loss_mlp": 0.01041981,
      "balance_loss_clip": 1.05769944,
      "balance_loss_mlp": 1.03047752,
      "epoch": 0.21331088799374737,
      "flos": 20959694277120.0,
      "grad_norm": 3.4095420527696505,
      "language_loss": 0.8632232,
      "learning_rate": 3.6578912046256702e-06,
      "loss": 0.88525856,
      "num_input_tokens_seen": 37730765,
      "step": 1774,
      "time_per_iteration": 2.815067768096924
    },
    {
      "auxiliary_loss_clip": 0.01195838,
      "auxiliary_loss_mlp": 0.01045929,
      "balance_loss_clip": 1.06700063,
      "balance_loss_mlp": 1.03454995,
      "epoch": 0.21343113088438645,
      "flos": 18624638937600.0,
      "grad_norm": 5.6329064071972486,
      "language_loss": 0.76289082,
      "learning_rate": 3.6574553772740083e-06,
      "loss": 0.78530848,
      "num_input_tokens_seen": 37748695,
      "step": 1775,
      "time_per_iteration": 2.8092005252838135
    },
    {
      "auxiliary_loss_clip": 0.01054897,
      "auxiliary_loss_mlp": 0.01003711,
      "balance_loss_clip": 1.02393949,
      "balance_loss_mlp": 1.00118339,
      "epoch": 0.21355137377502556,
      "flos": 67413128791680.0,
      "grad_norm": 0.8611145273651414,
      "language_loss": 0.61779416,
      "learning_rate": 3.657019298487684e-06,
      "loss": 0.63838017,
      "num_input_tokens_seen": 37813705,
      "step": 1776,
      "time_per_iteration": 3.62554931640625
    },
    {
      "auxiliary_loss_clip": 0.01193671,
      "auxiliary_loss_mlp": 0.00764742,
      "balance_loss_clip": 1.06270695,
      "balance_loss_mlp": 1.00056291,
      "epoch": 0.21367161666566464,
      "flos": 34532095697280.0,
      "grad_norm": 1.9736814746198432,
      "language_loss": 0.84211957,
      "learning_rate": 3.6565829683328495e-06,
      "loss": 0.86170369,
      "num_input_tokens_seen": 37836330,
      "step": 1777,
      "time_per_iteration": 3.28527569770813
    },
    {
      "auxiliary_loss_clip": 0.01095952,
      "auxiliary_loss_mlp": 0.01041534,
      "balance_loss_clip": 1.04887986,
      "balance_loss_mlp": 1.03094256,
      "epoch": 0.21379185955630373,
      "flos": 18989347680000.0,
      "grad_norm": 3.6228186716446884,
      "language_loss": 0.86100912,
      "learning_rate": 3.6561463868756965e-06,
      "loss": 0.88238394,
      "num_input_tokens_seen": 37855030,
      "step": 1778,
      "time_per_iteration": 2.960256338119507
    },
    {
      "auxiliary_loss_clip": 0.01198511,
      "auxiliary_loss_mlp": 0.01056313,
      "balance_loss_clip": 1.06568074,
      "balance_loss_mlp": 1.04563189,
      "epoch": 0.21391210244694284,
      "flos": 28218497207040.0,
      "grad_norm": 1.9794780945851491,
      "language_loss": 0.78160262,
      "learning_rate": 3.655709554182452e-06,
      "loss": 0.80415088,
      "num_input_tokens_seen": 37875370,
      "step": 1779,
      "time_per_iteration": 3.7205758094787598
    },
    {
      "auxiliary_loss_clip": 0.01230659,
      "auxiliary_loss_mlp": 0.0104005,
      "balance_loss_clip": 1.07141578,
      "balance_loss_mlp": 1.02930892,
      "epoch": 0.21403234533758192,
      "flos": 17455064192640.0,
      "grad_norm": 4.678868554787467,
      "language_loss": 0.84767222,
      "learning_rate": 3.6552724703193855e-06,
      "loss": 0.87037933,
      "num_input_tokens_seen": 37892560,
      "step": 1780,
      "time_per_iteration": 4.209951162338257
    },
    {
      "auxiliary_loss_clip": 0.01079596,
      "auxiliary_loss_mlp": 0.01000735,
      "balance_loss_clip": 1.02460492,
      "balance_loss_mlp": 0.9983986,
      "epoch": 0.214152588228221,
      "flos": 51637606686720.0,
      "grad_norm": 0.8062742582829387,
      "language_loss": 0.55931163,
      "learning_rate": 3.654835135352801e-06,
      "loss": 0.58011496,
      "num_input_tokens_seen": 37947370,
      "step": 1781,
      "time_per_iteration": 4.233088970184326
    },
    {
      "auxiliary_loss_clip": 0.01198945,
      "auxiliary_loss_mlp": 0.01037646,
      "balance_loss_clip": 1.06726766,
      "balance_loss_mlp": 1.02587414,
      "epoch": 0.21427283111886009,
      "flos": 19496154625920.0,
      "grad_norm": 3.8000742385319617,
      "language_loss": 0.8746168,
      "learning_rate": 3.654397549349043e-06,
      "loss": 0.89698279,
      "num_input_tokens_seen": 37964745,
      "step": 1782,
      "time_per_iteration": 2.794928550720215
    },
    {
      "auxiliary_loss_clip": 0.01199391,
      "auxiliary_loss_mlp": 0.01044959,
      "balance_loss_clip": 1.06951284,
      "balance_loss_mlp": 1.0341413,
      "epoch": 0.2143930740094992,
      "flos": 20084802710400.0,
      "grad_norm": 2.246482737803333,
      "language_loss": 0.75800431,
      "learning_rate": 3.653959712374491e-06,
      "loss": 0.78044784,
      "num_input_tokens_seen": 37982850,
      "step": 1783,
      "time_per_iteration": 2.794926643371582
    },
    {
      "auxiliary_loss_clip": 0.01165538,
      "auxiliary_loss_mlp": 0.01041996,
      "balance_loss_clip": 1.06125236,
      "balance_loss_mlp": 1.03120756,
      "epoch": 0.21451331690013828,
      "flos": 21798603394560.0,
      "grad_norm": 2.320884052571922,
      "language_loss": 0.82970548,
      "learning_rate": 3.6535216244955663e-06,
      "loss": 0.85178077,
      "num_input_tokens_seen": 38002745,
      "step": 1784,
      "time_per_iteration": 2.952995538711548
    },
    {
      "auxiliary_loss_clip": 0.01196649,
      "auxiliary_loss_mlp": 0.0103552,
      "balance_loss_clip": 1.06764662,
      "balance_loss_mlp": 1.02514911,
      "epoch": 0.21463355979077736,
      "flos": 32853882412800.0,
      "grad_norm": 2.4839682653371264,
      "language_loss": 0.70553064,
      "learning_rate": 3.653083285778726e-06,
      "loss": 0.72785223,
      "num_input_tokens_seen": 38024115,
      "step": 1785,
      "time_per_iteration": 2.8482627868652344
    },
    {
      "auxiliary_loss_clip": 0.01200002,
      "auxiliary_loss_mlp": 0.01043263,
      "balance_loss_clip": 1.0637536,
      "balance_loss_mlp": 1.03170013,
      "epoch": 0.21475380268141647,
      "flos": 21543817248000.0,
      "grad_norm": 5.933095342986092,
      "language_loss": 0.81367105,
      "learning_rate": 3.6526446962904653e-06,
      "loss": 0.83610368,
      "num_input_tokens_seen": 38042830,
      "step": 1786,
      "time_per_iteration": 3.9990382194519043
    },
    {
      "auxiliary_loss_clip": 0.01212999,
      "auxiliary_loss_mlp": 0.0105034,
      "balance_loss_clip": 1.07101691,
      "balance_loss_mlp": 1.04045796,
      "epoch": 0.21487404557205556,
      "flos": 32159082660480.0,
      "grad_norm": 1.6903888999008958,
      "language_loss": 0.74280125,
      "learning_rate": 3.652205856097318e-06,
      "loss": 0.76543462,
      "num_input_tokens_seen": 38066015,
      "step": 1787,
      "time_per_iteration": 2.8634417057037354
    },
    {
      "auxiliary_loss_clip": 0.01229249,
      "auxiliary_loss_mlp": 0.01039238,
      "balance_loss_clip": 1.07002592,
      "balance_loss_mlp": 1.0277462,
      "epoch": 0.21499428846269464,
      "flos": 12673091583360.0,
      "grad_norm": 2.4618687337768734,
      "language_loss": 0.79679072,
      "learning_rate": 3.651766765265856e-06,
      "loss": 0.81947559,
      "num_input_tokens_seen": 38083025,
      "step": 1788,
      "time_per_iteration": 2.613983631134033
    },
    {
      "auxiliary_loss_clip": 0.01217235,
      "auxiliary_loss_mlp": 0.01040165,
      "balance_loss_clip": 1.0669018,
      "balance_loss_mlp": 1.02974653,
      "epoch": 0.21511453135333372,
      "flos": 23471573293440.0,
      "grad_norm": 2.5216547198942685,
      "language_loss": 0.80998635,
      "learning_rate": 3.65132742386269e-06,
      "loss": 0.83256042,
      "num_input_tokens_seen": 38098245,
      "step": 1789,
      "time_per_iteration": 2.733802080154419
    },
    {
      "auxiliary_loss_clip": 0.01205832,
      "auxiliary_loss_mlp": 0.00764724,
      "balance_loss_clip": 1.06778073,
      "balance_loss_mlp": 1.00053787,
      "epoch": 0.21523477424397283,
      "flos": 26943560893440.0,
      "grad_norm": 1.9665940421353088,
      "language_loss": 0.84882671,
      "learning_rate": 3.6508878319544656e-06,
      "loss": 0.8685323,
      "num_input_tokens_seen": 38118460,
      "step": 1790,
      "time_per_iteration": 2.7762420177459717
    },
    {
      "auxiliary_loss_clip": 0.01233888,
      "auxiliary_loss_mlp": 0.01050012,
      "balance_loss_clip": 1.07565546,
      "balance_loss_mlp": 1.0395807,
      "epoch": 0.21535501713461191,
      "flos": 18916161719040.0,
      "grad_norm": 6.230122728749084,
      "language_loss": 0.81766438,
      "learning_rate": 3.65044798960787e-06,
      "loss": 0.84050345,
      "num_input_tokens_seen": 38136800,
      "step": 1791,
      "time_per_iteration": 2.7166764736175537
    },
    {
      "auxiliary_loss_clip": 0.01152078,
      "auxiliary_loss_mlp": 0.01044743,
      "balance_loss_clip": 1.05591774,
      "balance_loss_mlp": 1.03456831,
      "epoch": 0.215475260025251,
      "flos": 17895113712000.0,
      "grad_norm": 4.045406431413641,
      "language_loss": 0.78272331,
      "learning_rate": 3.650007896889627e-06,
      "loss": 0.80469149,
      "num_input_tokens_seen": 38155380,
      "step": 1792,
      "time_per_iteration": 2.866985559463501
    },
    {
      "auxiliary_loss_clip": 0.01228101,
      "auxiliary_loss_mlp": 0.01038734,
      "balance_loss_clip": 1.06993937,
      "balance_loss_mlp": 1.02779651,
      "epoch": 0.2155955029158901,
      "flos": 16654292340480.0,
      "grad_norm": 3.1497440427701626,
      "language_loss": 0.80929244,
      "learning_rate": 3.6495675538664974e-06,
      "loss": 0.83196074,
      "num_input_tokens_seen": 38174395,
      "step": 1793,
      "time_per_iteration": 2.766761302947998
    },
    {
      "auxiliary_loss_clip": 0.01218455,
      "auxiliary_loss_mlp": 0.00764671,
      "balance_loss_clip": 1.06912398,
      "balance_loss_mlp": 1.0006485,
      "epoch": 0.2157157458065292,
      "flos": 23621213352960.0,
      "grad_norm": 1.9347681787919357,
      "language_loss": 0.82301736,
      "learning_rate": 3.649126960605282e-06,
      "loss": 0.8428486,
      "num_input_tokens_seen": 38195380,
      "step": 1794,
      "time_per_iteration": 2.775775194168091
    },
    {
      "auxiliary_loss_clip": 0.0118614,
      "auxiliary_loss_mlp": 0.01040906,
      "balance_loss_clip": 1.06146455,
      "balance_loss_mlp": 1.03008747,
      "epoch": 0.21583598869716827,
      "flos": 22127078292480.0,
      "grad_norm": 6.208016296168433,
      "language_loss": 0.83981931,
      "learning_rate": 3.6486861171728174e-06,
      "loss": 0.86208975,
      "num_input_tokens_seen": 38213775,
      "step": 1795,
      "time_per_iteration": 2.688617706298828
    },
    {
      "auxiliary_loss_clip": 0.01182915,
      "auxiliary_loss_mlp": 0.01040016,
      "balance_loss_clip": 1.06776822,
      "balance_loss_mlp": 1.03019941,
      "epoch": 0.21595623158780738,
      "flos": 23441229279360.0,
      "grad_norm": 1.8348146326802262,
      "language_loss": 0.78626168,
      "learning_rate": 3.6482450236359803e-06,
      "loss": 0.80849105,
      "num_input_tokens_seen": 38235630,
      "step": 1796,
      "time_per_iteration": 2.780155658721924
    },
    {
      "auxiliary_loss_clip": 0.01162767,
      "auxiliary_loss_mlp": 0.01037105,
      "balance_loss_clip": 1.06576383,
      "balance_loss_mlp": 1.02698398,
      "epoch": 0.21607647447844647,
      "flos": 26906501036160.0,
      "grad_norm": 2.2377783456791196,
      "language_loss": 0.77817446,
      "learning_rate": 3.647803680061683e-06,
      "loss": 0.80017316,
      "num_input_tokens_seen": 38256045,
      "step": 1797,
      "time_per_iteration": 2.842031717300415
    },
    {
      "auxiliary_loss_clip": 0.01152146,
      "auxiliary_loss_mlp": 0.01034351,
      "balance_loss_clip": 1.05813503,
      "balance_loss_mlp": 1.02380145,
      "epoch": 0.21619671736908555,
      "flos": 14495378319360.0,
      "grad_norm": 2.753744639286658,
      "language_loss": 0.75125438,
      "learning_rate": 3.6473620865168776e-06,
      "loss": 0.77311933,
      "num_input_tokens_seen": 38272915,
      "step": 1798,
      "time_per_iteration": 2.742732048034668
    },
    {
      "auxiliary_loss_clip": 0.012016,
      "auxiliary_loss_mlp": 0.01037646,
      "balance_loss_clip": 1.07103312,
      "balance_loss_mlp": 1.02737069,
      "epoch": 0.21631696025972463,
      "flos": 17931096161280.0,
      "grad_norm": 2.1928333483833065,
      "language_loss": 0.81526607,
      "learning_rate": 3.646920243068554e-06,
      "loss": 0.83765852,
      "num_input_tokens_seen": 38290810,
      "step": 1799,
      "time_per_iteration": 2.6772677898406982
    },
    {
      "auxiliary_loss_clip": 0.01197855,
      "auxiliary_loss_mlp": 0.0103963,
      "balance_loss_clip": 1.06684506,
      "balance_loss_mlp": 1.02932429,
      "epoch": 0.21643720315036374,
      "flos": 24462385027200.0,
      "grad_norm": 2.094605210520824,
      "language_loss": 0.7447902,
      "learning_rate": 3.6464781497837384e-06,
      "loss": 0.76716506,
      "num_input_tokens_seen": 38312785,
      "step": 1800,
      "time_per_iteration": 2.7708394527435303
    },
    {
      "auxiliary_loss_clip": 0.01150913,
      "auxiliary_loss_mlp": 0.01052515,
      "balance_loss_clip": 1.05725408,
      "balance_loss_mlp": 1.04201269,
      "epoch": 0.21655744604100283,
      "flos": 28474432588800.0,
      "grad_norm": 1.8728244926332696,
      "language_loss": 0.72984415,
      "learning_rate": 3.6460358067294965e-06,
      "loss": 0.75187844,
      "num_input_tokens_seen": 38334015,
      "step": 1801,
      "time_per_iteration": 2.847959518432617
    },
    {
      "auxiliary_loss_clip": 0.01190077,
      "auxiliary_loss_mlp": 0.01042208,
      "balance_loss_clip": 1.06018949,
      "balance_loss_mlp": 1.03106189,
      "epoch": 0.2166776889316419,
      "flos": 20152960767360.0,
      "grad_norm": 3.1603101821151327,
      "language_loss": 0.77878523,
      "learning_rate": 3.645593213972932e-06,
      "loss": 0.80110806,
      "num_input_tokens_seen": 38352920,
      "step": 1802,
      "time_per_iteration": 2.740384578704834
    },
    {
      "auxiliary_loss_clip": 0.01219984,
      "auxiliary_loss_mlp": 0.01043353,
      "balance_loss_clip": 1.07235312,
      "balance_loss_mlp": 1.03261876,
      "epoch": 0.21679793182228102,
      "flos": 15193482122880.0,
      "grad_norm": 4.954866392540916,
      "language_loss": 0.80353761,
      "learning_rate": 3.6451503715811852e-06,
      "loss": 0.82617098,
      "num_input_tokens_seen": 38371230,
      "step": 1803,
      "time_per_iteration": 2.6951045989990234
    },
    {
      "auxiliary_loss_clip": 0.01227034,
      "auxiliary_loss_mlp": 0.01039992,
      "balance_loss_clip": 1.07161307,
      "balance_loss_mlp": 1.03001392,
      "epoch": 0.2169181747129201,
      "flos": 17384464010880.0,
      "grad_norm": 2.2750723620298605,
      "language_loss": 0.80223119,
      "learning_rate": 3.6447072796214345e-06,
      "loss": 0.82490146,
      "num_input_tokens_seen": 38389795,
      "step": 1804,
      "time_per_iteration": 2.664896249771118
    },
    {
      "auxiliary_loss_clip": 0.01072259,
      "auxiliary_loss_mlp": 0.01011737,
      "balance_loss_clip": 1.02680182,
      "balance_loss_mlp": 1.00945997,
      "epoch": 0.21703841760355919,
      "flos": 58760955429120.0,
      "grad_norm": 0.9606054811681773,
      "language_loss": 0.63133383,
      "learning_rate": 3.644263938160898e-06,
      "loss": 0.65217376,
      "num_input_tokens_seen": 38445760,
      "step": 1805,
      "time_per_iteration": 4.127001523971558
    },
    {
      "auxiliary_loss_clip": 0.01203094,
      "auxiliary_loss_mlp": 0.01047055,
      "balance_loss_clip": 1.06868172,
      "balance_loss_mlp": 1.03643322,
      "epoch": 0.21715866049419827,
      "flos": 22418457419520.0,
      "grad_norm": 2.0037809479291835,
      "language_loss": 0.71976745,
      "learning_rate": 3.6438203472668293e-06,
      "loss": 0.74226892,
      "num_input_tokens_seen": 38465405,
      "step": 1806,
      "time_per_iteration": 4.1510090827941895
    },
    {
      "auxiliary_loss_clip": 0.01195588,
      "auxiliary_loss_mlp": 0.01045487,
      "balance_loss_clip": 1.06524277,
      "balance_loss_mlp": 1.03584886,
      "epoch": 0.21727890338483738,
      "flos": 17237732952960.0,
      "grad_norm": 5.480395143800151,
      "language_loss": 0.81643301,
      "learning_rate": 3.6433765070065206e-06,
      "loss": 0.83884376,
      "num_input_tokens_seen": 38483195,
      "step": 1807,
      "time_per_iteration": 3.6029956340789795
    },
    {
      "auxiliary_loss_clip": 0.01215441,
      "auxiliary_loss_mlp": 0.01050505,
      "balance_loss_clip": 1.06534207,
      "balance_loss_mlp": 1.03935933,
      "epoch": 0.21739914627547646,
      "flos": 13434792416640.0,
      "grad_norm": 10.079797739521508,
      "language_loss": 0.87586784,
      "learning_rate": 3.6429324174473025e-06,
      "loss": 0.8985272,
      "num_input_tokens_seen": 38496735,
      "step": 1808,
      "time_per_iteration": 2.653534173965454
    },
    {
      "auxiliary_loss_clip": 0.01197687,
      "auxiliary_loss_mlp": 0.01040375,
      "balance_loss_clip": 1.06144083,
      "balance_loss_mlp": 1.02890158,
      "epoch": 0.21751938916611555,
      "flos": 20959514709120.0,
      "grad_norm": 2.1382882581449207,
      "language_loss": 0.85099262,
      "learning_rate": 3.6424880786565425e-06,
      "loss": 0.87337327,
      "num_input_tokens_seen": 38512880,
      "step": 1809,
      "time_per_iteration": 2.7571959495544434
    },
    {
      "auxiliary_loss_clip": 0.01174097,
      "auxiliary_loss_mlp": 0.01044154,
      "balance_loss_clip": 1.0629909,
      "balance_loss_mlp": 1.03330553,
      "epoch": 0.21763963205675466,
      "flos": 27599936071680.0,
      "grad_norm": 38.67785283942538,
      "language_loss": 0.7971158,
      "learning_rate": 3.6420434907016482e-06,
      "loss": 0.81929827,
      "num_input_tokens_seen": 38532570,
      "step": 1810,
      "time_per_iteration": 2.85209059715271
    },
    {
      "auxiliary_loss_clip": 0.01197776,
      "auxiliary_loss_mlp": 0.01037952,
      "balance_loss_clip": 1.06937492,
      "balance_loss_mlp": 1.02792668,
      "epoch": 0.21775987494739374,
      "flos": 21430411032960.0,
      "grad_norm": 2.498709899768392,
      "language_loss": 0.81125319,
      "learning_rate": 3.6415986536500606e-06,
      "loss": 0.83361053,
      "num_input_tokens_seen": 38550900,
      "step": 1811,
      "time_per_iteration": 2.790457010269165
    },
    {
      "auxiliary_loss_clip": 0.01182422,
      "auxiliary_loss_mlp": 0.01046949,
      "balance_loss_clip": 1.06467795,
      "balance_loss_mlp": 1.03695345,
      "epoch": 0.21788011783803282,
      "flos": 18332972501760.0,
      "grad_norm": 3.946370317636097,
      "language_loss": 0.80714393,
      "learning_rate": 3.641153567569263e-06,
      "loss": 0.82943761,
      "num_input_tokens_seen": 38569215,
      "step": 1812,
      "time_per_iteration": 3.755613088607788
    },
    {
      "auxiliary_loss_clip": 0.01132881,
      "auxiliary_loss_mlp": 0.01041549,
      "balance_loss_clip": 1.0544225,
      "balance_loss_mlp": 1.03156471,
      "epoch": 0.2180003607286719,
      "flos": 30262748037120.0,
      "grad_norm": 2.4155339883190687,
      "language_loss": 0.95448488,
      "learning_rate": 3.640708232526774e-06,
      "loss": 0.97622913,
      "num_input_tokens_seen": 38587870,
      "step": 1813,
      "time_per_iteration": 2.987138032913208
    },
    {
      "auxiliary_loss_clip": 0.01212846,
      "auxiliary_loss_mlp": 0.01044173,
      "balance_loss_clip": 1.06539857,
      "balance_loss_mlp": 1.03323555,
      "epoch": 0.21812060361931102,
      "flos": 25480272637440.0,
      "grad_norm": 1.9893793236694115,
      "language_loss": 0.78704059,
      "learning_rate": 3.6402626485901504e-06,
      "loss": 0.80961084,
      "num_input_tokens_seen": 38606965,
      "step": 1814,
      "time_per_iteration": 2.7351737022399902
    },
    {
      "auxiliary_loss_clip": 0.01162636,
      "auxiliary_loss_mlp": 0.01039818,
      "balance_loss_clip": 1.06395936,
      "balance_loss_mlp": 1.02944112,
      "epoch": 0.2182408465099501,
      "flos": 21908166854400.0,
      "grad_norm": 3.1686869431169624,
      "language_loss": 0.78492361,
      "learning_rate": 3.639816815826988e-06,
      "loss": 0.80694813,
      "num_input_tokens_seen": 38626290,
      "step": 1815,
      "time_per_iteration": 2.79482364654541
    },
    {
      "auxiliary_loss_clip": 0.01195218,
      "auxiliary_loss_mlp": 0.01034206,
      "balance_loss_clip": 1.06383419,
      "balance_loss_mlp": 1.02403188,
      "epoch": 0.21836108940058918,
      "flos": 23657339456640.0,
      "grad_norm": 2.622042856084782,
      "language_loss": 0.78110737,
      "learning_rate": 3.6393707343049176e-06,
      "loss": 0.80340165,
      "num_input_tokens_seen": 38646620,
      "step": 1816,
      "time_per_iteration": 2.7881336212158203
    },
    {
      "auxiliary_loss_clip": 0.01183556,
      "auxiliary_loss_mlp": 0.01035418,
      "balance_loss_clip": 1.06774545,
      "balance_loss_mlp": 1.0260005,
      "epoch": 0.2184813322912283,
      "flos": 24681009156480.0,
      "grad_norm": 2.5386598566184047,
      "language_loss": 0.73182762,
      "learning_rate": 3.6389244040916104e-06,
      "loss": 0.75401735,
      "num_input_tokens_seen": 38665695,
      "step": 1817,
      "time_per_iteration": 2.7695257663726807
    },
    {
      "auxiliary_loss_clip": 0.01203431,
      "auxiliary_loss_mlp": 0.01039802,
      "balance_loss_clip": 1.06742346,
      "balance_loss_mlp": 1.02968752,
      "epoch": 0.21860157518186737,
      "flos": 26574650259840.0,
      "grad_norm": 2.2503652235458746,
      "language_loss": 0.79129291,
      "learning_rate": 3.6384778252547747e-06,
      "loss": 0.81372523,
      "num_input_tokens_seen": 38681575,
      "step": 1818,
      "time_per_iteration": 2.813512086868286
    },
    {
      "auxiliary_loss_clip": 0.01169371,
      "auxiliary_loss_mlp": 0.01040494,
      "balance_loss_clip": 1.0649302,
      "balance_loss_mlp": 1.03074896,
      "epoch": 0.21872181807250646,
      "flos": 20886292834560.0,
      "grad_norm": 2.18439803307691,
      "language_loss": 0.78178841,
      "learning_rate": 3.638030997862155e-06,
      "loss": 0.80388707,
      "num_input_tokens_seen": 38700510,
      "step": 1819,
      "time_per_iteration": 2.92004656791687
    },
    {
      "auxiliary_loss_clip": 0.01096649,
      "auxiliary_loss_mlp": 0.01001472,
      "balance_loss_clip": 1.02491736,
      "balance_loss_mlp": 0.99939805,
      "epoch": 0.21884206096314554,
      "flos": 61209452897280.0,
      "grad_norm": 0.7626711801014292,
      "language_loss": 0.59372717,
      "learning_rate": 3.6375839219815356e-06,
      "loss": 0.61470836,
      "num_input_tokens_seen": 38758310,
      "step": 1820,
      "time_per_iteration": 3.2593328952789307
    },
    {
      "auxiliary_loss_clip": 0.0116852,
      "auxiliary_loss_mlp": 0.01033044,
      "balance_loss_clip": 1.06262672,
      "balance_loss_mlp": 1.02241063,
      "epoch": 0.21896230385378465,
      "flos": 23473835850240.0,
      "grad_norm": 2.248693575049806,
      "language_loss": 0.82804483,
      "learning_rate": 3.6371365976807375e-06,
      "loss": 0.85006046,
      "num_input_tokens_seen": 38778705,
      "step": 1821,
      "time_per_iteration": 2.915804147720337
    },
    {
      "auxiliary_loss_clip": 0.01203777,
      "auxiliary_loss_mlp": 0.01043636,
      "balance_loss_clip": 1.06685042,
      "balance_loss_mlp": 1.03340805,
      "epoch": 0.21908254674442373,
      "flos": 25081915829760.0,
      "grad_norm": 3.479819886642675,
      "language_loss": 0.83579123,
      "learning_rate": 3.6366890250276185e-06,
      "loss": 0.8582654,
      "num_input_tokens_seen": 38799660,
      "step": 1822,
      "time_per_iteration": 2.7419216632843018
    },
    {
      "auxiliary_loss_clip": 0.01169688,
      "auxiliary_loss_mlp": 0.01039164,
      "balance_loss_clip": 1.05695987,
      "balance_loss_mlp": 1.02808928,
      "epoch": 0.21920278963506282,
      "flos": 23513768795520.0,
      "grad_norm": 2.030250142711635,
      "language_loss": 0.90174973,
      "learning_rate": 3.6362412040900764e-06,
      "loss": 0.92383826,
      "num_input_tokens_seen": 38819450,
      "step": 1823,
      "time_per_iteration": 2.847809076309204
    },
    {
      "auxiliary_loss_clip": 0.01209972,
      "auxiliary_loss_mlp": 0.01040238,
      "balance_loss_clip": 1.06891763,
      "balance_loss_mlp": 1.02946699,
      "epoch": 0.21932303252570193,
      "flos": 29242238734080.0,
      "grad_norm": 2.6966724015062855,
      "language_loss": 0.80699813,
      "learning_rate": 3.635793134936044e-06,
      "loss": 0.8295002,
      "num_input_tokens_seen": 38840460,
      "step": 1824,
      "time_per_iteration": 2.747342586517334
    },
    {
      "auxiliary_loss_clip": 0.01223351,
      "auxiliary_loss_mlp": 0.01042474,
      "balance_loss_clip": 1.0685823,
      "balance_loss_mlp": 1.0324012,
      "epoch": 0.219443275416341,
      "flos": 20806857907200.0,
      "grad_norm": 2.3540992250694175,
      "language_loss": 0.72977376,
      "learning_rate": 3.635344817633494e-06,
      "loss": 0.75243199,
      "num_input_tokens_seen": 38859775,
      "step": 1825,
      "time_per_iteration": 2.6757712364196777
    },
    {
      "auxiliary_loss_clip": 0.01135776,
      "auxiliary_loss_mlp": 0.0103401,
      "balance_loss_clip": 1.05371428,
      "balance_loss_mlp": 1.02431226,
      "epoch": 0.2195635183069801,
      "flos": 14501555458560.0,
      "grad_norm": 3.3496333224480623,
      "language_loss": 0.75881952,
      "learning_rate": 3.634896252250436e-06,
      "loss": 0.7805174,
      "num_input_tokens_seen": 38876540,
      "step": 1826,
      "time_per_iteration": 2.8061296939849854
    },
    {
      "auxiliary_loss_clip": 0.01170943,
      "auxiliary_loss_mlp": 0.01042568,
      "balance_loss_clip": 1.0620532,
      "balance_loss_mlp": 1.03201222,
      "epoch": 0.2196837611976192,
      "flos": 24243473589120.0,
      "grad_norm": 2.4480166627964226,
      "language_loss": 0.82550997,
      "learning_rate": 3.6344474388549157e-06,
      "loss": 0.84764504,
      "num_input_tokens_seen": 38896195,
      "step": 1827,
      "time_per_iteration": 2.8221819400787354
    },
    {
      "auxiliary_loss_clip": 0.011703,
      "auxiliary_loss_mlp": 0.01040096,
      "balance_loss_clip": 1.06636381,
      "balance_loss_mlp": 1.0298624,
      "epoch": 0.2198040040882583,
      "flos": 18074523168000.0,
      "grad_norm": 5.481934328561655,
      "language_loss": 0.80535495,
      "learning_rate": 3.6339983775150183e-06,
      "loss": 0.82745898,
      "num_input_tokens_seen": 38912755,
      "step": 1828,
      "time_per_iteration": 2.810194730758667
    },
    {
      "auxiliary_loss_clip": 0.01198034,
      "auxiliary_loss_mlp": 0.01036823,
      "balance_loss_clip": 1.0635128,
      "balance_loss_mlp": 1.02633905,
      "epoch": 0.21992424697889737,
      "flos": 17784185535360.0,
      "grad_norm": 2.86349249892814,
      "language_loss": 0.84250867,
      "learning_rate": 3.6335490682988664e-06,
      "loss": 0.86485732,
      "num_input_tokens_seen": 38928365,
      "step": 1829,
      "time_per_iteration": 2.769172191619873
    },
    {
      "auxiliary_loss_clip": 0.01197126,
      "auxiliary_loss_mlp": 0.01041062,
      "balance_loss_clip": 1.06989574,
      "balance_loss_mlp": 1.03033352,
      "epoch": 0.22004448986953645,
      "flos": 17638495971840.0,
      "grad_norm": 2.072696661770759,
      "language_loss": 0.83102268,
      "learning_rate": 3.63309951127462e-06,
      "loss": 0.85340458,
      "num_input_tokens_seen": 38945275,
      "step": 1830,
      "time_per_iteration": 2.7377588748931885
    },
    {
      "auxiliary_loss_clip": 0.01182962,
      "auxiliary_loss_mlp": 0.01043277,
      "balance_loss_clip": 1.06067228,
      "balance_loss_mlp": 1.03201795,
      "epoch": 0.22016473276017556,
      "flos": 22275533203200.0,
      "grad_norm": 5.608688401279056,
      "language_loss": 0.76012224,
      "learning_rate": 3.6326497065104757e-06,
      "loss": 0.78238463,
      "num_input_tokens_seen": 38965740,
      "step": 1831,
      "time_per_iteration": 3.7785606384277344
    },
    {
      "auxiliary_loss_clip": 0.01182596,
      "auxiliary_loss_mlp": 0.01048524,
      "balance_loss_clip": 1.06467068,
      "balance_loss_mlp": 1.0374198,
      "epoch": 0.22028497565081465,
      "flos": 25556259859200.0,
      "grad_norm": 2.6511636649326054,
      "language_loss": 0.77995074,
      "learning_rate": 3.6321996540746697e-06,
      "loss": 0.80226195,
      "num_input_tokens_seen": 38984815,
      "step": 1832,
      "time_per_iteration": 4.297353744506836
    },
    {
      "auxiliary_loss_clip": 0.01199073,
      "auxiliary_loss_mlp": 0.0103466,
      "balance_loss_clip": 1.06466341,
      "balance_loss_mlp": 1.02473021,
      "epoch": 0.22040521854145373,
      "flos": 36247332925440.0,
      "grad_norm": 1.8364659936933956,
      "language_loss": 0.80574799,
      "learning_rate": 3.6317493540354733e-06,
      "loss": 0.8280853,
      "num_input_tokens_seen": 39008230,
      "step": 1833,
      "time_per_iteration": 3.8074190616607666
    },
    {
      "auxiliary_loss_clip": 0.01167063,
      "auxiliary_loss_mlp": 0.00764379,
      "balance_loss_clip": 1.06560481,
      "balance_loss_mlp": 1.0004952,
      "epoch": 0.22052546143209284,
      "flos": 11838420270720.0,
      "grad_norm": 2.1428138993691603,
      "language_loss": 0.77073878,
      "learning_rate": 3.6312988064611976e-06,
      "loss": 0.79005313,
      "num_input_tokens_seen": 39026540,
      "step": 1834,
      "time_per_iteration": 2.925872802734375
    },
    {
      "auxiliary_loss_clip": 0.01209699,
      "auxiliary_loss_mlp": 0.00764781,
      "balance_loss_clip": 1.06368256,
      "balance_loss_mlp": 1.00045657,
      "epoch": 0.22064570432273192,
      "flos": 24209250906240.0,
      "grad_norm": 6.466176901301433,
      "language_loss": 0.81484079,
      "learning_rate": 3.6308480114201896e-06,
      "loss": 0.83458567,
      "num_input_tokens_seen": 39048460,
      "step": 1835,
      "time_per_iteration": 2.7445926666259766
    },
    {
      "auxiliary_loss_clip": 0.01210328,
      "auxiliary_loss_mlp": 0.01033574,
      "balance_loss_clip": 1.06635547,
      "balance_loss_mlp": 1.0235244,
      "epoch": 0.220765947213371,
      "flos": 17931347556480.0,
      "grad_norm": 1.9838078129840129,
      "language_loss": 0.76275122,
      "learning_rate": 3.630396968980835e-06,
      "loss": 0.78519022,
      "num_input_tokens_seen": 39066335,
      "step": 1836,
      "time_per_iteration": 2.6856448650360107
    },
    {
      "auxiliary_loss_clip": 0.01226106,
      "auxiliary_loss_mlp": 0.0104854,
      "balance_loss_clip": 1.06842577,
      "balance_loss_mlp": 1.03795385,
      "epoch": 0.2208861901040101,
      "flos": 26757040544640.0,
      "grad_norm": 2.780522266381842,
      "language_loss": 0.83698022,
      "learning_rate": 3.6299456792115575e-06,
      "loss": 0.85972673,
      "num_input_tokens_seen": 39087590,
      "step": 1837,
      "time_per_iteration": 3.598675489425659
    },
    {
      "auxiliary_loss_clip": 0.01169545,
      "auxiliary_loss_mlp": 0.01046295,
      "balance_loss_clip": 1.06294179,
      "balance_loss_mlp": 1.03606057,
      "epoch": 0.2210064329946492,
      "flos": 17817977255040.0,
      "grad_norm": 4.529380801312879,
      "language_loss": 0.81160855,
      "learning_rate": 3.629494142180815e-06,
      "loss": 0.83376694,
      "num_input_tokens_seen": 39106335,
      "step": 1838,
      "time_per_iteration": 2.8042120933532715
    },
    {
      "auxiliary_loss_clip": 0.01209276,
      "auxiliary_loss_mlp": 0.01043126,
      "balance_loss_clip": 1.06872022,
      "balance_loss_mlp": 1.03236747,
      "epoch": 0.22112667588528828,
      "flos": 17967401832960.0,
      "grad_norm": 2.938145827782744,
      "language_loss": 0.85392296,
      "learning_rate": 3.6290423579571075e-06,
      "loss": 0.87644696,
      "num_input_tokens_seen": 39122875,
      "step": 1839,
      "time_per_iteration": 2.6500608921051025
    },
    {
      "auxiliary_loss_clip": 0.01213217,
      "auxiliary_loss_mlp": 0.01043201,
      "balance_loss_clip": 1.06427097,
      "balance_loss_mlp": 1.0318042,
      "epoch": 0.22124691877592736,
      "flos": 18369206346240.0,
      "grad_norm": 1.811448415037635,
      "language_loss": 0.80657268,
      "learning_rate": 3.6285903266089694e-06,
      "loss": 0.82913685,
      "num_input_tokens_seen": 39142150,
      "step": 1840,
      "time_per_iteration": 2.6790645122528076
    },
    {
      "auxiliary_loss_clip": 0.01166521,
      "auxiliary_loss_mlp": 0.01051499,
      "balance_loss_clip": 1.05993819,
      "balance_loss_mlp": 1.04043627,
      "epoch": 0.22136716166656648,
      "flos": 20813286441600.0,
      "grad_norm": 2.2504805310330562,
      "language_loss": 0.77267683,
      "learning_rate": 3.628138048204974e-06,
      "loss": 0.79485703,
      "num_input_tokens_seen": 39162835,
      "step": 1841,
      "time_per_iteration": 2.8559296131134033
    },
    {
      "auxiliary_loss_clip": 0.01156635,
      "auxiliary_loss_mlp": 0.01048888,
      "balance_loss_clip": 1.05576575,
      "balance_loss_mlp": 1.03792655,
      "epoch": 0.22148740455720556,
      "flos": 17675699483520.0,
      "grad_norm": 1.8472142203362196,
      "language_loss": 0.76261073,
      "learning_rate": 3.6276855228137304e-06,
      "loss": 0.78466594,
      "num_input_tokens_seen": 39181040,
      "step": 1842,
      "time_per_iteration": 2.8081138134002686
    },
    {
      "auxiliary_loss_clip": 0.01203536,
      "auxiliary_loss_mlp": 0.01042169,
      "balance_loss_clip": 1.06603312,
      "balance_loss_mlp": 1.03151798,
      "epoch": 0.22160764744784464,
      "flos": 21726710323200.0,
      "grad_norm": 2.929472513973111,
      "language_loss": 0.8227666,
      "learning_rate": 3.6272327505038874e-06,
      "loss": 0.84522367,
      "num_input_tokens_seen": 39197505,
      "step": 1843,
      "time_per_iteration": 2.8550267219543457
    },
    {
      "auxiliary_loss_clip": 0.01167045,
      "auxiliary_loss_mlp": 0.01040853,
      "balance_loss_clip": 1.06081676,
      "balance_loss_mlp": 1.03133976,
      "epoch": 0.22172789033848372,
      "flos": 23764712186880.0,
      "grad_norm": 2.224778844487863,
      "language_loss": 0.78139806,
      "learning_rate": 3.626779731344131e-06,
      "loss": 0.80347717,
      "num_input_tokens_seen": 39217295,
      "step": 1844,
      "time_per_iteration": 2.7441885471343994
    },
    {
      "auxiliary_loss_clip": 0.01191469,
      "auxiliary_loss_mlp": 0.01033447,
      "balance_loss_clip": 1.06304979,
      "balance_loss_mlp": 1.02290893,
      "epoch": 0.22184813322912283,
      "flos": 16982300361600.0,
      "grad_norm": 2.152586197473668,
      "language_loss": 0.85692585,
      "learning_rate": 3.6263264654031814e-06,
      "loss": 0.87917507,
      "num_input_tokens_seen": 39234195,
      "step": 1845,
      "time_per_iteration": 2.7003023624420166
    },
    {
      "auxiliary_loss_clip": 0.01083845,
      "auxiliary_loss_mlp": 0.01007946,
      "balance_loss_clip": 1.02805352,
      "balance_loss_mlp": 1.00585985,
      "epoch": 0.22196837611976192,
      "flos": 61823740314240.0,
      "grad_norm": 0.7142019180349632,
      "language_loss": 0.59080458,
      "learning_rate": 3.6258729527498008e-06,
      "loss": 0.61172247,
      "num_input_tokens_seen": 39295040,
      "step": 1846,
      "time_per_iteration": 3.2560014724731445
    },
    {
      "auxiliary_loss_clip": 0.01180473,
      "auxiliary_loss_mlp": 0.01039326,
      "balance_loss_clip": 1.06503284,
      "balance_loss_mlp": 1.02885365,
      "epoch": 0.222088619010401,
      "flos": 25558019625600.0,
      "grad_norm": 2.521594892689101,
      "language_loss": 0.65102983,
      "learning_rate": 3.6254191934527854e-06,
      "loss": 0.67322785,
      "num_input_tokens_seen": 39314395,
      "step": 1847,
      "time_per_iteration": 2.7808592319488525
    },
    {
      "auxiliary_loss_clip": 0.01216108,
      "auxiliary_loss_mlp": 0.01050255,
      "balance_loss_clip": 1.0708164,
      "balance_loss_mlp": 1.03878725,
      "epoch": 0.2222088619010401,
      "flos": 19318612677120.0,
      "grad_norm": 2.1500835948696357,
      "language_loss": 0.65167141,
      "learning_rate": 3.6249651875809715e-06,
      "loss": 0.67433506,
      "num_input_tokens_seen": 39334275,
      "step": 1848,
      "time_per_iteration": 2.7061266899108887
    },
    {
      "auxiliary_loss_clip": 0.01170246,
      "auxiliary_loss_mlp": 0.00764123,
      "balance_loss_clip": 1.06124806,
      "balance_loss_mlp": 1.0006001,
      "epoch": 0.2223291047916792,
      "flos": 19099342103040.0,
      "grad_norm": 3.2493742166296724,
      "language_loss": 0.89306223,
      "learning_rate": 3.62451093520323e-06,
      "loss": 0.91240591,
      "num_input_tokens_seen": 39352180,
      "step": 1849,
      "time_per_iteration": 2.725503444671631
    },
    {
      "auxiliary_loss_clip": 0.01142869,
      "auxiliary_loss_mlp": 0.01043007,
      "balance_loss_clip": 1.05894065,
      "balance_loss_mlp": 1.03274286,
      "epoch": 0.22244934768231828,
      "flos": 20850418126080.0,
      "grad_norm": 2.7121293512784233,
      "language_loss": 0.90745497,
      "learning_rate": 3.6240564363884714e-06,
      "loss": 0.92931372,
      "num_input_tokens_seen": 39372125,
      "step": 1850,
      "time_per_iteration": 2.8883230686187744
    },
    {
      "auxiliary_loss_clip": 0.01210229,
      "auxiliary_loss_mlp": 0.01037128,
      "balance_loss_clip": 1.0666213,
      "balance_loss_mlp": 1.02629161,
      "epoch": 0.2225695905729574,
      "flos": 15632921111040.0,
      "grad_norm": 3.476407466782312,
      "language_loss": 0.70768887,
      "learning_rate": 3.623601691205643e-06,
      "loss": 0.73016238,
      "num_input_tokens_seen": 39391200,
      "step": 1851,
      "time_per_iteration": 2.6403589248657227
    },
    {
      "auxiliary_loss_clip": 0.0115423,
      "auxiliary_loss_mlp": 0.00764216,
      "balance_loss_clip": 1.0593605,
      "balance_loss_mlp": 1.00070882,
      "epoch": 0.22268983346359647,
      "flos": 25373582265600.0,
      "grad_norm": 2.6934924907685565,
      "language_loss": 0.81567043,
      "learning_rate": 3.623146699723729e-06,
      "loss": 0.8348549,
      "num_input_tokens_seen": 39410660,
      "step": 1852,
      "time_per_iteration": 2.8244597911834717
    },
    {
      "auxiliary_loss_clip": 0.0122922,
      "auxiliary_loss_mlp": 0.01041646,
      "balance_loss_clip": 1.0730015,
      "balance_loss_mlp": 1.03023815,
      "epoch": 0.22281007635423555,
      "flos": 13261452359040.0,
      "grad_norm": 2.4184984893816717,
      "language_loss": 0.77679765,
      "learning_rate": 3.6226914620117507e-06,
      "loss": 0.79950631,
      "num_input_tokens_seen": 39429280,
      "step": 1853,
      "time_per_iteration": 2.5766115188598633
    },
    {
      "auxiliary_loss_clip": 0.01207081,
      "auxiliary_loss_mlp": 0.01040145,
      "balance_loss_clip": 1.06584859,
      "balance_loss_mlp": 1.02995276,
      "epoch": 0.22293031924487464,
      "flos": 15340536403200.0,
      "grad_norm": 1.9911544065209505,
      "language_loss": 0.80753714,
      "learning_rate": 3.622235978138768e-06,
      "loss": 0.83000946,
      "num_input_tokens_seen": 39446905,
      "step": 1854,
      "time_per_iteration": 2.6638662815093994
    },
    {
      "auxiliary_loss_clip": 0.0121383,
      "auxiliary_loss_mlp": 0.01044863,
      "balance_loss_clip": 1.06755102,
      "balance_loss_mlp": 1.0338006,
      "epoch": 0.22305056213551375,
      "flos": 22564649773440.0,
      "grad_norm": 2.0715902329667815,
      "language_loss": 0.81563151,
      "learning_rate": 3.621780248173877e-06,
      "loss": 0.83821845,
      "num_input_tokens_seen": 39465105,
      "step": 1855,
      "time_per_iteration": 2.680614948272705
    },
    {
      "auxiliary_loss_clip": 0.01075746,
      "auxiliary_loss_mlp": 0.01001601,
      "balance_loss_clip": 1.02705193,
      "balance_loss_mlp": 0.99924034,
      "epoch": 0.22317080502615283,
      "flos": 64880419887360.0,
      "grad_norm": 0.8818831299539281,
      "language_loss": 0.60956442,
      "learning_rate": 3.6213242721862125e-06,
      "loss": 0.63033789,
      "num_input_tokens_seen": 39523560,
      "step": 1856,
      "time_per_iteration": 4.298959732055664
    },
    {
      "auxiliary_loss_clip": 0.0114165,
      "auxiliary_loss_mlp": 0.00763931,
      "balance_loss_clip": 1.05875969,
      "balance_loss_mlp": 1.00058532,
      "epoch": 0.2232910479167919,
      "flos": 25775997310080.0,
      "grad_norm": 1.9232022797446175,
      "language_loss": 0.75516951,
      "learning_rate": 3.620868050244945e-06,
      "loss": 0.77422529,
      "num_input_tokens_seen": 39544040,
      "step": 1857,
      "time_per_iteration": 2.8607232570648193
    },
    {
      "auxiliary_loss_clip": 0.0117932,
      "auxiliary_loss_mlp": 0.01044547,
      "balance_loss_clip": 1.05936921,
      "balance_loss_mlp": 1.03236973,
      "epoch": 0.22341129080743102,
      "flos": 23251799928960.0,
      "grad_norm": 2.1158108440288976,
      "language_loss": 0.77860653,
      "learning_rate": 3.6204115824192817e-06,
      "loss": 0.80084527,
      "num_input_tokens_seen": 39561515,
      "step": 1858,
      "time_per_iteration": 3.7162559032440186
    },
    {
      "auxiliary_loss_clip": 0.01135004,
      "auxiliary_loss_mlp": 0.01044451,
      "balance_loss_clip": 1.05139089,
      "balance_loss_mlp": 1.0334177,
      "epoch": 0.2235315336980701,
      "flos": 21214552250880.0,
      "grad_norm": 3.782162357806844,
      "language_loss": 0.7687273,
      "learning_rate": 3.619954868778471e-06,
      "loss": 0.79052186,
      "num_input_tokens_seen": 39578210,
      "step": 1859,
      "time_per_iteration": 3.7191758155822754
    },
    {
      "auxiliary_loss_clip": 0.01182184,
      "auxiliary_loss_mlp": 0.01048715,
      "balance_loss_clip": 1.06330466,
      "balance_loss_mlp": 1.03861165,
      "epoch": 0.2236517765887092,
      "flos": 19901945548800.0,
      "grad_norm": 2.9222368602523763,
      "language_loss": 0.83075798,
      "learning_rate": 3.6194979093917944e-06,
      "loss": 0.85306698,
      "num_input_tokens_seen": 39597625,
      "step": 1860,
      "time_per_iteration": 2.755401849746704
    },
    {
      "auxiliary_loss_clip": 0.01194365,
      "auxiliary_loss_mlp": 0.01045067,
      "balance_loss_clip": 1.0646708,
      "balance_loss_mlp": 1.03452277,
      "epoch": 0.22377201947934827,
      "flos": 23214847812480.0,
      "grad_norm": 2.0574282852957504,
      "language_loss": 0.8734647,
      "learning_rate": 3.6190407043285724e-06,
      "loss": 0.895859,
      "num_input_tokens_seen": 39615360,
      "step": 1861,
      "time_per_iteration": 2.6785225868225098
    },
    {
      "auxiliary_loss_clip": 0.01177819,
      "auxiliary_loss_mlp": 0.01039814,
      "balance_loss_clip": 1.06315565,
      "balance_loss_mlp": 1.02906764,
      "epoch": 0.22389226236998738,
      "flos": 26794244056320.0,
      "grad_norm": 2.216378751186828,
      "language_loss": 0.75781691,
      "learning_rate": 3.618583253658163e-06,
      "loss": 0.7799933,
      "num_input_tokens_seen": 39635460,
      "step": 1862,
      "time_per_iteration": 2.767181158065796
    },
    {
      "auxiliary_loss_clip": 0.01211967,
      "auxiliary_loss_mlp": 0.01046226,
      "balance_loss_clip": 1.06800866,
      "balance_loss_mlp": 1.03529453,
      "epoch": 0.22401250526062647,
      "flos": 24170359455360.0,
      "grad_norm": 2.968893394603283,
      "language_loss": 0.86641884,
      "learning_rate": 3.618125557449961e-06,
      "loss": 0.88900071,
      "num_input_tokens_seen": 39653515,
      "step": 1863,
      "time_per_iteration": 3.654590129852295
    },
    {
      "auxiliary_loss_clip": 0.01226128,
      "auxiliary_loss_mlp": 0.01043581,
      "balance_loss_clip": 1.06831694,
      "balance_loss_mlp": 1.03382921,
      "epoch": 0.22413274815126555,
      "flos": 16759761649920.0,
      "grad_norm": 2.0672346465379854,
      "language_loss": 0.8354404,
      "learning_rate": 3.6176676157733983e-06,
      "loss": 0.85813749,
      "num_input_tokens_seen": 39668525,
      "step": 1864,
      "time_per_iteration": 2.551332712173462
    },
    {
      "auxiliary_loss_clip": 0.01182589,
      "auxiliary_loss_mlp": 0.01037365,
      "balance_loss_clip": 1.06411755,
      "balance_loss_mlp": 1.02639818,
      "epoch": 0.22425299104190466,
      "flos": 21360205900800.0,
      "grad_norm": 2.6270490758912475,
      "language_loss": 0.75805408,
      "learning_rate": 3.6172094286979443e-06,
      "loss": 0.78025365,
      "num_input_tokens_seen": 39685895,
      "step": 1865,
      "time_per_iteration": 2.754756450653076
    },
    {
      "auxiliary_loss_clip": 0.01195421,
      "auxiliary_loss_mlp": 0.01039401,
      "balance_loss_clip": 1.06294107,
      "balance_loss_mlp": 1.02895248,
      "epoch": 0.22437323393254374,
      "flos": 32165547108480.0,
      "grad_norm": 2.512607449935796,
      "language_loss": 0.81510401,
      "learning_rate": 3.6167509962931064e-06,
      "loss": 0.83745217,
      "num_input_tokens_seen": 39711595,
      "step": 1866,
      "time_per_iteration": 2.830676555633545
    },
    {
      "auxiliary_loss_clip": 0.01177378,
      "auxiliary_loss_mlp": 0.01036194,
      "balance_loss_clip": 1.06276405,
      "balance_loss_mlp": 1.02669299,
      "epoch": 0.22449347682318282,
      "flos": 18002809664640.0,
      "grad_norm": 2.93237425854836,
      "language_loss": 0.77655327,
      "learning_rate": 3.6162923186284276e-06,
      "loss": 0.79868901,
      "num_input_tokens_seen": 39727555,
      "step": 1867,
      "time_per_iteration": 2.759535789489746
    },
    {
      "auxiliary_loss_clip": 0.01207091,
      "auxiliary_loss_mlp": 0.01035987,
      "balance_loss_clip": 1.06662667,
      "balance_loss_mlp": 1.02517533,
      "epoch": 0.2246137197138219,
      "flos": 18697286194560.0,
      "grad_norm": 2.9015948791610264,
      "language_loss": 0.86121631,
      "learning_rate": 3.6158333957734888e-06,
      "loss": 0.88364708,
      "num_input_tokens_seen": 39746145,
      "step": 1868,
      "time_per_iteration": 2.672886371612549
    },
    {
      "auxiliary_loss_clip": 0.0120604,
      "auxiliary_loss_mlp": 0.01045492,
      "balance_loss_clip": 1.06583321,
      "balance_loss_mlp": 1.03518665,
      "epoch": 0.22473396260446102,
      "flos": 15590653781760.0,
      "grad_norm": 2.264956178687926,
      "language_loss": 0.82982051,
      "learning_rate": 3.6153742277979088e-06,
      "loss": 0.85233581,
      "num_input_tokens_seen": 39763575,
      "step": 1869,
      "time_per_iteration": 2.6746015548706055
    },
    {
      "auxiliary_loss_clip": 0.01176476,
      "auxiliary_loss_mlp": 0.01039964,
      "balance_loss_clip": 1.05975795,
      "balance_loss_mlp": 1.02908611,
      "epoch": 0.2248542054951001,
      "flos": 14465501182080.0,
      "grad_norm": 2.8275835837663976,
      "language_loss": 0.78342432,
      "learning_rate": 3.6149148147713434e-06,
      "loss": 0.80558872,
      "num_input_tokens_seen": 39781810,
      "step": 1870,
      "time_per_iteration": 2.7632927894592285
    },
    {
      "auxiliary_loss_clip": 0.01169665,
      "auxiliary_loss_mlp": 0.01039386,
      "balance_loss_clip": 1.06622076,
      "balance_loss_mlp": 1.02850211,
      "epoch": 0.22497444838573918,
      "flos": 19243882431360.0,
      "grad_norm": 2.3008380028449835,
      "language_loss": 0.86510921,
      "learning_rate": 3.614455156763484e-06,
      "loss": 0.88719964,
      "num_input_tokens_seen": 39800115,
      "step": 1871,
      "time_per_iteration": 2.7802371978759766
    },
    {
      "auxiliary_loss_clip": 0.01184087,
      "auxiliary_loss_mlp": 0.01036984,
      "balance_loss_clip": 1.06010175,
      "balance_loss_mlp": 1.02648187,
      "epoch": 0.2250946912763783,
      "flos": 16910299549440.0,
      "grad_norm": 2.226595035913541,
      "language_loss": 0.71033412,
      "learning_rate": 3.613995253844061e-06,
      "loss": 0.73254478,
      "num_input_tokens_seen": 39817795,
      "step": 1872,
      "time_per_iteration": 2.698249340057373
    },
    {
      "auxiliary_loss_clip": 0.01179437,
      "auxiliary_loss_mlp": 0.01036243,
      "balance_loss_clip": 1.05852962,
      "balance_loss_mlp": 1.02617002,
      "epoch": 0.22521493416701738,
      "flos": 24681368292480.0,
      "grad_norm": 2.811259414252998,
      "language_loss": 0.80816436,
      "learning_rate": 3.6135351060828414e-06,
      "loss": 0.83032113,
      "num_input_tokens_seen": 39838270,
      "step": 1873,
      "time_per_iteration": 2.7647147178649902
    },
    {
      "auxiliary_loss_clip": 0.01200427,
      "auxiliary_loss_mlp": 0.01045685,
      "balance_loss_clip": 1.06478024,
      "balance_loss_mlp": 1.03450942,
      "epoch": 0.22533517705765646,
      "flos": 17821963664640.0,
      "grad_norm": 8.106866375006518,
      "language_loss": 0.69333071,
      "learning_rate": 3.6130747135496285e-06,
      "loss": 0.71579182,
      "num_input_tokens_seen": 39857270,
      "step": 1874,
      "time_per_iteration": 2.714459180831909
    },
    {
      "auxiliary_loss_clip": 0.0116062,
      "auxiliary_loss_mlp": 0.01042744,
      "balance_loss_clip": 1.05626702,
      "balance_loss_mlp": 1.03084719,
      "epoch": 0.22545541994829554,
      "flos": 33691390899840.0,
      "grad_norm": 2.1540791999084705,
      "language_loss": 0.66127616,
      "learning_rate": 3.6126140763142646e-06,
      "loss": 0.68330979,
      "num_input_tokens_seen": 39882300,
      "step": 1875,
      "time_per_iteration": 2.8493235111236572
    },
    {
      "auxiliary_loss_clip": 0.01175876,
      "auxiliary_loss_mlp": 0.01039351,
      "balance_loss_clip": 1.06111264,
      "balance_loss_mlp": 1.0284791,
      "epoch": 0.22557566283893465,
      "flos": 19171594310400.0,
      "grad_norm": 4.563607000804462,
      "language_loss": 0.86374927,
      "learning_rate": 3.6121531944466275e-06,
      "loss": 0.88590151,
      "num_input_tokens_seen": 39899625,
      "step": 1876,
      "time_per_iteration": 2.722808599472046
    },
    {
      "auxiliary_loss_clip": 0.01177516,
      "auxiliary_loss_mlp": 0.01037076,
      "balance_loss_clip": 1.06053436,
      "balance_loss_mlp": 1.02669907,
      "epoch": 0.22569590572957374,
      "flos": 20773281669120.0,
      "grad_norm": 2.2636670028578054,
      "language_loss": 0.78866768,
      "learning_rate": 3.611692068016633e-06,
      "loss": 0.81081367,
      "num_input_tokens_seen": 39915955,
      "step": 1877,
      "time_per_iteration": 2.645982265472412
    },
    {
      "auxiliary_loss_clip": 0.01179268,
      "auxiliary_loss_mlp": 0.01044099,
      "balance_loss_clip": 1.06359565,
      "balance_loss_mlp": 1.03226733,
      "epoch": 0.22581614862021282,
      "flos": 18442715529600.0,
      "grad_norm": 2.4436943187178235,
      "language_loss": 0.75215155,
      "learning_rate": 3.611230697094233e-06,
      "loss": 0.77438521,
      "num_input_tokens_seen": 39932655,
      "step": 1878,
      "time_per_iteration": 2.762866735458374
    },
    {
      "auxiliary_loss_clip": 0.01172633,
      "auxiliary_loss_mlp": 0.01041592,
      "balance_loss_clip": 1.06574619,
      "balance_loss_mlp": 1.03174496,
      "epoch": 0.22593639151085193,
      "flos": 20048389297920.0,
      "grad_norm": 1.9912463477553362,
      "language_loss": 0.87449592,
      "learning_rate": 3.6107690817494173e-06,
      "loss": 0.89663815,
      "num_input_tokens_seen": 39952875,
      "step": 1879,
      "time_per_iteration": 2.8029463291168213
    },
    {
      "auxiliary_loss_clip": 0.01188067,
      "auxiliary_loss_mlp": 0.01043511,
      "balance_loss_clip": 1.06144619,
      "balance_loss_mlp": 1.0333302,
      "epoch": 0.226056634401491,
      "flos": 13115116350720.0,
      "grad_norm": 3.8136486975945263,
      "language_loss": 0.71269119,
      "learning_rate": 3.6103072220522117e-06,
      "loss": 0.73500693,
      "num_input_tokens_seen": 39968405,
      "step": 1880,
      "time_per_iteration": 2.6741104125976562
    },
    {
      "auxiliary_loss_clip": 0.01154864,
      "auxiliary_loss_mlp": 0.01041445,
      "balance_loss_clip": 1.05616748,
      "balance_loss_mlp": 1.03106833,
      "epoch": 0.2261768772921301,
      "flos": 18988378012800.0,
      "grad_norm": 2.2335717846185976,
      "language_loss": 0.92178988,
      "learning_rate": 3.609845118072682e-06,
      "loss": 0.94375294,
      "num_input_tokens_seen": 39987075,
      "step": 1881,
      "time_per_iteration": 2.743398666381836
    },
    {
      "auxiliary_loss_clip": 0.01184767,
      "auxiliary_loss_mlp": 0.01037616,
      "balance_loss_clip": 1.06154501,
      "balance_loss_mlp": 1.02701199,
      "epoch": 0.2262971201827692,
      "flos": 19974054101760.0,
      "grad_norm": 2.5317599215872604,
      "language_loss": 0.80163515,
      "learning_rate": 3.6093827698809276e-06,
      "loss": 0.82385892,
      "num_input_tokens_seen": 40006175,
      "step": 1882,
      "time_per_iteration": 2.702986240386963
    },
    {
      "auxiliary_loss_clip": 0.0116923,
      "auxiliary_loss_mlp": 0.01035814,
      "balance_loss_clip": 1.06136715,
      "balance_loss_mlp": 1.02580082,
      "epoch": 0.2264173630734083,
      "flos": 16654543735680.0,
      "grad_norm": 2.4479648250372006,
      "language_loss": 0.84588742,
      "learning_rate": 3.6089201775470864e-06,
      "loss": 0.86793786,
      "num_input_tokens_seen": 40021630,
      "step": 1883,
      "time_per_iteration": 3.7651984691619873
    },
    {
      "auxiliary_loss_clip": 0.0114976,
      "auxiliary_loss_mlp": 0.01032902,
      "balance_loss_clip": 1.06169403,
      "balance_loss_mlp": 1.02261412,
      "epoch": 0.22653760596404737,
      "flos": 24389809597440.0,
      "grad_norm": 1.4692103936569307,
      "language_loss": 0.77470511,
      "learning_rate": 3.6084573411413334e-06,
      "loss": 0.79653174,
      "num_input_tokens_seen": 40041025,
      "step": 1884,
      "time_per_iteration": 3.854905366897583
    },
    {
      "auxiliary_loss_clip": 0.01171285,
      "auxiliary_loss_mlp": 0.01039588,
      "balance_loss_clip": 1.06040502,
      "balance_loss_mlp": 1.02891254,
      "epoch": 0.22665784885468646,
      "flos": 18332541538560.0,
      "grad_norm": 2.413663237256814,
      "language_loss": 0.80773175,
      "learning_rate": 3.607994260733881e-06,
      "loss": 0.82984048,
      "num_input_tokens_seen": 40060265,
      "step": 1885,
      "time_per_iteration": 3.6695711612701416
    },
    {
      "auxiliary_loss_clip": 0.01181278,
      "auxiliary_loss_mlp": 0.01036827,
      "balance_loss_clip": 1.06185746,
      "balance_loss_mlp": 1.02719498,
      "epoch": 0.22677809174532557,
      "flos": 24058102475520.0,
      "grad_norm": 2.3472929677207097,
      "language_loss": 0.74812293,
      "learning_rate": 3.6075309363949776e-06,
      "loss": 0.77030402,
      "num_input_tokens_seen": 40079435,
      "step": 1886,
      "time_per_iteration": 2.7536582946777344
    },
    {
      "auxiliary_loss_clip": 0.01211076,
      "auxiliary_loss_mlp": 0.01036176,
      "balance_loss_clip": 1.06501913,
      "balance_loss_mlp": 1.02557206,
      "epoch": 0.22689833463596465,
      "flos": 20374242503040.0,
      "grad_norm": 4.15391015525508,
      "language_loss": 0.8141098,
      "learning_rate": 3.6070673681949094e-06,
      "loss": 0.8365823,
      "num_input_tokens_seen": 40097800,
      "step": 1887,
      "time_per_iteration": 2.632739305496216
    },
    {
      "auxiliary_loss_clip": 0.01120803,
      "auxiliary_loss_mlp": 0.01045007,
      "balance_loss_clip": 1.05673456,
      "balance_loss_mlp": 1.03572702,
      "epoch": 0.22701857752660373,
      "flos": 30120398438400.0,
      "grad_norm": 2.1127054204085103,
      "language_loss": 0.81786275,
      "learning_rate": 3.606603556203999e-06,
      "loss": 0.83952087,
      "num_input_tokens_seen": 40122745,
      "step": 1888,
      "time_per_iteration": 3.9007813930511475
    },
    {
      "auxiliary_loss_clip": 0.01138331,
      "auxiliary_loss_mlp": 0.0076419,
      "balance_loss_clip": 1.05153489,
      "balance_loss_mlp": 1.0007689,
      "epoch": 0.22713882041724284,
      "flos": 22492182084480.0,
      "grad_norm": 3.7955510361149,
      "language_loss": 0.83598435,
      "learning_rate": 3.6061395004926066e-06,
      "loss": 0.85500956,
      "num_input_tokens_seen": 40141680,
      "step": 1889,
      "time_per_iteration": 2.90921688079834
    },
    {
      "auxiliary_loss_clip": 0.0121329,
      "auxiliary_loss_mlp": 0.0103356,
      "balance_loss_clip": 1.06298399,
      "balance_loss_mlp": 1.02261662,
      "epoch": 0.22725906330788193,
      "flos": 20521548178560.0,
      "grad_norm": 3.40919182238203,
      "language_loss": 0.85218132,
      "learning_rate": 3.605675201131129e-06,
      "loss": 0.87464976,
      "num_input_tokens_seen": 40160140,
      "step": 1890,
      "time_per_iteration": 2.6491358280181885
    },
    {
      "auxiliary_loss_clip": 0.01214111,
      "auxiliary_loss_mlp": 0.01039655,
      "balance_loss_clip": 1.07017016,
      "balance_loss_mlp": 1.02967119,
      "epoch": 0.227379306198521,
      "flos": 18989922297600.0,
      "grad_norm": 4.454269233354582,
      "language_loss": 0.79809487,
      "learning_rate": 3.60521065819e-06,
      "loss": 0.82063246,
      "num_input_tokens_seen": 40177450,
      "step": 1891,
      "time_per_iteration": 2.7146003246307373
    },
    {
      "auxiliary_loss_clip": 0.01207019,
      "auxiliary_loss_mlp": 0.01046508,
      "balance_loss_clip": 1.06596994,
      "balance_loss_mlp": 1.03599358,
      "epoch": 0.2274995490891601,
      "flos": 21798351999360.0,
      "grad_norm": 2.9363364191028696,
      "language_loss": 0.88391399,
      "learning_rate": 3.60474587173969e-06,
      "loss": 0.90644926,
      "num_input_tokens_seen": 40195935,
      "step": 1892,
      "time_per_iteration": 2.637737512588501
    },
    {
      "auxiliary_loss_clip": 0.01108253,
      "auxiliary_loss_mlp": 0.01040079,
      "balance_loss_clip": 1.04969907,
      "balance_loss_mlp": 1.03017831,
      "epoch": 0.2276197919797992,
      "flos": 19058654972160.0,
      "grad_norm": 2.1161954636973315,
      "language_loss": 0.84374839,
      "learning_rate": 3.6042808418507084e-06,
      "loss": 0.86523169,
      "num_input_tokens_seen": 40213620,
      "step": 1893,
      "time_per_iteration": 3.1102468967437744
    },
    {
      "auxiliary_loss_clip": 0.01209181,
      "auxiliary_loss_mlp": 0.01050428,
      "balance_loss_clip": 1.06663179,
      "balance_loss_mlp": 1.03972864,
      "epoch": 0.22774003487043828,
      "flos": 18806777827200.0,
      "grad_norm": 3.1602781982565116,
      "language_loss": 0.77272069,
      "learning_rate": 3.6038155685935976e-06,
      "loss": 0.79531682,
      "num_input_tokens_seen": 40230190,
      "step": 1894,
      "time_per_iteration": 3.069122552871704
    },
    {
      "auxiliary_loss_clip": 0.01210226,
      "auxiliary_loss_mlp": 0.01035691,
      "balance_loss_clip": 1.06580651,
      "balance_loss_mlp": 1.02483153,
      "epoch": 0.22786027776107737,
      "flos": 23002544476800.0,
      "grad_norm": 4.475424925523545,
      "language_loss": 0.70707297,
      "learning_rate": 3.6033500520389404e-06,
      "loss": 0.72953212,
      "num_input_tokens_seen": 40246860,
      "step": 1895,
      "time_per_iteration": 2.7134294509887695
    },
    {
      "auxiliary_loss_clip": 0.01088549,
      "auxiliary_loss_mlp": 0.01005551,
      "balance_loss_clip": 1.031528,
      "balance_loss_mlp": 1.00338125,
      "epoch": 0.22798052065171648,
      "flos": 66706872600960.0,
      "grad_norm": 0.7905203077941925,
      "language_loss": 0.64766264,
      "learning_rate": 3.6028842922573553e-06,
      "loss": 0.66860354,
      "num_input_tokens_seen": 40311005,
      "step": 1896,
      "time_per_iteration": 3.4483425617218018
    },
    {
      "auxiliary_loss_clip": 0.01056812,
      "auxiliary_loss_mlp": 0.01011439,
      "balance_loss_clip": 1.02600861,
      "balance_loss_mlp": 1.00917399,
      "epoch": 0.22810076354235556,
      "flos": 62080896758400.0,
      "grad_norm": 0.8614496890015833,
      "language_loss": 0.62801754,
      "learning_rate": 3.602418289319497e-06,
      "loss": 0.64870006,
      "num_input_tokens_seen": 40369560,
      "step": 1897,
      "time_per_iteration": 3.340824842453003
    },
    {
      "auxiliary_loss_clip": 0.01211666,
      "auxiliary_loss_mlp": 0.01047715,
      "balance_loss_clip": 1.06961775,
      "balance_loss_mlp": 1.03728426,
      "epoch": 0.22822100643299464,
      "flos": 23876358635520.0,
      "grad_norm": 2.5672747499798274,
      "language_loss": 0.73257709,
      "learning_rate": 3.601952043296059e-06,
      "loss": 0.75517094,
      "num_input_tokens_seen": 40389555,
      "step": 1898,
      "time_per_iteration": 2.6938133239746094
    },
    {
      "auxiliary_loss_clip": 0.01214572,
      "auxiliary_loss_mlp": 0.01036781,
      "balance_loss_clip": 1.06911564,
      "balance_loss_mlp": 1.02703547,
      "epoch": 0.22834124932363373,
      "flos": 20991331180800.0,
      "grad_norm": 3.0761408544417974,
      "language_loss": 0.80735207,
      "learning_rate": 3.6014855542577696e-06,
      "loss": 0.82986557,
      "num_input_tokens_seen": 40406765,
      "step": 1899,
      "time_per_iteration": 2.7505810260772705
    },
    {
      "auxiliary_loss_clip": 0.01200614,
      "auxiliary_loss_mlp": 0.00764425,
      "balance_loss_clip": 1.06762075,
      "balance_loss_mlp": 1.00063705,
      "epoch": 0.22846149221427284,
      "flos": 24901572620160.0,
      "grad_norm": 3.4682677747696653,
      "language_loss": 0.84267402,
      "learning_rate": 3.6010188222753943e-06,
      "loss": 0.86232436,
      "num_input_tokens_seen": 40427535,
      "step": 1900,
      "time_per_iteration": 2.7239532470703125
    },
    {
      "auxiliary_loss_clip": 0.01078906,
      "auxiliary_loss_mlp": 0.01000219,
      "balance_loss_clip": 1.02953744,
      "balance_loss_mlp": 0.99800187,
      "epoch": 0.22858173510491192,
      "flos": 56132294319360.0,
      "grad_norm": 0.9081286720376891,
      "language_loss": 0.64110374,
      "learning_rate": 3.6005518474197372e-06,
      "loss": 0.66189498,
      "num_input_tokens_seen": 40479580,
      "step": 1901,
      "time_per_iteration": 3.281759023666382
    },
    {
      "auxiliary_loss_clip": 0.01202356,
      "auxiliary_loss_mlp": 0.01044556,
      "balance_loss_clip": 1.06628811,
      "balance_loss_mlp": 1.03349376,
      "epoch": 0.228701977995551,
      "flos": 24170826332160.0,
      "grad_norm": 4.098561866619031,
      "language_loss": 0.78652841,
      "learning_rate": 3.6000846297616373e-06,
      "loss": 0.80899751,
      "num_input_tokens_seen": 40497880,
      "step": 1902,
      "time_per_iteration": 2.7036592960357666
    },
    {
      "auxiliary_loss_clip": 0.01171298,
      "auxiliary_loss_mlp": 0.01050668,
      "balance_loss_clip": 1.06593621,
      "balance_loss_mlp": 1.04029083,
      "epoch": 0.22882222088619011,
      "flos": 21387892308480.0,
      "grad_norm": 2.329659074005726,
      "language_loss": 0.72879994,
      "learning_rate": 3.5996171693719717e-06,
      "loss": 0.7510196,
      "num_input_tokens_seen": 40513975,
      "step": 1903,
      "time_per_iteration": 2.7419052124023438
    },
    {
      "auxiliary_loss_clip": 0.01098345,
      "auxiliary_loss_mlp": 0.01008129,
      "balance_loss_clip": 1.03450871,
      "balance_loss_mlp": 1.00574493,
      "epoch": 0.2289424637768292,
      "flos": 64589615377920.0,
      "grad_norm": 0.8319921786053258,
      "language_loss": 0.64714777,
      "learning_rate": 3.5991494663216528e-06,
      "loss": 0.66821253,
      "num_input_tokens_seen": 40576960,
      "step": 1904,
      "time_per_iteration": 3.324284553527832
    },
    {
      "auxiliary_loss_clip": 0.01184591,
      "auxiliary_loss_mlp": 0.01036235,
      "balance_loss_clip": 1.06160486,
      "balance_loss_mlp": 1.02656686,
      "epoch": 0.22906270666746828,
      "flos": 22163419877760.0,
      "grad_norm": 2.39794287214683,
      "language_loss": 0.87880313,
      "learning_rate": 3.5986815206816314e-06,
      "loss": 0.90101147,
      "num_input_tokens_seen": 40595780,
      "step": 1905,
      "time_per_iteration": 2.7380950450897217
    },
    {
      "auxiliary_loss_clip": 0.01189554,
      "auxiliary_loss_mlp": 0.01041286,
      "balance_loss_clip": 1.05947828,
      "balance_loss_mlp": 1.03132057,
      "epoch": 0.2291829495581074,
      "flos": 25772334122880.0,
      "grad_norm": 1.9187668468813273,
      "language_loss": 0.74688119,
      "learning_rate": 3.598213332522895e-06,
      "loss": 0.7691896,
      "num_input_tokens_seen": 40615810,
      "step": 1906,
      "time_per_iteration": 2.7859270572662354
    },
    {
      "auxiliary_loss_clip": 0.01120779,
      "auxiliary_loss_mlp": 0.01035169,
      "balance_loss_clip": 1.0546186,
      "balance_loss_mlp": 1.02595448,
      "epoch": 0.22930319244874647,
      "flos": 31172760126720.0,
      "grad_norm": 2.104923437163601,
      "language_loss": 0.77487898,
      "learning_rate": 3.597744901916466e-06,
      "loss": 0.79643846,
      "num_input_tokens_seen": 40637095,
      "step": 1907,
      "time_per_iteration": 2.90014386177063
    },
    {
      "auxiliary_loss_clip": 0.01212412,
      "auxiliary_loss_mlp": 0.01038586,
      "balance_loss_clip": 1.06562769,
      "balance_loss_mlp": 1.02783322,
      "epoch": 0.22942343533938556,
      "flos": 23254098399360.0,
      "grad_norm": 2.436077735904654,
      "language_loss": 0.76511139,
      "learning_rate": 3.5972762289334058e-06,
      "loss": 0.78762138,
      "num_input_tokens_seen": 40656725,
      "step": 1908,
      "time_per_iteration": 4.087263584136963
    },
    {
      "auxiliary_loss_clip": 0.01208978,
      "auxiliary_loss_mlp": 0.0104534,
      "balance_loss_clip": 1.0688951,
      "balance_loss_mlp": 1.03567803,
      "epoch": 0.22954367823002464,
      "flos": 14610903436800.0,
      "grad_norm": 1.9641462450397467,
      "language_loss": 0.85343289,
      "learning_rate": 3.5968073136448116e-06,
      "loss": 0.87597597,
      "num_input_tokens_seen": 40674745,
      "step": 1909,
      "time_per_iteration": 2.722790241241455
    },
    {
      "auxiliary_loss_clip": 0.0121136,
      "auxiliary_loss_mlp": 0.01045194,
      "balance_loss_clip": 1.0679841,
      "balance_loss_mlp": 1.03471589,
      "epoch": 0.22966392112066375,
      "flos": 16763604405120.0,
      "grad_norm": 2.3640725134965144,
      "language_loss": 0.9110502,
      "learning_rate": 3.596338156121818e-06,
      "loss": 0.93361568,
      "num_input_tokens_seen": 40693630,
      "step": 1910,
      "time_per_iteration": 3.6797027587890625
    },
    {
      "auxiliary_loss_clip": 0.01075699,
      "auxiliary_loss_mlp": 0.01002789,
      "balance_loss_clip": 1.02792263,
      "balance_loss_mlp": 1.00061929,
      "epoch": 0.22978416401130283,
      "flos": 67474247783040.0,
      "grad_norm": 0.7801453835999994,
      "language_loss": 0.59290987,
      "learning_rate": 3.595868756435595e-06,
      "loss": 0.61369473,
      "num_input_tokens_seen": 40761310,
      "step": 1911,
      "time_per_iteration": 4.394990921020508
    },
    {
      "auxiliary_loss_clip": 0.01214087,
      "auxiliary_loss_mlp": 0.0103969,
      "balance_loss_clip": 1.0680443,
      "balance_loss_mlp": 1.0286932,
      "epoch": 0.22990440690194192,
      "flos": 19865137086720.0,
      "grad_norm": 2.5344093311409406,
      "language_loss": 0.80305642,
      "learning_rate": 3.5953991146573504e-06,
      "loss": 0.82559425,
      "num_input_tokens_seen": 40779955,
      "step": 1912,
      "time_per_iteration": 2.7264554500579834
    },
    {
      "auxiliary_loss_clip": 0.01211937,
      "auxiliary_loss_mlp": 0.01032258,
      "balance_loss_clip": 1.06980705,
      "balance_loss_mlp": 1.02215457,
      "epoch": 0.23002464979258103,
      "flos": 13289246507520.0,
      "grad_norm": 4.485703895186528,
      "language_loss": 0.83546627,
      "learning_rate": 3.5949292308583294e-06,
      "loss": 0.85790825,
      "num_input_tokens_seen": 40793200,
      "step": 1913,
      "time_per_iteration": 2.608097553253174
    },
    {
      "auxiliary_loss_clip": 0.0122517,
      "auxiliary_loss_mlp": 0.01036396,
      "balance_loss_clip": 1.06929231,
      "balance_loss_mlp": 1.02618623,
      "epoch": 0.2301448926832201,
      "flos": 22163779013760.0,
      "grad_norm": 2.2159450472841757,
      "language_loss": 0.81167448,
      "learning_rate": 3.594459105109811e-06,
      "loss": 0.83429015,
      "num_input_tokens_seen": 40812380,
      "step": 1914,
      "time_per_iteration": 3.6441361904144287
    },
    {
      "auxiliary_loss_clip": 0.01214772,
      "auxiliary_loss_mlp": 0.01035397,
      "balance_loss_clip": 1.06959546,
      "balance_loss_mlp": 1.02562809,
      "epoch": 0.2302651355738592,
      "flos": 20704477167360.0,
      "grad_norm": 2.3125551772451605,
      "language_loss": 0.8173638,
      "learning_rate": 3.593988737483115e-06,
      "loss": 0.83986551,
      "num_input_tokens_seen": 40832320,
      "step": 1915,
      "time_per_iteration": 2.7635271549224854
    },
    {
      "auxiliary_loss_clip": 0.01179428,
      "auxiliary_loss_mlp": 0.01038947,
      "balance_loss_clip": 1.063802,
      "balance_loss_mlp": 1.02884388,
      "epoch": 0.23038537846449827,
      "flos": 18588943797120.0,
      "grad_norm": 2.693844248108563,
      "language_loss": 0.78440762,
      "learning_rate": 3.5935181280495947e-06,
      "loss": 0.80659133,
      "num_input_tokens_seen": 40850900,
      "step": 1916,
      "time_per_iteration": 2.7956910133361816
    },
    {
      "auxiliary_loss_clip": 0.01071889,
      "auxiliary_loss_mlp": 0.01006701,
      "balance_loss_clip": 1.02588797,
      "balance_loss_mlp": 1.00478196,
      "epoch": 0.23050562135513739,
      "flos": 64224260190720.0,
      "grad_norm": 0.8035596804998746,
      "language_loss": 0.54244745,
      "learning_rate": 3.5930472768806412e-06,
      "loss": 0.56323326,
      "num_input_tokens_seen": 40909570,
      "step": 1917,
      "time_per_iteration": 3.2905426025390625
    },
    {
      "auxiliary_loss_clip": 0.01194485,
      "auxiliary_loss_mlp": 0.01039023,
      "balance_loss_clip": 1.06884503,
      "balance_loss_mlp": 1.02883077,
      "epoch": 0.23062586424577647,
      "flos": 17313396952320.0,
      "grad_norm": 5.58954404906789,
      "language_loss": 0.77632499,
      "learning_rate": 3.5925761840476826e-06,
      "loss": 0.79866004,
      "num_input_tokens_seen": 40928180,
      "step": 1918,
      "time_per_iteration": 2.721961498260498
    },
    {
      "auxiliary_loss_clip": 0.01181638,
      "auxiliary_loss_mlp": 0.01045727,
      "balance_loss_clip": 1.06492448,
      "balance_loss_mlp": 1.03581464,
      "epoch": 0.23074610713641555,
      "flos": 27855979194240.0,
      "grad_norm": 2.503962373530742,
      "language_loss": 0.81647325,
      "learning_rate": 3.592104849622183e-06,
      "loss": 0.83874691,
      "num_input_tokens_seen": 40950435,
      "step": 1919,
      "time_per_iteration": 2.8163869380950928
    },
    {
      "auxiliary_loss_clip": 0.01222109,
      "auxiliary_loss_mlp": 0.01037412,
      "balance_loss_clip": 1.0666945,
      "balance_loss_mlp": 1.0268563,
      "epoch": 0.23086635002705466,
      "flos": 28841798937600.0,
      "grad_norm": 2.23387665472359,
      "language_loss": 0.73151219,
      "learning_rate": 3.591633273675644e-06,
      "loss": 0.75410736,
      "num_input_tokens_seen": 40972670,
      "step": 1920,
      "time_per_iteration": 2.7857770919799805
    },
    {
      "auxiliary_loss_clip": 0.01098391,
      "auxiliary_loss_mlp": 0.01000707,
      "balance_loss_clip": 1.021819,
      "balance_loss_mlp": 0.99863291,
      "epoch": 0.23098659291769374,
      "flos": 62923681566720.0,
      "grad_norm": 0.9021554814658758,
      "language_loss": 0.58138251,
      "learning_rate": 3.591161456279602e-06,
      "loss": 0.60237348,
      "num_input_tokens_seen": 41018215,
      "step": 1921,
      "time_per_iteration": 3.066370725631714
    },
    {
      "auxiliary_loss_clip": 0.01177637,
      "auxiliary_loss_mlp": 0.01043233,
      "balance_loss_clip": 1.06119394,
      "balance_loss_mlp": 1.03358293,
      "epoch": 0.23110683580833283,
      "flos": 23476816679040.0,
      "grad_norm": 10.063577297091177,
      "language_loss": 0.80318201,
      "learning_rate": 3.590689397505633e-06,
      "loss": 0.8253907,
      "num_input_tokens_seen": 41039125,
      "step": 1922,
      "time_per_iteration": 2.770897388458252
    },
    {
      "auxiliary_loss_clip": 0.01161075,
      "auxiliary_loss_mlp": 0.00764143,
      "balance_loss_clip": 1.05763173,
      "balance_loss_mlp": 1.00074518,
      "epoch": 0.2312270786989719,
      "flos": 27271066124160.0,
      "grad_norm": 4.616879621784079,
      "language_loss": 0.86933506,
      "learning_rate": 3.590217097425347e-06,
      "loss": 0.88858724,
      "num_input_tokens_seen": 41059025,
      "step": 1923,
      "time_per_iteration": 2.8505256175994873
    },
    {
      "auxiliary_loss_clip": 0.01208186,
      "auxiliary_loss_mlp": 0.01042598,
      "balance_loss_clip": 1.06886721,
      "balance_loss_mlp": 1.03154123,
      "epoch": 0.23134732158961102,
      "flos": 13261344618240.0,
      "grad_norm": 2.315071212024194,
      "language_loss": 0.7105366,
      "learning_rate": 3.589744556110391e-06,
      "loss": 0.73304439,
      "num_input_tokens_seen": 41077015,
      "step": 1924,
      "time_per_iteration": 2.701972246170044
    },
    {
      "auxiliary_loss_clip": 0.01158988,
      "auxiliary_loss_mlp": 0.01034827,
      "balance_loss_clip": 1.06031752,
      "balance_loss_mlp": 1.02516508,
      "epoch": 0.2314675644802501,
      "flos": 36977648250240.0,
      "grad_norm": 1.9804893702126478,
      "language_loss": 0.84283507,
      "learning_rate": 3.58927177363245e-06,
      "loss": 0.86477321,
      "num_input_tokens_seen": 41099840,
      "step": 1925,
      "time_per_iteration": 2.916490077972412
    },
    {
      "auxiliary_loss_clip": 0.01181713,
      "auxiliary_loss_mlp": 0.01037284,
      "balance_loss_clip": 1.06217289,
      "balance_loss_mlp": 1.02554762,
      "epoch": 0.2315878073708892,
      "flos": 23842207779840.0,
      "grad_norm": 4.501342505894998,
      "language_loss": 0.72278762,
      "learning_rate": 3.5887987500632447e-06,
      "loss": 0.74497759,
      "num_input_tokens_seen": 41117845,
      "step": 1926,
      "time_per_iteration": 2.812741279602051
    },
    {
      "auxiliary_loss_clip": 0.010951,
      "auxiliary_loss_mlp": 0.01046363,
      "balance_loss_clip": 1.04914665,
      "balance_loss_mlp": 1.03691602,
      "epoch": 0.2317080502615283,
      "flos": 23039424766080.0,
      "grad_norm": 1.7495399194777077,
      "language_loss": 0.84326404,
      "learning_rate": 3.5883254854745325e-06,
      "loss": 0.86467862,
      "num_input_tokens_seen": 41136235,
      "step": 1927,
      "time_per_iteration": 3.159984588623047
    },
    {
      "auxiliary_loss_clip": 0.01189989,
      "auxiliary_loss_mlp": 0.01035427,
      "balance_loss_clip": 1.064991,
      "balance_loss_mlp": 1.0252943,
      "epoch": 0.23182829315216738,
      "flos": 11254656435840.0,
      "grad_norm": 2.345966032593877,
      "language_loss": 0.75254744,
      "learning_rate": 3.587851979938107e-06,
      "loss": 0.77480161,
      "num_input_tokens_seen": 41153125,
      "step": 1928,
      "time_per_iteration": 2.9706568717956543
    },
    {
      "auxiliary_loss_clip": 0.01108325,
      "auxiliary_loss_mlp": 0.01041737,
      "balance_loss_clip": 1.05293298,
      "balance_loss_mlp": 1.03117537,
      "epoch": 0.23194853604280646,
      "flos": 19828939155840.0,
      "grad_norm": 3.347578310652382,
      "language_loss": 0.77896208,
      "learning_rate": 3.5873782335257985e-06,
      "loss": 0.80046272,
      "num_input_tokens_seen": 41171290,
      "step": 1929,
      "time_per_iteration": 3.0004847049713135
    },
    {
      "auxiliary_loss_clip": 0.01187479,
      "auxiliary_loss_mlp": 0.01036796,
      "balance_loss_clip": 1.06256557,
      "balance_loss_mlp": 1.02526259,
      "epoch": 0.23206877893344555,
      "flos": 15305020830720.0,
      "grad_norm": 3.192546474403519,
      "language_loss": 0.78556573,
      "learning_rate": 3.5869042463094744e-06,
      "loss": 0.80780852,
      "num_input_tokens_seen": 41189005,
      "step": 1930,
      "time_per_iteration": 3.1214277744293213
    },
    {
      "auxiliary_loss_clip": 0.01194876,
      "auxiliary_loss_mlp": 0.01037621,
      "balance_loss_clip": 1.06355047,
      "balance_loss_mlp": 1.02777982,
      "epoch": 0.23218902182408466,
      "flos": 22711488572160.0,
      "grad_norm": 2.531319494945736,
      "language_loss": 0.76652884,
      "learning_rate": 3.586430018361038e-06,
      "loss": 0.78885376,
      "num_input_tokens_seen": 41208775,
      "step": 1931,
      "time_per_iteration": 2.8532049655914307
    },
    {
      "auxiliary_loss_clip": 0.01226901,
      "auxiliary_loss_mlp": 0.01038822,
      "balance_loss_clip": 1.0715456,
      "balance_loss_mlp": 1.02919626,
      "epoch": 0.23230926471472374,
      "flos": 22710734386560.0,
      "grad_norm": 2.7771098542617163,
      "language_loss": 0.76738167,
      "learning_rate": 3.5859555497524283e-06,
      "loss": 0.79003882,
      "num_input_tokens_seen": 41226010,
      "step": 1932,
      "time_per_iteration": 2.6792385578155518
    },
    {
      "auxiliary_loss_clip": 0.011964,
      "auxiliary_loss_mlp": 0.00764157,
      "balance_loss_clip": 1.06644082,
      "balance_loss_mlp": 1.00084543,
      "epoch": 0.23242950760536282,
      "flos": 20375499479040.0,
      "grad_norm": 2.2728132712563003,
      "language_loss": 0.92463279,
      "learning_rate": 3.5854808405556237e-06,
      "loss": 0.94423842,
      "num_input_tokens_seen": 41245245,
      "step": 1933,
      "time_per_iteration": 2.7467575073242188
    },
    {
      "auxiliary_loss_clip": 0.01160694,
      "auxiliary_loss_mlp": 0.01039411,
      "balance_loss_clip": 1.05404472,
      "balance_loss_mlp": 1.02852166,
      "epoch": 0.23254975049600193,
      "flos": 16908324301440.0,
      "grad_norm": 12.614228662616297,
      "language_loss": 0.75235218,
      "learning_rate": 3.5850058908426355e-06,
      "loss": 0.77435321,
      "num_input_tokens_seen": 41263795,
      "step": 1934,
      "time_per_iteration": 3.8829431533813477
    },
    {
      "auxiliary_loss_clip": 0.01209724,
      "auxiliary_loss_mlp": 0.01040136,
      "balance_loss_clip": 1.06456757,
      "balance_loss_mlp": 1.03023601,
      "epoch": 0.23266999338664102,
      "flos": 23294821443840.0,
      "grad_norm": 4.580268405094075,
      "language_loss": 0.85603368,
      "learning_rate": 3.584530700685514e-06,
      "loss": 0.87853229,
      "num_input_tokens_seen": 41284055,
      "step": 1935,
      "time_per_iteration": 2.684300422668457
    },
    {
      "auxiliary_loss_clip": 0.01186531,
      "auxiliary_loss_mlp": 0.01039133,
      "balance_loss_clip": 1.06067276,
      "balance_loss_mlp": 1.02946496,
      "epoch": 0.2327902362772801,
      "flos": 19569987031680.0,
      "grad_norm": 2.036935062347896,
      "language_loss": 0.88838947,
      "learning_rate": 3.5840552701563448e-06,
      "loss": 0.91064608,
      "num_input_tokens_seen": 41300255,
      "step": 1936,
      "time_per_iteration": 3.657386302947998
    },
    {
      "auxiliary_loss_clip": 0.01199562,
      "auxiliary_loss_mlp": 0.01042924,
      "balance_loss_clip": 1.06392777,
      "balance_loss_mlp": 1.03251696,
      "epoch": 0.2329104791679192,
      "flos": 16727514215040.0,
      "grad_norm": 2.1874737313587533,
      "language_loss": 0.81435812,
      "learning_rate": 3.5835795993272513e-06,
      "loss": 0.83678293,
      "num_input_tokens_seen": 41318540,
      "step": 1937,
      "time_per_iteration": 3.548078775405884
    },
    {
      "auxiliary_loss_clip": 0.0115857,
      "auxiliary_loss_mlp": 0.00763603,
      "balance_loss_clip": 1.05751801,
      "balance_loss_mlp": 1.00089335,
      "epoch": 0.2330307220585583,
      "flos": 22163743100160.0,
      "grad_norm": 2.613491198636571,
      "language_loss": 0.70842743,
      "learning_rate": 3.583103688270391e-06,
      "loss": 0.72764915,
      "num_input_tokens_seen": 41338320,
      "step": 1938,
      "time_per_iteration": 2.784445285797119
    },
    {
      "auxiliary_loss_clip": 0.01129115,
      "auxiliary_loss_mlp": 0.01045098,
      "balance_loss_clip": 1.05250132,
      "balance_loss_mlp": 1.03447652,
      "epoch": 0.23315096494919738,
      "flos": 19317319787520.0,
      "grad_norm": 2.394783379355195,
      "language_loss": 0.89771545,
      "learning_rate": 3.58262753705796e-06,
      "loss": 0.91945755,
      "num_input_tokens_seen": 41353210,
      "step": 1939,
      "time_per_iteration": 2.8175852298736572
    },
    {
      "auxiliary_loss_clip": 0.01097435,
      "auxiliary_loss_mlp": 0.00998917,
      "balance_loss_clip": 1.02474952,
      "balance_loss_mlp": 0.99685496,
      "epoch": 0.23327120783983646,
      "flos": 53031048946560.0,
      "grad_norm": 0.7610458402149918,
      "language_loss": 0.55436468,
      "learning_rate": 3.5821511457621902e-06,
      "loss": 0.57532817,
      "num_input_tokens_seen": 41410510,
      "step": 1940,
      "time_per_iteration": 4.142487049102783
    },
    {
      "auxiliary_loss_clip": 0.01167996,
      "auxiliary_loss_mlp": 0.01045669,
      "balance_loss_clip": 1.05997479,
      "balance_loss_mlp": 1.03450489,
      "epoch": 0.23339145073047557,
      "flos": 17126984344320.0,
      "grad_norm": 5.883686482030216,
      "language_loss": 0.80657572,
      "learning_rate": 3.5816745144553497e-06,
      "loss": 0.82871234,
      "num_input_tokens_seen": 41425830,
      "step": 1941,
      "time_per_iteration": 2.697645902633667
    },
    {
      "auxiliary_loss_clip": 0.01210036,
      "auxiliary_loss_mlp": 0.01039263,
      "balance_loss_clip": 1.06526792,
      "balance_loss_mlp": 1.02907634,
      "epoch": 0.23351169362111465,
      "flos": 13078918419840.0,
      "grad_norm": 2.6285635095910713,
      "language_loss": 0.757411,
      "learning_rate": 3.5811976432097424e-06,
      "loss": 0.77990401,
      "num_input_tokens_seen": 41443500,
      "step": 1942,
      "time_per_iteration": 2.633315086364746
    },
    {
      "auxiliary_loss_clip": 0.0122342,
      "auxiliary_loss_mlp": 0.00763904,
      "balance_loss_clip": 1.06949484,
      "balance_loss_mlp": 1.00068688,
      "epoch": 0.23363193651175373,
      "flos": 15851257931520.0,
      "grad_norm": 2.323853129846723,
      "language_loss": 0.84529334,
      "learning_rate": 3.58072053209771e-06,
      "loss": 0.86516654,
      "num_input_tokens_seen": 41460055,
      "step": 1943,
      "time_per_iteration": 2.6071417331695557
    },
    {
      "auxiliary_loss_clip": 0.01138398,
      "auxiliary_loss_mlp": 0.0104184,
      "balance_loss_clip": 1.05454683,
      "balance_loss_mlp": 1.03078914,
      "epoch": 0.23375217940239285,
      "flos": 21025769345280.0,
      "grad_norm": 7.186882990353247,
      "language_loss": 0.79155934,
      "learning_rate": 3.5802431811916296e-06,
      "loss": 0.81336176,
      "num_input_tokens_seen": 41476665,
      "step": 1944,
      "time_per_iteration": 2.846282958984375
    },
    {
      "auxiliary_loss_clip": 0.01179657,
      "auxiliary_loss_mlp": 0.01032379,
      "balance_loss_clip": 1.05697632,
      "balance_loss_mlp": 1.0214237,
      "epoch": 0.23387242229303193,
      "flos": 20594698225920.0,
      "grad_norm": 2.7854637731216774,
      "language_loss": 0.80606961,
      "learning_rate": 3.579765590563916e-06,
      "loss": 0.82818997,
      "num_input_tokens_seen": 41496065,
      "step": 1945,
      "time_per_iteration": 2.655947685241699
    },
    {
      "auxiliary_loss_clip": 0.0119762,
      "auxiliary_loss_mlp": 0.01040988,
      "balance_loss_clip": 1.06611037,
      "balance_loss_mlp": 1.03109944,
      "epoch": 0.233992665183671,
      "flos": 24279491952000.0,
      "grad_norm": 2.344160334563518,
      "language_loss": 0.81882191,
      "learning_rate": 3.579287760287017e-06,
      "loss": 0.84120798,
      "num_input_tokens_seen": 41516815,
      "step": 1946,
      "time_per_iteration": 2.7798478603363037
    },
    {
      "auxiliary_loss_clip": 0.01185616,
      "auxiliary_loss_mlp": 0.01042896,
      "balance_loss_clip": 1.06085563,
      "balance_loss_mlp": 1.03272724,
      "epoch": 0.2341129080743101,
      "flos": 30154621121280.0,
      "grad_norm": 1.9065008231963787,
      "language_loss": 0.72771949,
      "learning_rate": 3.578809690433421e-06,
      "loss": 0.75000459,
      "num_input_tokens_seen": 41538525,
      "step": 1947,
      "time_per_iteration": 2.752173662185669
    },
    {
      "auxiliary_loss_clip": 0.01205618,
      "auxiliary_loss_mlp": 0.01039598,
      "balance_loss_clip": 1.06671262,
      "balance_loss_mlp": 1.02878571,
      "epoch": 0.2342331509649492,
      "flos": 22784135829120.0,
      "grad_norm": 2.668099860057924,
      "language_loss": 0.8154003,
      "learning_rate": 3.578331381075651e-06,
      "loss": 0.83785248,
      "num_input_tokens_seen": 41559025,
      "step": 1948,
      "time_per_iteration": 2.685189723968506
    },
    {
      "auxiliary_loss_clip": 0.01180196,
      "auxiliary_loss_mlp": 0.01038134,
      "balance_loss_clip": 1.06151092,
      "balance_loss_mlp": 1.02815628,
      "epoch": 0.2343533938555883,
      "flos": 23623152687360.0,
      "grad_norm": 2.4438571343924043,
      "language_loss": 0.69989628,
      "learning_rate": 3.5778528322862646e-06,
      "loss": 0.72207958,
      "num_input_tokens_seen": 41577845,
      "step": 1949,
      "time_per_iteration": 2.7390127182006836
    },
    {
      "auxiliary_loss_clip": 0.01222609,
      "auxiliary_loss_mlp": 0.01035967,
      "balance_loss_clip": 1.06825089,
      "balance_loss_mlp": 1.02591228,
      "epoch": 0.23447363674622737,
      "flos": 24570332375040.0,
      "grad_norm": 3.321924632891526,
      "language_loss": 0.86811423,
      "learning_rate": 3.5773740441378585e-06,
      "loss": 0.89069998,
      "num_input_tokens_seen": 41598600,
      "step": 1950,
      "time_per_iteration": 2.6635658740997314
    },
    {
      "auxiliary_loss_clip": 0.01208139,
      "auxiliary_loss_mlp": 0.01039993,
      "balance_loss_clip": 1.06701708,
      "balance_loss_mlp": 1.03026509,
      "epoch": 0.23459387963686648,
      "flos": 53140322119680.0,
      "grad_norm": 2.5149603206842652,
      "language_loss": 0.74104863,
      "learning_rate": 3.5768950167030633e-06,
      "loss": 0.7635299,
      "num_input_tokens_seen": 41623300,
      "step": 1951,
      "time_per_iteration": 2.996309995651245
    },
    {
      "auxiliary_loss_clip": 0.01172141,
      "auxiliary_loss_mlp": 0.01044902,
      "balance_loss_clip": 1.06190634,
      "balance_loss_mlp": 1.03452528,
      "epoch": 0.23471412252750556,
      "flos": 23951412103680.0,
      "grad_norm": 3.276650003177166,
      "language_loss": 0.78607094,
      "learning_rate": 3.576415750054548e-06,
      "loss": 0.80824137,
      "num_input_tokens_seen": 41643420,
      "step": 1952,
      "time_per_iteration": 2.810891628265381
    },
    {
      "auxiliary_loss_clip": 0.01189612,
      "auxiliary_loss_mlp": 0.01046817,
      "balance_loss_clip": 1.06512773,
      "balance_loss_mlp": 1.036762,
      "epoch": 0.23483436541814465,
      "flos": 15706573948800.0,
      "grad_norm": 1.9910639460947108,
      "language_loss": 0.85809821,
      "learning_rate": 3.5759362442650172e-06,
      "loss": 0.88046253,
      "num_input_tokens_seen": 41660170,
      "step": 1953,
      "time_per_iteration": 2.707736015319824
    },
    {
      "auxiliary_loss_clip": 0.01192147,
      "auxiliary_loss_mlp": 0.01043735,
      "balance_loss_clip": 1.0629828,
      "balance_loss_mlp": 1.03379846,
      "epoch": 0.23495460830878373,
      "flos": 24936262179840.0,
      "grad_norm": 2.800207518147886,
      "language_loss": 0.85720527,
      "learning_rate": 3.5754564994072113e-06,
      "loss": 0.87956405,
      "num_input_tokens_seen": 41679010,
      "step": 1954,
      "time_per_iteration": 2.8078439235687256
    },
    {
      "auxiliary_loss_clip": 0.011952,
      "auxiliary_loss_mlp": 0.0103443,
      "balance_loss_clip": 1.06404674,
      "balance_loss_mlp": 1.02392745,
      "epoch": 0.23507485119942284,
      "flos": 30482665056000.0,
      "grad_norm": 6.334615232185839,
      "language_loss": 0.59713042,
      "learning_rate": 3.5749765155539067e-06,
      "loss": 0.61942673,
      "num_input_tokens_seen": 41699495,
      "step": 1955,
      "time_per_iteration": 2.8305108547210693
    },
    {
      "auxiliary_loss_clip": 0.01172091,
      "auxiliary_loss_mlp": 0.01043162,
      "balance_loss_clip": 1.05967546,
      "balance_loss_mlp": 1.03243923,
      "epoch": 0.23519509409006192,
      "flos": 18329129746560.0,
      "grad_norm": 4.1065523656348635,
      "language_loss": 0.92361164,
      "learning_rate": 3.574496292777917e-06,
      "loss": 0.94576412,
      "num_input_tokens_seen": 41717705,
      "step": 1956,
      "time_per_iteration": 2.7089173793792725
    },
    {
      "auxiliary_loss_clip": 0.01214464,
      "auxiliary_loss_mlp": 0.01039695,
      "balance_loss_clip": 1.06555748,
      "balance_loss_mlp": 1.02793527,
      "epoch": 0.235315336980701,
      "flos": 29643217234560.0,
      "grad_norm": 2.776570205235359,
      "language_loss": 0.7203213,
      "learning_rate": 3.574015831152092e-06,
      "loss": 0.74286282,
      "num_input_tokens_seen": 41738120,
      "step": 1957,
      "time_per_iteration": 2.729628801345825
    },
    {
      "auxiliary_loss_clip": 0.0115742,
      "auxiliary_loss_mlp": 0.01040778,
      "balance_loss_clip": 1.06027186,
      "balance_loss_mlp": 1.03023434,
      "epoch": 0.23543557987134012,
      "flos": 18551704371840.0,
      "grad_norm": 4.508482993000119,
      "language_loss": 0.83268118,
      "learning_rate": 3.573535130749316e-06,
      "loss": 0.85466313,
      "num_input_tokens_seen": 41756070,
      "step": 1958,
      "time_per_iteration": 2.7749762535095215
    },
    {
      "auxiliary_loss_clip": 0.01164022,
      "auxiliary_loss_mlp": 0.0104078,
      "balance_loss_clip": 1.06219077,
      "balance_loss_mlp": 1.03079581,
      "epoch": 0.2355558227619792,
      "flos": 24679033908480.0,
      "grad_norm": 2.677610310641086,
      "language_loss": 0.73979211,
      "learning_rate": 3.5730541916425127e-06,
      "loss": 0.76184011,
      "num_input_tokens_seen": 41777550,
      "step": 1959,
      "time_per_iteration": 2.8087193965911865
    },
    {
      "auxiliary_loss_clip": 0.01210057,
      "auxiliary_loss_mlp": 0.01041491,
      "balance_loss_clip": 1.06857669,
      "balance_loss_mlp": 1.03086376,
      "epoch": 0.23567606565261828,
      "flos": 21944795748480.0,
      "grad_norm": 2.074008389798478,
      "language_loss": 0.85678011,
      "learning_rate": 3.572573013904639e-06,
      "loss": 0.87929553,
      "num_input_tokens_seen": 41797460,
      "step": 1960,
      "time_per_iteration": 3.7266809940338135
    },
    {
      "auxiliary_loss_clip": 0.01183999,
      "auxiliary_loss_mlp": 0.01040056,
      "balance_loss_clip": 1.06160808,
      "balance_loss_mlp": 1.03012002,
      "epoch": 0.2357963085432574,
      "flos": 13589352639360.0,
      "grad_norm": 2.557466023817922,
      "language_loss": 0.9247781,
      "learning_rate": 3.572091597608689e-06,
      "loss": 0.94701862,
      "num_input_tokens_seen": 41815585,
      "step": 1961,
      "time_per_iteration": 3.640892744064331
    },
    {
      "auxiliary_loss_clip": 0.01180132,
      "auxiliary_loss_mlp": 0.01032949,
      "balance_loss_clip": 1.0660851,
      "balance_loss_mlp": 1.02223253,
      "epoch": 0.23591655143389648,
      "flos": 22088689632000.0,
      "grad_norm": 2.3713395097698666,
      "language_loss": 0.73680532,
      "learning_rate": 3.571609942827694e-06,
      "loss": 0.75893617,
      "num_input_tokens_seen": 41834700,
      "step": 1962,
      "time_per_iteration": 2.7272140979766846
    },
    {
      "auxiliary_loss_clip": 0.01177024,
      "auxiliary_loss_mlp": 0.01041484,
      "balance_loss_clip": 1.06344509,
      "balance_loss_mlp": 1.03126144,
      "epoch": 0.23603679432453556,
      "flos": 17017349057280.0,
      "grad_norm": 2.1881507580973993,
      "language_loss": 0.88502878,
      "learning_rate": 3.57112804963472e-06,
      "loss": 0.90721381,
      "num_input_tokens_seen": 41852915,
      "step": 1963,
      "time_per_iteration": 3.6697723865509033
    },
    {
      "auxiliary_loss_clip": 0.01143734,
      "auxiliary_loss_mlp": 0.01048587,
      "balance_loss_clip": 1.05917168,
      "balance_loss_mlp": 1.03952134,
      "epoch": 0.23615703721517464,
      "flos": 19171307001600.0,
      "grad_norm": 2.1273826058312006,
      "language_loss": 0.76485276,
      "learning_rate": 3.57064591810287e-06,
      "loss": 0.78677595,
      "num_input_tokens_seen": 41870415,
      "step": 1964,
      "time_per_iteration": 2.7882959842681885
    },
    {
      "auxiliary_loss_clip": 0.01224381,
      "auxiliary_loss_mlp": 0.0104243,
      "balance_loss_clip": 1.07109022,
      "balance_loss_mlp": 1.0328697,
      "epoch": 0.23627728010581375,
      "flos": 19098803399040.0,
      "grad_norm": 2.190026436079446,
      "language_loss": 0.80244982,
      "learning_rate": 3.570163548305284e-06,
      "loss": 0.82511795,
      "num_input_tokens_seen": 41889345,
      "step": 1965,
      "time_per_iteration": 2.6466872692108154
    },
    {
      "auxiliary_loss_clip": 0.01202578,
      "auxiliary_loss_mlp": 0.01039369,
      "balance_loss_clip": 1.06626081,
      "balance_loss_mlp": 1.02871215,
      "epoch": 0.23639752299645284,
      "flos": 14282213057280.0,
      "grad_norm": 3.200879617958273,
      "language_loss": 0.69919384,
      "learning_rate": 3.569680940315135e-06,
      "loss": 0.72161329,
      "num_input_tokens_seen": 41905745,
      "step": 1966,
      "time_per_iteration": 3.6314609050750732
    },
    {
      "auxiliary_loss_clip": 0.01192549,
      "auxiliary_loss_mlp": 0.01039505,
      "balance_loss_clip": 1.06576753,
      "balance_loss_mlp": 1.02865696,
      "epoch": 0.23651776588709192,
      "flos": 22893411980160.0,
      "grad_norm": 2.7208665646780577,
      "language_loss": 0.82011068,
      "learning_rate": 3.5691980942056356e-06,
      "loss": 0.84243119,
      "num_input_tokens_seen": 41925115,
      "step": 1967,
      "time_per_iteration": 2.7284152507781982
    },
    {
      "auxiliary_loss_clip": 0.01168762,
      "auxiliary_loss_mlp": 0.01042683,
      "balance_loss_clip": 1.06586432,
      "balance_loss_mlp": 1.03299713,
      "epoch": 0.23663800877773103,
      "flos": 18624531196800.0,
      "grad_norm": 1.775707220061921,
      "language_loss": 0.80018842,
      "learning_rate": 3.5687150100500332e-06,
      "loss": 0.82230294,
      "num_input_tokens_seen": 41944815,
      "step": 1968,
      "time_per_iteration": 2.810779333114624
    },
    {
      "auxiliary_loss_clip": 0.01154984,
      "auxiliary_loss_mlp": 0.01039955,
      "balance_loss_clip": 1.06163931,
      "balance_loss_mlp": 1.03017998,
      "epoch": 0.2367582516683701,
      "flos": 25555828896000.0,
      "grad_norm": 2.425901286865552,
      "language_loss": 0.74563265,
      "learning_rate": 3.568231687921611e-06,
      "loss": 0.76758206,
      "num_input_tokens_seen": 41964990,
      "step": 1969,
      "time_per_iteration": 2.771275520324707
    },
    {
      "auxiliary_loss_clip": 0.01177337,
      "auxiliary_loss_mlp": 0.01037156,
      "balance_loss_clip": 1.05857301,
      "balance_loss_mlp": 1.02771509,
      "epoch": 0.2368784945590092,
      "flos": 23295072839040.0,
      "grad_norm": 2.300767901083434,
      "language_loss": 0.80385792,
      "learning_rate": 3.5677481278936883e-06,
      "loss": 0.82600284,
      "num_input_tokens_seen": 41984570,
      "step": 1970,
      "time_per_iteration": 2.7332284450531006
    },
    {
      "auxiliary_loss_clip": 0.01110534,
      "auxiliary_loss_mlp": 0.01007804,
      "balance_loss_clip": 1.03180826,
      "balance_loss_mlp": 1.00598037,
      "epoch": 0.23699873744964828,
      "flos": 69859291875840.0,
      "grad_norm": 0.828766435635211,
      "language_loss": 0.57790124,
      "learning_rate": 3.5672643300396214e-06,
      "loss": 0.59908462,
      "num_input_tokens_seen": 42053715,
      "step": 1971,
      "time_per_iteration": 3.2640087604522705
    },
    {
      "auxiliary_loss_clip": 0.01165672,
      "auxiliary_loss_mlp": 0.01032837,
      "balance_loss_clip": 1.06133544,
      "balance_loss_mlp": 1.02341938,
      "epoch": 0.2371189803402874,
      "flos": 21835052720640.0,
      "grad_norm": 3.1246014968721947,
      "language_loss": 0.67196381,
      "learning_rate": 3.566780294432802e-06,
      "loss": 0.69394886,
      "num_input_tokens_seen": 42070890,
      "step": 1972,
      "time_per_iteration": 2.6010258197784424
    },
    {
      "auxiliary_loss_clip": 0.01192514,
      "auxiliary_loss_mlp": 0.01041413,
      "balance_loss_clip": 1.06179261,
      "balance_loss_mlp": 1.03129244,
      "epoch": 0.23723922323092647,
      "flos": 21908490076800.0,
      "grad_norm": 2.5917777858760007,
      "language_loss": 0.74481559,
      "learning_rate": 3.566296021146657e-06,
      "loss": 0.76715487,
      "num_input_tokens_seen": 42090270,
      "step": 1973,
      "time_per_iteration": 2.600001573562622
    },
    {
      "auxiliary_loss_clip": 0.01162216,
      "auxiliary_loss_mlp": 0.01035129,
      "balance_loss_clip": 1.0591917,
      "balance_loss_mlp": 1.024508,
      "epoch": 0.23735946612156555,
      "flos": 32708803380480.0,
      "grad_norm": 2.282191820079571,
      "language_loss": 0.73262131,
      "learning_rate": 3.565811510254652e-06,
      "loss": 0.75459468,
      "num_input_tokens_seen": 42111150,
      "step": 1974,
      "time_per_iteration": 2.7568933963775635
    },
    {
      "auxiliary_loss_clip": 0.01043679,
      "auxiliary_loss_mlp": 0.01016476,
      "balance_loss_clip": 1.02848887,
      "balance_loss_mlp": 1.01461637,
      "epoch": 0.23747970901220466,
      "flos": 70546944821760.0,
      "grad_norm": 0.8301749341895739,
      "language_loss": 0.58169228,
      "learning_rate": 3.5653267618302845e-06,
      "loss": 0.60229385,
      "num_input_tokens_seen": 42178730,
      "step": 1975,
      "time_per_iteration": 3.6155643463134766
    },
    {
      "auxiliary_loss_clip": 0.01207206,
      "auxiliary_loss_mlp": 0.00763923,
      "balance_loss_clip": 1.06615329,
      "balance_loss_mlp": 1.00103641,
      "epoch": 0.23759995190284375,
      "flos": 20849807594880.0,
      "grad_norm": 5.806820371936132,
      "language_loss": 0.8584761,
      "learning_rate": 3.564841775947093e-06,
      "loss": 0.87818742,
      "num_input_tokens_seen": 42199620,
      "step": 1976,
      "time_per_iteration": 2.9170174598693848
    },
    {
      "auxiliary_loss_clip": 0.01205817,
      "auxiliary_loss_mlp": 0.01039365,
      "balance_loss_clip": 1.0623914,
      "balance_loss_mlp": 1.02965498,
      "epoch": 0.23772019479348283,
      "flos": 32921645420160.0,
      "grad_norm": 2.578309377416829,
      "language_loss": 0.76251268,
      "learning_rate": 3.5643565526786475e-06,
      "loss": 0.78496444,
      "num_input_tokens_seen": 42219560,
      "step": 1977,
      "time_per_iteration": 2.7427802085876465
    },
    {
      "auxiliary_loss_clip": 0.01168523,
      "auxiliary_loss_mlp": 0.01035953,
      "balance_loss_clip": 1.06256282,
      "balance_loss_mlp": 1.02599883,
      "epoch": 0.2378404376841219,
      "flos": 32342765834880.0,
      "grad_norm": 1.772348393944755,
      "language_loss": 0.77172959,
      "learning_rate": 3.5638710920985574e-06,
      "loss": 0.79377437,
      "num_input_tokens_seen": 42241020,
      "step": 1978,
      "time_per_iteration": 2.797088146209717
    },
    {
      "auxiliary_loss_clip": 0.01177087,
      "auxiliary_loss_mlp": 0.01043263,
      "balance_loss_clip": 1.06228149,
      "balance_loss_mlp": 1.03343976,
      "epoch": 0.23796068057476102,
      "flos": 22997624313600.0,
      "grad_norm": 2.005702256037837,
      "language_loss": 0.81742358,
      "learning_rate": 3.5633853942804655e-06,
      "loss": 0.83962715,
      "num_input_tokens_seen": 42259345,
      "step": 1979,
      "time_per_iteration": 2.6921768188476562
    },
    {
      "auxiliary_loss_clip": 0.01186416,
      "auxiliary_loss_mlp": 0.01035285,
      "balance_loss_clip": 1.06392193,
      "balance_loss_mlp": 1.02514029,
      "epoch": 0.2380809234654001,
      "flos": 13480938414720.0,
      "grad_norm": 2.5852908051739045,
      "language_loss": 0.7624774,
      "learning_rate": 3.5628994592980527e-06,
      "loss": 0.78469443,
      "num_input_tokens_seen": 42277250,
      "step": 1980,
      "time_per_iteration": 2.7019383907318115
    },
    {
      "auxiliary_loss_clip": 0.01207546,
      "auxiliary_loss_mlp": 0.00762876,
      "balance_loss_clip": 1.06550956,
      "balance_loss_mlp": 1.00062811,
      "epoch": 0.2382011663560392,
      "flos": 16871803148160.0,
      "grad_norm": 2.302305064361389,
      "language_loss": 0.70305711,
      "learning_rate": 3.562413287225034e-06,
      "loss": 0.72276133,
      "num_input_tokens_seen": 42295360,
      "step": 1981,
      "time_per_iteration": 2.561224937438965
    },
    {
      "auxiliary_loss_clip": 0.01157898,
      "auxiliary_loss_mlp": 0.01045602,
      "balance_loss_clip": 1.06223476,
      "balance_loss_mlp": 1.03557658,
      "epoch": 0.2383214092466783,
      "flos": 18441135331200.0,
      "grad_norm": 3.485637607498906,
      "language_loss": 0.89218915,
      "learning_rate": 3.5619268781351623e-06,
      "loss": 0.91422409,
      "num_input_tokens_seen": 42313430,
      "step": 1982,
      "time_per_iteration": 2.778637647628784
    },
    {
      "auxiliary_loss_clip": 0.01154934,
      "auxiliary_loss_mlp": 0.01033003,
      "balance_loss_clip": 1.05913389,
      "balance_loss_mlp": 1.02344835,
      "epoch": 0.23844165213731738,
      "flos": 19755717281280.0,
      "grad_norm": 2.1342200791851416,
      "language_loss": 0.76792777,
      "learning_rate": 3.5614402321022256e-06,
      "loss": 0.78980714,
      "num_input_tokens_seen": 42331260,
      "step": 1983,
      "time_per_iteration": 2.7474586963653564
    },
    {
      "auxiliary_loss_clip": 0.01152401,
      "auxiliary_loss_mlp": 0.00762874,
      "balance_loss_clip": 1.06077671,
      "balance_loss_mlp": 1.00090623,
      "epoch": 0.23856189502795647,
      "flos": 23367360960000.0,
      "grad_norm": 1.9460813391953002,
      "language_loss": 0.87266695,
      "learning_rate": 3.5609533492000463e-06,
      "loss": 0.89181972,
      "num_input_tokens_seen": 42350150,
      "step": 1984,
      "time_per_iteration": 2.82659649848938
    },
    {
      "auxiliary_loss_clip": 0.01189726,
      "auxiliary_loss_mlp": 0.01044368,
      "balance_loss_clip": 1.06596279,
      "balance_loss_mlp": 1.03483713,
      "epoch": 0.23868213791859555,
      "flos": 23475056912640.0,
      "grad_norm": 2.3340290326332327,
      "language_loss": 0.79057086,
      "learning_rate": 3.560466229502485e-06,
      "loss": 0.81291175,
      "num_input_tokens_seen": 42369495,
      "step": 1985,
      "time_per_iteration": 3.6070148944854736
    },
    {
      "auxiliary_loss_clip": 0.01204846,
      "auxiliary_loss_mlp": 0.01035251,
      "balance_loss_clip": 1.06834674,
      "balance_loss_mlp": 1.02589357,
      "epoch": 0.23880238080923466,
      "flos": 16617340224000.0,
      "grad_norm": 2.1562404209730763,
      "language_loss": 0.89769942,
      "learning_rate": 3.5599788730834384e-06,
      "loss": 0.92010045,
      "num_input_tokens_seen": 42387455,
      "step": 1986,
      "time_per_iteration": 3.6078147888183594
    },
    {
      "auxiliary_loss_clip": 0.011711,
      "auxiliary_loss_mlp": 0.00763192,
      "balance_loss_clip": 1.05881155,
      "balance_loss_mlp": 1.00080442,
      "epoch": 0.23892262369987374,
      "flos": 17348409734400.0,
      "grad_norm": 3.1082977931973024,
      "language_loss": 0.79160476,
      "learning_rate": 3.559491280016836e-06,
      "loss": 0.81094772,
      "num_input_tokens_seen": 42405400,
      "step": 1987,
      "time_per_iteration": 3.896418571472168
    },
    {
      "auxiliary_loss_clip": 0.01181393,
      "auxiliary_loss_mlp": 0.01036151,
      "balance_loss_clip": 1.06407928,
      "balance_loss_mlp": 1.02674484,
      "epoch": 0.23904286659051283,
      "flos": 22309899540480.0,
      "grad_norm": 2.1218882138229023,
      "language_loss": 0.70924997,
      "learning_rate": 3.5590034503766465e-06,
      "loss": 0.7314254,
      "num_input_tokens_seen": 42425065,
      "step": 1988,
      "time_per_iteration": 2.7704193592071533
    },
    {
      "auxiliary_loss_clip": 0.01217928,
      "auxiliary_loss_mlp": 0.01034406,
      "balance_loss_clip": 1.06747556,
      "balance_loss_mlp": 1.02525699,
      "epoch": 0.23916310948115194,
      "flos": 21178246579200.0,
      "grad_norm": 2.354892465182421,
      "language_loss": 0.81202912,
      "learning_rate": 3.558515384236874e-06,
      "loss": 0.83455241,
      "num_input_tokens_seen": 42442495,
      "step": 1989,
      "time_per_iteration": 3.7843313217163086
    },
    {
      "auxiliary_loss_clip": 0.01206755,
      "auxiliary_loss_mlp": 0.01042004,
      "balance_loss_clip": 1.06768656,
      "balance_loss_mlp": 1.0321573,
      "epoch": 0.23928335237179102,
      "flos": 14137349506560.0,
      "grad_norm": 12.97069533525323,
      "language_loss": 0.83431393,
      "learning_rate": 3.558027081671556e-06,
      "loss": 0.85680151,
      "num_input_tokens_seen": 42459480,
      "step": 1990,
      "time_per_iteration": 2.7007980346679688
    },
    {
      "auxiliary_loss_clip": 0.01203783,
      "auxiliary_loss_mlp": 0.01036628,
      "balance_loss_clip": 1.06428337,
      "balance_loss_mlp": 1.02700806,
      "epoch": 0.2394035952624301,
      "flos": 23769596436480.0,
      "grad_norm": 2.1877430109130303,
      "language_loss": 0.68679655,
      "learning_rate": 3.557538542754769e-06,
      "loss": 0.70920062,
      "num_input_tokens_seen": 42479175,
      "step": 1991,
      "time_per_iteration": 2.683703660964966
    },
    {
      "auxiliary_loss_clip": 0.01144726,
      "auxiliary_loss_mlp": 0.01036818,
      "balance_loss_clip": 1.05794597,
      "balance_loss_mlp": 1.0274837,
      "epoch": 0.2395238381530692,
      "flos": 24206198250240.0,
      "grad_norm": 3.1676998083402546,
      "language_loss": 0.67321688,
      "learning_rate": 3.557049767560623e-06,
      "loss": 0.69503236,
      "num_input_tokens_seen": 42498090,
      "step": 1992,
      "time_per_iteration": 3.745292901992798
    },
    {
      "auxiliary_loss_clip": 0.01168122,
      "auxiliary_loss_mlp": 0.01046986,
      "balance_loss_clip": 1.05732846,
      "balance_loss_mlp": 1.03724611,
      "epoch": 0.2396440810437083,
      "flos": 25295763450240.0,
      "grad_norm": 2.1769166731315788,
      "language_loss": 0.86039317,
      "learning_rate": 3.5565607561632655e-06,
      "loss": 0.88254428,
      "num_input_tokens_seen": 42516930,
      "step": 1993,
      "time_per_iteration": 2.8038854598999023
    },
    {
      "auxiliary_loss_clip": 0.01164561,
      "auxiliary_loss_mlp": 0.01032144,
      "balance_loss_clip": 1.05639601,
      "balance_loss_mlp": 1.02253556,
      "epoch": 0.23976432393434738,
      "flos": 28543093436160.0,
      "grad_norm": 2.661251433086316,
      "language_loss": 0.79978085,
      "learning_rate": 3.5560715086368787e-06,
      "loss": 0.82174802,
      "num_input_tokens_seen": 42534800,
      "step": 1994,
      "time_per_iteration": 2.8349316120147705
    },
    {
      "auxiliary_loss_clip": 0.0121739,
      "auxiliary_loss_mlp": 0.01033469,
      "balance_loss_clip": 1.06645489,
      "balance_loss_mlp": 1.02470708,
      "epoch": 0.23988456682498646,
      "flos": 19494358945920.0,
      "grad_norm": 3.7217515295339023,
      "language_loss": 0.82195318,
      "learning_rate": 3.5555820250556816e-06,
      "loss": 0.8444618,
      "num_input_tokens_seen": 42552000,
      "step": 1995,
      "time_per_iteration": 2.6963722705841064
    },
    {
      "auxiliary_loss_clip": 0.01143671,
      "auxiliary_loss_mlp": 0.01041984,
      "balance_loss_clip": 1.05697107,
      "balance_loss_mlp": 1.03199458,
      "epoch": 0.24000480971562557,
      "flos": 20266331068800.0,
      "grad_norm": 2.418810178091214,
      "language_loss": 0.69883239,
      "learning_rate": 3.5550923054939278e-06,
      "loss": 0.72068894,
      "num_input_tokens_seen": 42571455,
      "step": 1996,
      "time_per_iteration": 2.7795519828796387
    },
    {
      "auxiliary_loss_clip": 0.01140793,
      "auxiliary_loss_mlp": 0.00762895,
      "balance_loss_clip": 1.05715942,
      "balance_loss_mlp": 1.00089157,
      "epoch": 0.24012505260626466,
      "flos": 25443176866560.0,
      "grad_norm": 2.091969754325559,
      "language_loss": 0.74697739,
      "learning_rate": 3.5546023500259083e-06,
      "loss": 0.76601428,
      "num_input_tokens_seen": 42592550,
      "step": 1997,
      "time_per_iteration": 2.9992623329162598
    },
    {
      "auxiliary_loss_clip": 0.01190125,
      "auxiliary_loss_mlp": 0.01040183,
      "balance_loss_clip": 1.06294179,
      "balance_loss_mlp": 1.03073609,
      "epoch": 0.24024529549690374,
      "flos": 15553342529280.0,
      "grad_norm": 6.835990116019922,
      "language_loss": 0.81044173,
      "learning_rate": 3.5541121587259477e-06,
      "loss": 0.83274478,
      "num_input_tokens_seen": 42610385,
      "step": 1998,
      "time_per_iteration": 3.0458481311798096
    },
    {
      "auxiliary_loss_clip": 0.0108048,
      "auxiliary_loss_mlp": 0.00755607,
      "balance_loss_clip": 1.02581406,
      "balance_loss_mlp": 1.0016259,
      "epoch": 0.24036553838754285,
      "flos": 57122351867520.0,
      "grad_norm": 0.8361305585365975,
      "language_loss": 0.5784142,
      "learning_rate": 3.553621731668408e-06,
      "loss": 0.59677505,
      "num_input_tokens_seen": 42673595,
      "step": 1999,
      "time_per_iteration": 3.3511264324188232
    },
    {
      "auxiliary_loss_clip": 0.0119023,
      "auxiliary_loss_mlp": 0.0103443,
      "balance_loss_clip": 1.0615226,
      "balance_loss_mlp": 1.02548337,
      "epoch": 0.24048578127818193,
      "flos": 24969946158720.0,
      "grad_norm": 2.0419571405329835,
      "language_loss": 0.83458984,
      "learning_rate": 3.553131068927688e-06,
      "loss": 0.85683638,
      "num_input_tokens_seen": 42692000,
      "step": 2000,
      "time_per_iteration": 2.758882761001587
    },
    {
      "auxiliary_loss_clip": 0.01139049,
      "auxiliary_loss_mlp": 0.01034773,
      "balance_loss_clip": 1.05876088,
      "balance_loss_mlp": 1.02585614,
      "epoch": 0.24060602416882101,
      "flos": 23330947547520.0,
      "grad_norm": 1.801699059350038,
      "language_loss": 0.80669284,
      "learning_rate": 3.552640170578219e-06,
      "loss": 0.82843101,
      "num_input_tokens_seen": 42712250,
      "step": 2001,
      "time_per_iteration": 2.7826437950134277
    },
    {
      "auxiliary_loss_clip": 0.01154009,
      "auxiliary_loss_mlp": 0.01032038,
      "balance_loss_clip": 1.05392838,
      "balance_loss_mlp": 1.02356875,
      "epoch": 0.2407262670594601,
      "flos": 14173260128640.0,
      "grad_norm": 1.970600139327174,
      "language_loss": 0.78448331,
      "learning_rate": 3.5521490366944703e-06,
      "loss": 0.80634379,
      "num_input_tokens_seen": 42729900,
      "step": 2002,
      "time_per_iteration": 2.7307915687561035
    },
    {
      "auxiliary_loss_clip": 0.01161251,
      "auxiliary_loss_mlp": 0.01048645,
      "balance_loss_clip": 1.05990517,
      "balance_loss_mlp": 1.03938198,
      "epoch": 0.2408465099500992,
      "flos": 13663113217920.0,
      "grad_norm": 2.3114454328350047,
      "language_loss": 0.80321264,
      "learning_rate": 3.5516576673509474e-06,
      "loss": 0.82531166,
      "num_input_tokens_seen": 42747900,
      "step": 2003,
      "time_per_iteration": 2.7387075424194336
    },
    {
      "auxiliary_loss_clip": 0.01195083,
      "auxiliary_loss_mlp": 0.0103678,
      "balance_loss_clip": 1.06158412,
      "balance_loss_mlp": 1.02628934,
      "epoch": 0.2409667528407383,
      "flos": 31248029076480.0,
      "grad_norm": 1.8597014179455729,
      "language_loss": 0.86354852,
      "learning_rate": 3.5511660626221896e-06,
      "loss": 0.88586718,
      "num_input_tokens_seen": 42768540,
      "step": 2004,
      "time_per_iteration": 2.710313081741333
    },
    {
      "auxiliary_loss_clip": 0.0118823,
      "auxiliary_loss_mlp": 0.01033995,
      "balance_loss_clip": 1.06206775,
      "balance_loss_mlp": 1.02433324,
      "epoch": 0.24108699573137737,
      "flos": 22199941031040.0,
      "grad_norm": 3.769473671407566,
      "language_loss": 0.89594299,
      "learning_rate": 3.5506742225827744e-06,
      "loss": 0.91816521,
      "num_input_tokens_seen": 42785395,
      "step": 2005,
      "time_per_iteration": 2.79610538482666
    },
    {
      "auxiliary_loss_clip": 0.01166106,
      "auxiliary_loss_mlp": 0.01034449,
      "balance_loss_clip": 1.06055951,
      "balance_loss_mlp": 1.02498949,
      "epoch": 0.24120723862201648,
      "flos": 26103035664000.0,
      "grad_norm": 2.323488481191171,
      "language_loss": 0.90518618,
      "learning_rate": 3.5501821473073116e-06,
      "loss": 0.92719173,
      "num_input_tokens_seen": 42801980,
      "step": 2006,
      "time_per_iteration": 2.8170604705810547
    },
    {
      "auxiliary_loss_clip": 0.01201073,
      "auxiliary_loss_mlp": 0.01037286,
      "balance_loss_clip": 1.06544602,
      "balance_loss_mlp": 1.02802944,
      "epoch": 0.24132748151265557,
      "flos": 18624926246400.0,
      "grad_norm": 2.9514632874679387,
      "language_loss": 0.86923367,
      "learning_rate": 3.54968983687045e-06,
      "loss": 0.8916173,
      "num_input_tokens_seen": 42818850,
      "step": 2007,
      "time_per_iteration": 2.7108941078186035
    },
    {
      "auxiliary_loss_clip": 0.01170155,
      "auxiliary_loss_mlp": 0.01036658,
      "balance_loss_clip": 1.0614382,
      "balance_loss_mlp": 1.02663851,
      "epoch": 0.24144772440329465,
      "flos": 15267673664640.0,
      "grad_norm": 3.138491234260582,
      "language_loss": 0.8997947,
      "learning_rate": 3.549197291346872e-06,
      "loss": 0.92186284,
      "num_input_tokens_seen": 42835375,
      "step": 2008,
      "time_per_iteration": 2.769010543823242
    },
    {
      "auxiliary_loss_clip": 0.01159544,
      "auxiliary_loss_mlp": 0.00763114,
      "balance_loss_clip": 1.05880773,
      "balance_loss_mlp": 1.00060415,
      "epoch": 0.24156796729393373,
      "flos": 24024274842240.0,
      "grad_norm": 4.654710299407884,
      "language_loss": 0.79180533,
      "learning_rate": 3.548704510811297e-06,
      "loss": 0.81103194,
      "num_input_tokens_seen": 42854570,
      "step": 2009,
      "time_per_iteration": 2.7848968505859375
    },
    {
      "auxiliary_loss_clip": 0.01161231,
      "auxiliary_loss_mlp": 0.01027142,
      "balance_loss_clip": 1.05996799,
      "balance_loss_mlp": 1.01777232,
      "epoch": 0.24168821018457284,
      "flos": 26286790665600.0,
      "grad_norm": 2.3407936756034413,
      "language_loss": 0.74328661,
      "learning_rate": 3.5482114953384787e-06,
      "loss": 0.76517034,
      "num_input_tokens_seen": 42873800,
      "step": 2010,
      "time_per_iteration": 2.832789897918701
    },
    {
      "auxiliary_loss_clip": 0.01178222,
      "auxiliary_loss_mlp": 0.01031393,
      "balance_loss_clip": 1.06372082,
      "balance_loss_mlp": 1.0226903,
      "epoch": 0.24180845307521193,
      "flos": 18223193560320.0,
      "grad_norm": 2.143538879517564,
      "language_loss": 0.84666753,
      "learning_rate": 3.5477182450032077e-06,
      "loss": 0.86876363,
      "num_input_tokens_seen": 42892400,
      "step": 2011,
      "time_per_iteration": 2.7500526905059814
    },
    {
      "auxiliary_loss_clip": 0.01172415,
      "auxiliary_loss_mlp": 0.01028944,
      "balance_loss_clip": 1.06386995,
      "balance_loss_mlp": 1.01960957,
      "epoch": 0.241928695965851,
      "flos": 20449260057600.0,
      "grad_norm": 3.3979996587170254,
      "language_loss": 0.83694673,
      "learning_rate": 3.5472247598803097e-06,
      "loss": 0.85896027,
      "num_input_tokens_seen": 42911745,
      "step": 2012,
      "time_per_iteration": 3.9828221797943115
    },
    {
      "auxiliary_loss_clip": 0.01173419,
      "auxiliary_loss_mlp": 0.01038379,
      "balance_loss_clip": 1.05647612,
      "balance_loss_mlp": 1.02761412,
      "epoch": 0.24204893885649012,
      "flos": 25556475340800.0,
      "grad_norm": 6.528391198888471,
      "language_loss": 0.85058528,
      "learning_rate": 3.546731040044645e-06,
      "loss": 0.87270325,
      "num_input_tokens_seen": 42926915,
      "step": 2013,
      "time_per_iteration": 2.7565200328826904
    },
    {
      "auxiliary_loss_clip": 0.01157942,
      "auxiliary_loss_mlp": 0.01034449,
      "balance_loss_clip": 1.05780554,
      "balance_loss_mlp": 1.02486455,
      "epoch": 0.2421691817471292,
      "flos": 30660207004800.0,
      "grad_norm": 2.1195467390404485,
      "language_loss": 0.75722605,
      "learning_rate": 3.546237085571112e-06,
      "loss": 0.77914995,
      "num_input_tokens_seen": 42945350,
      "step": 2014,
      "time_per_iteration": 3.763009786605835
    },
    {
      "auxiliary_loss_clip": 0.0121832,
      "auxiliary_loss_mlp": 0.01037474,
      "balance_loss_clip": 1.06802833,
      "balance_loss_mlp": 1.02747178,
      "epoch": 0.24228942463776829,
      "flos": 21945011230080.0,
      "grad_norm": 2.2854161381599734,
      "language_loss": 0.72788739,
      "learning_rate": 3.5457428965346425e-06,
      "loss": 0.75044537,
      "num_input_tokens_seen": 42964290,
      "step": 2015,
      "time_per_iteration": 2.7180356979370117
    },
    {
      "auxiliary_loss_clip": 0.01171056,
      "auxiliary_loss_mlp": 0.01035957,
      "balance_loss_clip": 1.06277645,
      "balance_loss_mlp": 1.0264622,
      "epoch": 0.2424096675284074,
      "flos": 33984493879680.0,
      "grad_norm": 1.9600470260993628,
      "language_loss": 0.749084,
      "learning_rate": 3.545248473010205e-06,
      "loss": 0.77115417,
      "num_input_tokens_seen": 42987095,
      "step": 2016,
      "time_per_iteration": 2.8148505687713623
    },
    {
      "auxiliary_loss_clip": 0.01207522,
      "auxiliary_loss_mlp": 0.0076438,
      "balance_loss_clip": 1.06584406,
      "balance_loss_mlp": 1.00054288,
      "epoch": 0.24252991041904648,
      "flos": 21653416621440.0,
      "grad_norm": 2.6881030029168604,
      "language_loss": 0.87923622,
      "learning_rate": 3.544753815072802e-06,
      "loss": 0.89895529,
      "num_input_tokens_seen": 43005750,
      "step": 2017,
      "time_per_iteration": 3.7138993740081787
    },
    {
      "auxiliary_loss_clip": 0.01203125,
      "auxiliary_loss_mlp": 0.01034707,
      "balance_loss_clip": 1.06410503,
      "balance_loss_mlp": 1.02533698,
      "epoch": 0.24265015330968556,
      "flos": 21870065502720.0,
      "grad_norm": 3.293261438283625,
      "language_loss": 0.88121378,
      "learning_rate": 3.544258922797474e-06,
      "loss": 0.90359211,
      "num_input_tokens_seen": 43023870,
      "step": 2018,
      "time_per_iteration": 2.7386252880096436
    },
    {
      "auxiliary_loss_clip": 0.01178743,
      "auxiliary_loss_mlp": 0.01042087,
      "balance_loss_clip": 1.06204486,
      "balance_loss_mlp": 1.03253818,
      "epoch": 0.24277039620032465,
      "flos": 25628260671360.0,
      "grad_norm": 1.8550664198174411,
      "language_loss": 0.78583801,
      "learning_rate": 3.543763796259295e-06,
      "loss": 0.80804622,
      "num_input_tokens_seen": 43043825,
      "step": 2019,
      "time_per_iteration": 2.7562594413757324
    },
    {
      "auxiliary_loss_clip": 0.01203706,
      "auxiliary_loss_mlp": 0.01035459,
      "balance_loss_clip": 1.06426477,
      "balance_loss_mlp": 1.02575493,
      "epoch": 0.24289063909096376,
      "flos": 26286575184000.0,
      "grad_norm": 4.425279639964545,
      "language_loss": 0.9091146,
      "learning_rate": 3.5432684355333754e-06,
      "loss": 0.93150628,
      "num_input_tokens_seen": 43062480,
      "step": 2020,
      "time_per_iteration": 2.776545286178589
    },
    {
      "auxiliary_loss_clip": 0.01158512,
      "auxiliary_loss_mlp": 0.0103054,
      "balance_loss_clip": 1.05836391,
      "balance_loss_mlp": 1.02176595,
      "epoch": 0.24301088198160284,
      "flos": 25075056332160.0,
      "grad_norm": 6.906074343079,
      "language_loss": 0.77016968,
      "learning_rate": 3.5427728406948613e-06,
      "loss": 0.79206014,
      "num_input_tokens_seen": 43081595,
      "step": 2021,
      "time_per_iteration": 2.7707953453063965
    },
    {
      "auxiliary_loss_clip": 0.01104459,
      "auxiliary_loss_mlp": 0.00999382,
      "balance_loss_clip": 1.03111207,
      "balance_loss_mlp": 0.99758214,
      "epoch": 0.24313112487224192,
      "flos": 69900948673920.0,
      "grad_norm": 0.7549703351760663,
      "language_loss": 0.57894206,
      "learning_rate": 3.542277011818934e-06,
      "loss": 0.59998053,
      "num_input_tokens_seen": 43145430,
      "step": 2022,
      "time_per_iteration": 3.407597303390503
    },
    {
      "auxiliary_loss_clip": 0.01178499,
      "auxiliary_loss_mlp": 0.00762854,
      "balance_loss_clip": 1.06015038,
      "balance_loss_mlp": 1.00080848,
      "epoch": 0.24325136776288103,
      "flos": 40662334235520.0,
      "grad_norm": 4.714173633681863,
      "language_loss": 0.74328816,
      "learning_rate": 3.5417809489808104e-06,
      "loss": 0.76270175,
      "num_input_tokens_seen": 43167040,
      "step": 2023,
      "time_per_iteration": 2.858349084854126
    },
    {
      "auxiliary_loss_clip": 0.01176077,
      "auxiliary_loss_mlp": 0.01036421,
      "balance_loss_clip": 1.06419992,
      "balance_loss_mlp": 1.02675867,
      "epoch": 0.24337161065352012,
      "flos": 25046400257280.0,
      "grad_norm": 1.876915128101451,
      "language_loss": 0.72652495,
      "learning_rate": 3.5412846522557422e-06,
      "loss": 0.74864995,
      "num_input_tokens_seen": 43187930,
      "step": 2024,
      "time_per_iteration": 2.8107287883758545
    },
    {
      "auxiliary_loss_clip": 0.01207581,
      "auxiliary_loss_mlp": 0.00763497,
      "balance_loss_clip": 1.06904864,
      "balance_loss_mlp": 1.00090289,
      "epoch": 0.2434918535441592,
      "flos": 18661160090880.0,
      "grad_norm": 2.074187753475717,
      "language_loss": 0.74296141,
      "learning_rate": 3.540788121719018e-06,
      "loss": 0.76267219,
      "num_input_tokens_seen": 43206350,
      "step": 2025,
      "time_per_iteration": 2.702608585357666
    },
    {
      "auxiliary_loss_clip": 0.0121765,
      "auxiliary_loss_mlp": 0.01039141,
      "balance_loss_clip": 1.06762791,
      "balance_loss_mlp": 1.02930021,
      "epoch": 0.24361209643479828,
      "flos": 23915142345600.0,
      "grad_norm": 2.1249744489690854,
      "language_loss": 0.82069433,
      "learning_rate": 3.5402913574459604e-06,
      "loss": 0.84326226,
      "num_input_tokens_seen": 43226255,
      "step": 2026,
      "time_per_iteration": 2.681980609893799
    },
    {
      "auxiliary_loss_clip": 0.01185235,
      "auxiliary_loss_mlp": 0.01037905,
      "balance_loss_clip": 1.06000888,
      "balance_loss_mlp": 1.02872586,
      "epoch": 0.2437323393254374,
      "flos": 28657505232000.0,
      "grad_norm": 1.9131312807113745,
      "language_loss": 0.8608762,
      "learning_rate": 3.5397943595119297e-06,
      "loss": 0.88310754,
      "num_input_tokens_seen": 43247675,
      "step": 2027,
      "time_per_iteration": 2.84261417388916
    },
    {
      "auxiliary_loss_clip": 0.01192089,
      "auxiliary_loss_mlp": 0.01037905,
      "balance_loss_clip": 1.06293058,
      "balance_loss_mlp": 1.0284158,
      "epoch": 0.24385258221607647,
      "flos": 23550325862400.0,
      "grad_norm": 2.427893053347569,
      "language_loss": 0.77841556,
      "learning_rate": 3.5392971279923177e-06,
      "loss": 0.80071551,
      "num_input_tokens_seen": 43265895,
      "step": 2028,
      "time_per_iteration": 2.738255739212036
    },
    {
      "auxiliary_loss_clip": 0.01157238,
      "auxiliary_loss_mlp": 0.01037743,
      "balance_loss_clip": 1.05409813,
      "balance_loss_mlp": 1.02753854,
      "epoch": 0.24397282510671556,
      "flos": 25336091445120.0,
      "grad_norm": 10.935123622125893,
      "language_loss": 0.82770789,
      "learning_rate": 3.5387996629625557e-06,
      "loss": 0.84965771,
      "num_input_tokens_seen": 43283485,
      "step": 2029,
      "time_per_iteration": 2.793569326400757
    },
    {
      "auxiliary_loss_clip": 0.01052963,
      "auxiliary_loss_mlp": 0.0100445,
      "balance_loss_clip": 1.02799642,
      "balance_loss_mlp": 1.00268579,
      "epoch": 0.24409306799735467,
      "flos": 65187421430400.0,
      "grad_norm": 0.8104099121153387,
      "language_loss": 0.54975736,
      "learning_rate": 3.5383019644981083e-06,
      "loss": 0.57033145,
      "num_input_tokens_seen": 43347180,
      "step": 2030,
      "time_per_iteration": 3.5285537242889404
    },
    {
      "auxiliary_loss_clip": 0.01187698,
      "auxiliary_loss_mlp": 0.01034581,
      "balance_loss_clip": 1.06563842,
      "balance_loss_mlp": 1.02503264,
      "epoch": 0.24421331088799375,
      "flos": 19537093152000.0,
      "grad_norm": 3.8310591849685545,
      "language_loss": 0.73840725,
      "learning_rate": 3.5378040326744763e-06,
      "loss": 0.76063007,
      "num_input_tokens_seen": 43366665,
      "step": 2031,
      "time_per_iteration": 2.864470958709717
    },
    {
      "auxiliary_loss_clip": 0.0116387,
      "auxiliary_loss_mlp": 0.01035888,
      "balance_loss_clip": 1.05980849,
      "balance_loss_mlp": 1.02701247,
      "epoch": 0.24433355377863283,
      "flos": 21068575378560.0,
      "grad_norm": 2.5796632690431545,
      "language_loss": 0.85740006,
      "learning_rate": 3.5373058675671946e-06,
      "loss": 0.87939763,
      "num_input_tokens_seen": 43384670,
      "step": 2032,
      "time_per_iteration": 2.837864398956299
    },
    {
      "auxiliary_loss_clip": 0.0114461,
      "auxiliary_loss_mlp": 0.01039129,
      "balance_loss_clip": 1.05530095,
      "balance_loss_mlp": 1.02915144,
      "epoch": 0.24445379666927192,
      "flos": 22637189289600.0,
      "grad_norm": 3.826497431399797,
      "language_loss": 0.72426367,
      "learning_rate": 3.536807469251836e-06,
      "loss": 0.74610102,
      "num_input_tokens_seen": 43403825,
      "step": 2033,
      "time_per_iteration": 2.802086591720581
    },
    {
      "auxiliary_loss_clip": 0.0108049,
      "auxiliary_loss_mlp": 0.01036561,
      "balance_loss_clip": 1.0407443,
      "balance_loss_mlp": 1.0266726,
      "epoch": 0.24457403955991103,
      "flos": 21251612108160.0,
      "grad_norm": 2.1428287296978774,
      "language_loss": 0.8324393,
      "learning_rate": 3.5363088378040055e-06,
      "loss": 0.8536098,
      "num_input_tokens_seen": 43422715,
      "step": 2034,
      "time_per_iteration": 2.9812610149383545
    },
    {
      "auxiliary_loss_clip": 0.01105055,
      "auxiliary_loss_mlp": 0.00755006,
      "balance_loss_clip": 1.02743816,
      "balance_loss_mlp": 1.00164318,
      "epoch": 0.2446942824505501,
      "flos": 66997820764800.0,
      "grad_norm": 0.7644152676642546,
      "language_loss": 0.64370102,
      "learning_rate": 3.5358099732993463e-06,
      "loss": 0.66230172,
      "num_input_tokens_seen": 43481825,
      "step": 2035,
      "time_per_iteration": 3.405007839202881
    },
    {
      "auxiliary_loss_clip": 0.01204734,
      "auxiliary_loss_mlp": 0.01037991,
      "balance_loss_clip": 1.0690614,
      "balance_loss_mlp": 1.02875209,
      "epoch": 0.2448145253411892,
      "flos": 20411122792320.0,
      "grad_norm": 2.3154603119445683,
      "language_loss": 0.90050387,
      "learning_rate": 3.535310875813535e-06,
      "loss": 0.92293108,
      "num_input_tokens_seen": 43500220,
      "step": 2036,
      "time_per_iteration": 2.6725337505340576
    },
    {
      "auxiliary_loss_clip": 0.01188833,
      "auxiliary_loss_mlp": 0.01032554,
      "balance_loss_clip": 1.06311142,
      "balance_loss_mlp": 1.02319646,
      "epoch": 0.2449347682318283,
      "flos": 28804739080320.0,
      "grad_norm": 3.103346672641278,
      "language_loss": 0.81822777,
      "learning_rate": 3.5348115454222843e-06,
      "loss": 0.84044164,
      "num_input_tokens_seen": 43522805,
      "step": 2037,
      "time_per_iteration": 2.7793095111846924
    },
    {
      "auxiliary_loss_clip": 0.01190174,
      "auxiliary_loss_mlp": 0.01038948,
      "balance_loss_clip": 1.06246805,
      "balance_loss_mlp": 1.02891016,
      "epoch": 0.2450550111224674,
      "flos": 22528990546560.0,
      "grad_norm": 2.4144893581568367,
      "language_loss": 0.86617053,
      "learning_rate": 3.5343119822013425e-06,
      "loss": 0.88846171,
      "num_input_tokens_seen": 43541915,
      "step": 2038,
      "time_per_iteration": 5.022164344787598
    },
    {
      "auxiliary_loss_clip": 0.0118138,
      "auxiliary_loss_mlp": 0.00763981,
      "balance_loss_clip": 1.0650965,
      "balance_loss_mlp": 1.0009979,
      "epoch": 0.24517525401310647,
      "flos": 21759137326080.0,
      "grad_norm": 1.7645962528355155,
      "language_loss": 0.77802819,
      "learning_rate": 3.533812186226493e-06,
      "loss": 0.79748178,
      "num_input_tokens_seen": 43562625,
      "step": 2039,
      "time_per_iteration": 2.7466742992401123
    },
    {
      "auxiliary_loss_clip": 0.01196534,
      "auxiliary_loss_mlp": 0.01039915,
      "balance_loss_clip": 1.06415427,
      "balance_loss_mlp": 1.03005648,
      "epoch": 0.24529549690374555,
      "flos": 25043311687680.0,
      "grad_norm": 1.9919025041126455,
      "language_loss": 0.75850344,
      "learning_rate": 3.5333121575735545e-06,
      "loss": 0.78086793,
      "num_input_tokens_seen": 43582265,
      "step": 2040,
      "time_per_iteration": 3.6725142002105713
    },
    {
      "auxiliary_loss_clip": 0.01202403,
      "auxiliary_loss_mlp": 0.01032949,
      "balance_loss_clip": 1.06526971,
      "balance_loss_mlp": 1.02416933,
      "epoch": 0.24541573979438466,
      "flos": 32123638915200.0,
      "grad_norm": 2.1925785834325415,
      "language_loss": 0.75834632,
      "learning_rate": 3.532811896318381e-06,
      "loss": 0.78069985,
      "num_input_tokens_seen": 43604335,
      "step": 2041,
      "time_per_iteration": 2.7999868392944336
    },
    {
      "auxiliary_loss_clip": 0.01180024,
      "auxiliary_loss_mlp": 0.01037126,
      "balance_loss_clip": 1.06153822,
      "balance_loss_mlp": 1.02655172,
      "epoch": 0.24553598268502375,
      "flos": 31357556622720.0,
      "grad_norm": 2.4743419536441973,
      "language_loss": 0.82005328,
      "learning_rate": 3.5323114025368615e-06,
      "loss": 0.84222478,
      "num_input_tokens_seen": 43619400,
      "step": 2042,
      "time_per_iteration": 2.7911922931671143
    },
    {
      "auxiliary_loss_clip": 0.01166189,
      "auxiliary_loss_mlp": 0.01033594,
      "balance_loss_clip": 1.05984509,
      "balance_loss_mlp": 1.0242604,
      "epoch": 0.24565622557566283,
      "flos": 14027462824320.0,
      "grad_norm": 2.4515675865312505,
      "language_loss": 0.82147563,
      "learning_rate": 3.53181067630492e-06,
      "loss": 0.84347349,
      "num_input_tokens_seen": 43636870,
      "step": 2043,
      "time_per_iteration": 3.676359176635742
    },
    {
      "auxiliary_loss_clip": 0.01195361,
      "auxiliary_loss_mlp": 0.00763188,
      "balance_loss_clip": 1.06162095,
      "balance_loss_mlp": 1.0008018,
      "epoch": 0.24577646846630194,
      "flos": 16581465515520.0,
      "grad_norm": 2.0385620846759145,
      "language_loss": 0.76154184,
      "learning_rate": 3.5313097176985175e-06,
      "loss": 0.78112733,
      "num_input_tokens_seen": 43655180,
      "step": 2044,
      "time_per_iteration": 2.6671953201293945
    },
    {
      "auxiliary_loss_clip": 0.01191703,
      "auxiliary_loss_mlp": 0.01041888,
      "balance_loss_clip": 1.06122506,
      "balance_loss_mlp": 1.03252983,
      "epoch": 0.24589671135694102,
      "flos": 18807424272000.0,
      "grad_norm": 6.226390227091089,
      "language_loss": 0.81281096,
      "learning_rate": 3.5308085267936482e-06,
      "loss": 0.8351469,
      "num_input_tokens_seen": 43672895,
      "step": 2045,
      "time_per_iteration": 2.761835813522339
    },
    {
      "auxiliary_loss_clip": 0.01197713,
      "auxiliary_loss_mlp": 0.01036492,
      "balance_loss_clip": 1.06562567,
      "balance_loss_mlp": 1.02833223,
      "epoch": 0.2460169542475801,
      "flos": 19938538529280.0,
      "grad_norm": 1.809508484662963,
      "language_loss": 0.89934194,
      "learning_rate": 3.530307103666342e-06,
      "loss": 0.92168403,
      "num_input_tokens_seen": 43691975,
      "step": 2046,
      "time_per_iteration": 2.660688638687134
    },
    {
      "auxiliary_loss_clip": 0.01157641,
      "auxiliary_loss_mlp": 0.01037218,
      "balance_loss_clip": 1.05413961,
      "balance_loss_mlp": 1.02716291,
      "epoch": 0.24613719713821922,
      "flos": 24171221381760.0,
      "grad_norm": 2.278031812418075,
      "language_loss": 0.80444872,
      "learning_rate": 3.5298054483926658e-06,
      "loss": 0.8263973,
      "num_input_tokens_seen": 43712670,
      "step": 2047,
      "time_per_iteration": 2.7751641273498535
    },
    {
      "auxiliary_loss_clip": 0.01133437,
      "auxiliary_loss_mlp": 0.00763303,
      "balance_loss_clip": 1.05195379,
      "balance_loss_mlp": 1.00085592,
      "epoch": 0.2462574400288583,
      "flos": 30221055325440.0,
      "grad_norm": 2.3422393141095115,
      "language_loss": 0.82977533,
      "learning_rate": 3.5293035610487187e-06,
      "loss": 0.84874272,
      "num_input_tokens_seen": 43732035,
      "step": 2048,
      "time_per_iteration": 2.8402678966522217
    },
    {
      "auxiliary_loss_clip": 0.01081864,
      "auxiliary_loss_mlp": 0.01009303,
      "balance_loss_clip": 1.02346766,
      "balance_loss_mlp": 1.0075984,
      "epoch": 0.24637768291949738,
      "flos": 68943030819840.0,
      "grad_norm": 0.7310695572864089,
      "language_loss": 0.61916494,
      "learning_rate": 3.5288014417106374e-06,
      "loss": 0.64007664,
      "num_input_tokens_seen": 43798055,
      "step": 2049,
      "time_per_iteration": 3.321866273880005
    },
    {
      "auxiliary_loss_clip": 0.0117444,
      "auxiliary_loss_mlp": 0.01033193,
      "balance_loss_clip": 1.06174254,
      "balance_loss_mlp": 1.02343011,
      "epoch": 0.24649792581013646,
      "flos": 34383999922560.0,
      "grad_norm": 9.168049808737093,
      "language_loss": 0.75356299,
      "learning_rate": 3.528299090454593e-06,
      "loss": 0.77563936,
      "num_input_tokens_seen": 43818590,
      "step": 2050,
      "time_per_iteration": 2.8778188228607178
    },
    {
      "auxiliary_loss_clip": 0.01148902,
      "auxiliary_loss_mlp": 0.01035612,
      "balance_loss_clip": 1.05697179,
      "balance_loss_mlp": 1.02517498,
      "epoch": 0.24661816870077558,
      "flos": 19680448331520.0,
      "grad_norm": 2.2053184035174906,
      "language_loss": 0.82908994,
      "learning_rate": 3.527796507356792e-06,
      "loss": 0.8509351,
      "num_input_tokens_seen": 43832480,
      "step": 2051,
      "time_per_iteration": 2.7136893272399902
    },
    {
      "auxiliary_loss_clip": 0.01150313,
      "auxiliary_loss_mlp": 0.01040599,
      "balance_loss_clip": 1.05354595,
      "balance_loss_mlp": 1.03074002,
      "epoch": 0.24673841159141466,
      "flos": 20002279213440.0,
      "grad_norm": 2.5838352759021284,
      "language_loss": 0.90894133,
      "learning_rate": 3.527293692493475e-06,
      "loss": 0.93085045,
      "num_input_tokens_seen": 43848345,
      "step": 2052,
      "time_per_iteration": 2.7242965698242188
    },
    {
      "auxiliary_loss_clip": 0.01222371,
      "auxiliary_loss_mlp": 0.01038682,
      "balance_loss_clip": 1.06671214,
      "balance_loss_mlp": 1.02915072,
      "epoch": 0.24685865448205374,
      "flos": 21646593037440.0,
      "grad_norm": 7.381064619140641,
      "language_loss": 0.73259079,
      "learning_rate": 3.52679064594092e-06,
      "loss": 0.75520134,
      "num_input_tokens_seen": 43865685,
      "step": 2053,
      "time_per_iteration": 2.5682852268218994
    },
    {
      "auxiliary_loss_clip": 0.01183508,
      "auxiliary_loss_mlp": 0.01030226,
      "balance_loss_clip": 1.05969393,
      "balance_loss_mlp": 1.02079666,
      "epoch": 0.24697889737269285,
      "flos": 17960470508160.0,
      "grad_norm": 7.739246763125188,
      "language_loss": 0.7559908,
      "learning_rate": 3.5262873677754375e-06,
      "loss": 0.77812809,
      "num_input_tokens_seen": 43883690,
      "step": 2054,
      "time_per_iteration": 2.7276761531829834
    },
    {
      "auxiliary_loss_clip": 0.01145628,
      "auxiliary_loss_mlp": 0.01038498,
      "balance_loss_clip": 1.05842543,
      "balance_loss_mlp": 1.02900863,
      "epoch": 0.24709914026333193,
      "flos": 27344611221120.0,
      "grad_norm": 2.146674895289124,
      "language_loss": 0.80645549,
      "learning_rate": 3.5257838580733745e-06,
      "loss": 0.82829678,
      "num_input_tokens_seen": 43903295,
      "step": 2055,
      "time_per_iteration": 2.7679100036621094
    },
    {
      "auxiliary_loss_clip": 0.01186696,
      "auxiliary_loss_mlp": 0.01040365,
      "balance_loss_clip": 1.0653559,
      "balance_loss_mlp": 1.0308578,
      "epoch": 0.24721938315397102,
      "flos": 19275519335040.0,
      "grad_norm": 6.569706323553933,
      "language_loss": 0.87442219,
      "learning_rate": 3.5252801169111138e-06,
      "loss": 0.89669275,
      "num_input_tokens_seen": 43920960,
      "step": 2056,
      "time_per_iteration": 2.7131187915802
    },
    {
      "auxiliary_loss_clip": 0.01150474,
      "auxiliary_loss_mlp": 0.01039722,
      "balance_loss_clip": 1.05632222,
      "balance_loss_mlp": 1.0303942,
      "epoch": 0.2473396260446101,
      "flos": 23185796688000.0,
      "grad_norm": 1.9881551512931395,
      "language_loss": 0.8008936,
      "learning_rate": 3.524776144365072e-06,
      "loss": 0.82279551,
      "num_input_tokens_seen": 43939415,
      "step": 2057,
      "time_per_iteration": 2.7306811809539795
    },
    {
      "auxiliary_loss_clip": 0.01188407,
      "auxiliary_loss_mlp": 0.01040191,
      "balance_loss_clip": 1.06355882,
      "balance_loss_mlp": 1.03048706,
      "epoch": 0.2474598689352492,
      "flos": 21142443697920.0,
      "grad_norm": 2.6638121102322474,
      "language_loss": 0.7939111,
      "learning_rate": 3.5242719405117016e-06,
      "loss": 0.81619704,
      "num_input_tokens_seen": 43959220,
      "step": 2058,
      "time_per_iteration": 2.6969449520111084
    },
    {
      "auxiliary_loss_clip": 0.01222359,
      "auxiliary_loss_mlp": 0.01045814,
      "balance_loss_clip": 1.06858242,
      "balance_loss_mlp": 1.03594303,
      "epoch": 0.2475801118258883,
      "flos": 21648352803840.0,
      "grad_norm": 9.465719030322383,
      "language_loss": 0.75275171,
      "learning_rate": 3.5237675054274893e-06,
      "loss": 0.77543348,
      "num_input_tokens_seen": 43978420,
      "step": 2059,
      "time_per_iteration": 2.6200826168060303
    },
    {
      "auxiliary_loss_clip": 0.01223612,
      "auxiliary_loss_mlp": 0.010477,
      "balance_loss_clip": 1.06900775,
      "balance_loss_mlp": 1.03746021,
      "epoch": 0.24770035471652738,
      "flos": 22674500542080.0,
      "grad_norm": 4.696837059785322,
      "language_loss": 0.80402732,
      "learning_rate": 3.5232628391889584e-06,
      "loss": 0.82674044,
      "num_input_tokens_seen": 43996710,
      "step": 2060,
      "time_per_iteration": 2.672823667526245
    },
    {
      "auxiliary_loss_clip": 0.01170648,
      "auxiliary_loss_mlp": 0.01038284,
      "balance_loss_clip": 1.06028616,
      "balance_loss_mlp": 1.02903986,
      "epoch": 0.2478205976071665,
      "flos": 22163814927360.0,
      "grad_norm": 3.1397445639143617,
      "language_loss": 0.64280885,
      "learning_rate": 3.522757941872666e-06,
      "loss": 0.66489816,
      "num_input_tokens_seen": 44014865,
      "step": 2061,
      "time_per_iteration": 2.736926317214966
    },
    {
      "auxiliary_loss_clip": 0.01167082,
      "auxiliary_loss_mlp": 0.0105173,
      "balance_loss_clip": 1.0619123,
      "balance_loss_mlp": 1.04171014,
      "epoch": 0.24794084049780557,
      "flos": 24973106555520.0,
      "grad_norm": 1.888505408984537,
      "language_loss": 0.82658458,
      "learning_rate": 3.5222528135552042e-06,
      "loss": 0.84877264,
      "num_input_tokens_seen": 44036325,
      "step": 2062,
      "time_per_iteration": 2.7480287551879883
    },
    {
      "auxiliary_loss_clip": 0.01204173,
      "auxiliary_loss_mlp": 0.01038554,
      "balance_loss_clip": 1.06723309,
      "balance_loss_mlp": 1.02833152,
      "epoch": 0.24806108338844465,
      "flos": 18296379521280.0,
      "grad_norm": 2.7884311131724586,
      "language_loss": 0.81001616,
      "learning_rate": 3.521747454313201e-06,
      "loss": 0.83244342,
      "num_input_tokens_seen": 44055005,
      "step": 2063,
      "time_per_iteration": 3.576676368713379
    },
    {
      "auxiliary_loss_clip": 0.01150913,
      "auxiliary_loss_mlp": 0.0103816,
      "balance_loss_clip": 1.05844736,
      "balance_loss_mlp": 1.02797353,
      "epoch": 0.24818132627908374,
      "flos": 19282163351040.0,
      "grad_norm": 2.558463223538029,
      "language_loss": 0.67594844,
      "learning_rate": 3.521241864223319e-06,
      "loss": 0.6978392,
      "num_input_tokens_seen": 44073965,
      "step": 2064,
      "time_per_iteration": 3.8537042140960693
    },
    {
      "auxiliary_loss_clip": 0.01098823,
      "auxiliary_loss_mlp": 0.01009749,
      "balance_loss_clip": 1.02417791,
      "balance_loss_mlp": 1.0079608,
      "epoch": 0.24830156916972285,
      "flos": 70285837881600.0,
      "grad_norm": 0.795001395200979,
      "language_loss": 0.61876237,
      "learning_rate": 3.5207360433622552e-06,
      "loss": 0.63984811,
      "num_input_tokens_seen": 44135965,
      "step": 2065,
      "time_per_iteration": 3.2141788005828857
    },
    {
      "auxiliary_loss_clip": 0.01136718,
      "auxiliary_loss_mlp": 0.01047483,
      "balance_loss_clip": 1.05532694,
      "balance_loss_mlp": 1.0381484,
      "epoch": 0.24842181206036193,
      "flos": 40409128287360.0,
      "grad_norm": 1.774633621728927,
      "language_loss": 0.74991572,
      "learning_rate": 3.5202299918067437e-06,
      "loss": 0.77175772,
      "num_input_tokens_seen": 44159560,
      "step": 2066,
      "time_per_iteration": 3.9359402656555176
    },
    {
      "auxiliary_loss_clip": 0.01191661,
      "auxiliary_loss_mlp": 0.01039256,
      "balance_loss_clip": 1.06100774,
      "balance_loss_mlp": 1.02973163,
      "epoch": 0.248542054951001,
      "flos": 20082432412800.0,
      "grad_norm": 6.724549959720837,
      "language_loss": 0.69750923,
      "learning_rate": 3.519723709633551e-06,
      "loss": 0.71981841,
      "num_input_tokens_seen": 44178320,
      "step": 2067,
      "time_per_iteration": 2.6810457706451416
    },
    {
      "auxiliary_loss_clip": 0.01177455,
      "auxiliary_loss_mlp": 0.01042495,
      "balance_loss_clip": 1.05988812,
      "balance_loss_mlp": 1.03307152,
      "epoch": 0.24866229784164012,
      "flos": 23513948363520.0,
      "grad_norm": 2.1914648815687556,
      "language_loss": 0.83763409,
      "learning_rate": 3.519217196919479e-06,
      "loss": 0.8598336,
      "num_input_tokens_seen": 44197305,
      "step": 2068,
      "time_per_iteration": 2.6967737674713135
    },
    {
      "auxiliary_loss_clip": 0.01167695,
      "auxiliary_loss_mlp": 0.01043205,
      "balance_loss_clip": 1.05833578,
      "balance_loss_mlp": 1.03340566,
      "epoch": 0.2487825407322792,
      "flos": 19865101173120.0,
      "grad_norm": 2.095694969031014,
      "language_loss": 0.73936701,
      "learning_rate": 3.518710453741367e-06,
      "loss": 0.76147598,
      "num_input_tokens_seen": 44216505,
      "step": 2069,
      "time_per_iteration": 3.8179917335510254
    },
    {
      "auxiliary_loss_clip": 0.01195338,
      "auxiliary_loss_mlp": 0.01031439,
      "balance_loss_clip": 1.06507599,
      "balance_loss_mlp": 1.02157474,
      "epoch": 0.2489027836229183,
      "flos": 22017622573440.0,
      "grad_norm": 2.320291383513601,
      "language_loss": 0.67856067,
      "learning_rate": 3.518203480176086e-06,
      "loss": 0.70082843,
      "num_input_tokens_seen": 44235435,
      "step": 2070,
      "time_per_iteration": 2.707885980606079
    },
    {
      "auxiliary_loss_clip": 0.01111228,
      "auxiliary_loss_mlp": 0.0104021,
      "balance_loss_clip": 1.04300904,
      "balance_loss_mlp": 1.03000522,
      "epoch": 0.2490230265135574,
      "flos": 23294354567040.0,
      "grad_norm": 1.9969159606912568,
      "language_loss": 0.80892926,
      "learning_rate": 3.517696276300545e-06,
      "loss": 0.83044362,
      "num_input_tokens_seen": 44256975,
      "step": 2071,
      "time_per_iteration": 2.8532865047454834
    },
    {
      "auxiliary_loss_clip": 0.01190268,
      "auxiliary_loss_mlp": 0.01030728,
      "balance_loss_clip": 1.06648278,
      "balance_loss_mlp": 1.02098846,
      "epoch": 0.24914326940419648,
      "flos": 19826784339840.0,
      "grad_norm": 3.72488657298156,
      "language_loss": 0.69745755,
      "learning_rate": 3.517188842191685e-06,
      "loss": 0.71966749,
      "num_input_tokens_seen": 44275125,
      "step": 2072,
      "time_per_iteration": 2.7220795154571533
    },
    {
      "auxiliary_loss_clip": 0.01222577,
      "auxiliary_loss_mlp": 0.01046534,
      "balance_loss_clip": 1.06786442,
      "balance_loss_mlp": 1.03640735,
      "epoch": 0.24926351229483557,
      "flos": 20229271211520.0,
      "grad_norm": 1.6920683263806706,
      "language_loss": 0.73744822,
      "learning_rate": 3.5166811779264837e-06,
      "loss": 0.76013935,
      "num_input_tokens_seen": 44295445,
      "step": 2073,
      "time_per_iteration": 2.633014440536499
    },
    {
      "auxiliary_loss_clip": 0.01122021,
      "auxiliary_loss_mlp": 0.010449,
      "balance_loss_clip": 1.04487717,
      "balance_loss_mlp": 1.03465438,
      "epoch": 0.24938375518547465,
      "flos": 23294570048640.0,
      "grad_norm": 2.668326705984164,
      "language_loss": 0.78040338,
      "learning_rate": 3.5161732835819545e-06,
      "loss": 0.80207264,
      "num_input_tokens_seen": 44314755,
      "step": 2074,
      "time_per_iteration": 2.8579137325286865
    },
    {
      "auxiliary_loss_clip": 0.01202814,
      "auxiliary_loss_mlp": 0.01039465,
      "balance_loss_clip": 1.06554615,
      "balance_loss_mlp": 1.02927268,
      "epoch": 0.24950399807611376,
      "flos": 17311673099520.0,
      "grad_norm": 2.114725655710589,
      "language_loss": 0.83434129,
      "learning_rate": 3.515665159235143e-06,
      "loss": 0.85676408,
      "num_input_tokens_seen": 44333640,
      "step": 2075,
      "time_per_iteration": 2.7190544605255127
    },
    {
      "auxiliary_loss_clip": 0.01160249,
      "auxiliary_loss_mlp": 0.00763546,
      "balance_loss_clip": 1.05572653,
      "balance_loss_mlp": 1.0005523,
      "epoch": 0.24962424096675284,
      "flos": 19024863252480.0,
      "grad_norm": 1.6587566765822026,
      "language_loss": 0.75246489,
      "learning_rate": 3.5151568049631318e-06,
      "loss": 0.77170277,
      "num_input_tokens_seen": 44352355,
      "step": 2076,
      "time_per_iteration": 2.7485527992248535
    },
    {
      "auxiliary_loss_clip": 0.01174977,
      "auxiliary_loss_mlp": 0.01044087,
      "balance_loss_clip": 1.05977035,
      "balance_loss_mlp": 1.03439522,
      "epoch": 0.24974448385739192,
      "flos": 33398790710400.0,
      "grad_norm": 1.8689507630868225,
      "language_loss": 0.80229324,
      "learning_rate": 3.5146482208430385e-06,
      "loss": 0.82448387,
      "num_input_tokens_seen": 44374185,
      "step": 2077,
      "time_per_iteration": 2.805734157562256
    },
    {
      "auxiliary_loss_clip": 0.01145059,
      "auxiliary_loss_mlp": 0.01044881,
      "balance_loss_clip": 1.05654681,
      "balance_loss_mlp": 1.03425395,
      "epoch": 0.24986472674803104,
      "flos": 30007279532160.0,
      "grad_norm": 2.2120474262032395,
      "language_loss": 0.68117434,
      "learning_rate": 3.514139406952014e-06,
      "loss": 0.70307374,
      "num_input_tokens_seen": 44396210,
      "step": 2078,
      "time_per_iteration": 2.893500328063965
    },
    {
      "auxiliary_loss_clip": 0.01207197,
      "auxiliary_loss_mlp": 0.01034811,
      "balance_loss_clip": 1.06496143,
      "balance_loss_mlp": 1.02539349,
      "epoch": 0.24998496963867012,
      "flos": 26613074833920.0,
      "grad_norm": 3.562429146271903,
      "language_loss": 0.83350778,
      "learning_rate": 3.5136303633672454e-06,
      "loss": 0.85592794,
      "num_input_tokens_seen": 44416340,
      "step": 2079,
      "time_per_iteration": 2.670997142791748
    },
    {
      "auxiliary_loss_clip": 0.01126392,
      "auxiliary_loss_mlp": 0.01039172,
      "balance_loss_clip": 1.05482054,
      "balance_loss_mlp": 1.02886605,
      "epoch": 0.25010521252930923,
      "flos": 23553989049600.0,
      "grad_norm": 1.8380111507956263,
      "language_loss": 0.74544102,
      "learning_rate": 3.5131210901659544e-06,
      "loss": 0.76709664,
      "num_input_tokens_seen": 44438095,
      "step": 2080,
      "time_per_iteration": 2.8681864738464355
    },
    {
      "auxiliary_loss_clip": 0.01153589,
      "auxiliary_loss_mlp": 0.01038839,
      "balance_loss_clip": 1.05384016,
      "balance_loss_mlp": 1.02815843,
      "epoch": 0.2502254554199483,
      "flos": 23441193365760.0,
      "grad_norm": 4.455839595492368,
      "language_loss": 0.82615542,
      "learning_rate": 3.5126115874253967e-06,
      "loss": 0.84807968,
      "num_input_tokens_seen": 44457650,
      "step": 2081,
      "time_per_iteration": 2.842315673828125
    },
    {
      "auxiliary_loss_clip": 0.01196346,
      "auxiliary_loss_mlp": 0.01042167,
      "balance_loss_clip": 1.0612812,
      "balance_loss_mlp": 1.03214693,
      "epoch": 0.2503456983105874,
      "flos": 28761681651840.0,
      "grad_norm": 2.067003416288652,
      "language_loss": 0.80660117,
      "learning_rate": 3.5121018552228644e-06,
      "loss": 0.82898629,
      "num_input_tokens_seen": 44476155,
      "step": 2082,
      "time_per_iteration": 2.6973540782928467
    },
    {
      "auxiliary_loss_clip": 0.01198946,
      "auxiliary_loss_mlp": 0.01034275,
      "balance_loss_clip": 1.06431985,
      "balance_loss_mlp": 1.02441645,
      "epoch": 0.2504659412012265,
      "flos": 18770256673920.0,
      "grad_norm": 6.1554781814219,
      "language_loss": 0.76457441,
      "learning_rate": 3.5115918936356827e-06,
      "loss": 0.7869066,
      "num_input_tokens_seen": 44492910,
      "step": 2083,
      "time_per_iteration": 2.6942241191864014
    },
    {
      "auxiliary_loss_clip": 0.01194621,
      "auxiliary_loss_mlp": 0.00764077,
      "balance_loss_clip": 1.06636333,
      "balance_loss_mlp": 1.00047517,
      "epoch": 0.25058618409186556,
      "flos": 16873383346560.0,
      "grad_norm": 3.766630729299553,
      "language_loss": 0.78950644,
      "learning_rate": 3.5110817027412123e-06,
      "loss": 0.80909348,
      "num_input_tokens_seen": 44512000,
      "step": 2084,
      "time_per_iteration": 2.6881861686706543
    },
    {
      "auxiliary_loss_clip": 0.01186467,
      "auxiliary_loss_mlp": 0.00763055,
      "balance_loss_clip": 1.06282079,
      "balance_loss_mlp": 1.00050938,
      "epoch": 0.25070642698250467,
      "flos": 24425540651520.0,
      "grad_norm": 2.2032315978993346,
      "language_loss": 0.69045258,
      "learning_rate": 3.5105712826168493e-06,
      "loss": 0.70994782,
      "num_input_tokens_seen": 44531650,
      "step": 2085,
      "time_per_iteration": 2.7989377975463867
    },
    {
      "auxiliary_loss_clip": 0.01192955,
      "auxiliary_loss_mlp": 0.01043603,
      "balance_loss_clip": 1.06343162,
      "balance_loss_mlp": 1.03366661,
      "epoch": 0.2508266698731437,
      "flos": 20260944028800.0,
      "grad_norm": 2.1888094194419123,
      "language_loss": 0.70991242,
      "learning_rate": 3.5100606333400235e-06,
      "loss": 0.73227799,
      "num_input_tokens_seen": 44548785,
      "step": 2086,
      "time_per_iteration": 2.676851749420166
    },
    {
      "auxiliary_loss_clip": 0.0120667,
      "auxiliary_loss_mlp": 0.01047809,
      "balance_loss_clip": 1.06671071,
      "balance_loss_mlp": 1.03692508,
      "epoch": 0.25094691276378284,
      "flos": 19245318975360.0,
      "grad_norm": 2.198484452715515,
      "language_loss": 0.77532125,
      "learning_rate": 3.5095497549882006e-06,
      "loss": 0.79786611,
      "num_input_tokens_seen": 44567230,
      "step": 2087,
      "time_per_iteration": 2.716472625732422
    },
    {
      "auxiliary_loss_clip": 0.01183362,
      "auxiliary_loss_mlp": 0.01040091,
      "balance_loss_clip": 1.06278205,
      "balance_loss_mlp": 1.03043544,
      "epoch": 0.25106715565442195,
      "flos": 26943237671040.0,
      "grad_norm": 4.519529958098726,
      "language_loss": 0.72863215,
      "learning_rate": 3.50903864763888e-06,
      "loss": 0.75086671,
      "num_input_tokens_seen": 44588020,
      "step": 2088,
      "time_per_iteration": 2.7126388549804688
    },
    {
      "auxiliary_loss_clip": 0.01172708,
      "auxiliary_loss_mlp": 0.01036891,
      "balance_loss_clip": 1.06002307,
      "balance_loss_mlp": 1.02703786,
      "epoch": 0.251187398545061,
      "flos": 48359570572800.0,
      "grad_norm": 2.276725103174622,
      "language_loss": 0.76713479,
      "learning_rate": 3.5085273113695965e-06,
      "loss": 0.78923082,
      "num_input_tokens_seen": 44612590,
      "step": 2089,
      "time_per_iteration": 3.921560287475586
    },
    {
      "auxiliary_loss_clip": 0.01208068,
      "auxiliary_loss_mlp": 0.00763457,
      "balance_loss_clip": 1.06428552,
      "balance_loss_mlp": 1.00050175,
      "epoch": 0.2513076414357001,
      "flos": 27016100409600.0,
      "grad_norm": 2.2273547165048555,
      "language_loss": 0.78680515,
      "learning_rate": 3.508015746257919e-06,
      "loss": 0.8065204,
      "num_input_tokens_seen": 44631630,
      "step": 2090,
      "time_per_iteration": 3.689624547958374
    },
    {
      "auxiliary_loss_clip": 0.01159875,
      "auxiliary_loss_mlp": 0.0103557,
      "balance_loss_clip": 1.06098831,
      "balance_loss_mlp": 1.02512693,
      "epoch": 0.2514278843263392,
      "flos": 19463619882240.0,
      "grad_norm": 2.6922122743673635,
      "language_loss": 0.83700746,
      "learning_rate": 3.5075039523814518e-06,
      "loss": 0.85896194,
      "num_input_tokens_seen": 44650820,
      "step": 2091,
      "time_per_iteration": 2.9272961616516113
    },
    {
      "auxiliary_loss_clip": 0.01189271,
      "auxiliary_loss_mlp": 0.01039725,
      "balance_loss_clip": 1.0647912,
      "balance_loss_mlp": 1.02842963,
      "epoch": 0.2515481272169783,
      "flos": 16866092885760.0,
      "grad_norm": 10.792563571521395,
      "language_loss": 0.81612051,
      "learning_rate": 3.506991929817834e-06,
      "loss": 0.83841044,
      "num_input_tokens_seen": 44667540,
      "step": 2092,
      "time_per_iteration": 3.610879898071289
    },
    {
      "auxiliary_loss_clip": 0.01159268,
      "auxiliary_loss_mlp": 0.01035634,
      "balance_loss_clip": 1.06396878,
      "balance_loss_mlp": 1.02700913,
      "epoch": 0.2516683701076174,
      "flos": 23732464752000.0,
      "grad_norm": 2.23638037242975,
      "language_loss": 0.83030021,
      "learning_rate": 3.506479678644738e-06,
      "loss": 0.85224926,
      "num_input_tokens_seen": 44687935,
      "step": 2093,
      "time_per_iteration": 2.7844135761260986
    },
    {
      "auxiliary_loss_clip": 0.01220772,
      "auxiliary_loss_mlp": 0.01045845,
      "balance_loss_clip": 1.06803,
      "balance_loss_mlp": 1.0353483,
      "epoch": 0.2517886129982565,
      "flos": 27635954434560.0,
      "grad_norm": 2.4525934233952946,
      "language_loss": 0.74397987,
      "learning_rate": 3.505967198939873e-06,
      "loss": 0.76664603,
      "num_input_tokens_seen": 44704975,
      "step": 2094,
      "time_per_iteration": 2.6482980251312256
    },
    {
      "auxiliary_loss_clip": 0.01169977,
      "auxiliary_loss_mlp": 0.01036433,
      "balance_loss_clip": 1.05953074,
      "balance_loss_mlp": 1.02633035,
      "epoch": 0.25190885588889556,
      "flos": 38104596529920.0,
      "grad_norm": 2.7022200218892443,
      "language_loss": 0.78064209,
      "learning_rate": 3.5054544907809813e-06,
      "loss": 0.80270612,
      "num_input_tokens_seen": 44725475,
      "step": 2095,
      "time_per_iteration": 3.791110038757324
    },
    {
      "auxiliary_loss_clip": 0.01192421,
      "auxiliary_loss_mlp": 0.01036461,
      "balance_loss_clip": 1.06447101,
      "balance_loss_mlp": 1.02602386,
      "epoch": 0.25202909877953467,
      "flos": 22269894768000.0,
      "grad_norm": 2.389365320059419,
      "language_loss": 0.80921215,
      "learning_rate": 3.50494155424584e-06,
      "loss": 0.83150101,
      "num_input_tokens_seen": 44744380,
      "step": 2096,
      "time_per_iteration": 2.7285122871398926
    },
    {
      "auxiliary_loss_clip": 0.01147504,
      "auxiliary_loss_mlp": 0.01037665,
      "balance_loss_clip": 1.05787683,
      "balance_loss_mlp": 1.02763999,
      "epoch": 0.2521493416701738,
      "flos": 21761759018880.0,
      "grad_norm": 1.8845545360782843,
      "language_loss": 0.83380312,
      "learning_rate": 3.504428389412262e-06,
      "loss": 0.85565478,
      "num_input_tokens_seen": 44765190,
      "step": 2097,
      "time_per_iteration": 2.7788214683532715
    },
    {
      "auxiliary_loss_clip": 0.01112739,
      "auxiliary_loss_mlp": 0.01042632,
      "balance_loss_clip": 1.04310989,
      "balance_loss_mlp": 1.03213537,
      "epoch": 0.25226958456081283,
      "flos": 27746738956800.0,
      "grad_norm": 2.6490080731400223,
      "language_loss": 0.73364282,
      "learning_rate": 3.5039149963580927e-06,
      "loss": 0.75519657,
      "num_input_tokens_seen": 44785210,
      "step": 2098,
      "time_per_iteration": 2.9082579612731934
    },
    {
      "auxiliary_loss_clip": 0.0118919,
      "auxiliary_loss_mlp": 0.01045555,
      "balance_loss_clip": 1.06375933,
      "balance_loss_mlp": 1.03560102,
      "epoch": 0.25238982745145194,
      "flos": 30732171903360.0,
      "grad_norm": 2.215227889412174,
      "language_loss": 0.70337665,
      "learning_rate": 3.503401375161215e-06,
      "loss": 0.7257241,
      "num_input_tokens_seen": 44804955,
      "step": 2099,
      "time_per_iteration": 2.7374677658081055
    },
    {
      "auxiliary_loss_clip": 0.0110071,
      "auxiliary_loss_mlp": 0.00764166,
      "balance_loss_clip": 1.05119979,
      "balance_loss_mlp": 1.00045252,
      "epoch": 0.252510070342091,
      "flos": 20266331068800.0,
      "grad_norm": 1.6784385507751962,
      "language_loss": 0.8412292,
      "learning_rate": 3.502887525899544e-06,
      "loss": 0.85987794,
      "num_input_tokens_seen": 44823935,
      "step": 2100,
      "time_per_iteration": 2.8499016761779785
    },
    {
      "auxiliary_loss_clip": 0.01151396,
      "auxiliary_loss_mlp": 0.0103346,
      "balance_loss_clip": 1.0575726,
      "balance_loss_mlp": 1.02330303,
      "epoch": 0.2526303132327301,
      "flos": 22747399194240.0,
      "grad_norm": 1.8966306743458536,
      "language_loss": 0.82942504,
      "learning_rate": 3.50237344865103e-06,
      "loss": 0.8512736,
      "num_input_tokens_seen": 44844935,
      "step": 2101,
      "time_per_iteration": 2.8020455837249756
    },
    {
      "auxiliary_loss_clip": 0.01169027,
      "auxiliary_loss_mlp": 0.0104125,
      "balance_loss_clip": 1.06328857,
      "balance_loss_mlp": 1.03088474,
      "epoch": 0.2527505561233692,
      "flos": 30263466309120.0,
      "grad_norm": 3.153363070949564,
      "language_loss": 0.76831764,
      "learning_rate": 3.501859143493658e-06,
      "loss": 0.79042041,
      "num_input_tokens_seen": 44865565,
      "step": 2102,
      "time_per_iteration": 2.8678267002105713
    },
    {
      "auxiliary_loss_clip": 0.01095879,
      "auxiliary_loss_mlp": 0.01001815,
      "balance_loss_clip": 1.03401065,
      "balance_loss_mlp": 0.9999674,
      "epoch": 0.2528707990140083,
      "flos": 58492917164160.0,
      "grad_norm": 0.9092609720854564,
      "language_loss": 0.60490435,
      "learning_rate": 3.5013446105054488e-06,
      "loss": 0.62588131,
      "num_input_tokens_seen": 44918485,
      "step": 2103,
      "time_per_iteration": 3.0367701053619385
    },
    {
      "auxiliary_loss_clip": 0.01182318,
      "auxiliary_loss_mlp": 0.01040431,
      "balance_loss_clip": 1.0643661,
      "balance_loss_mlp": 1.03047049,
      "epoch": 0.2529910419046474,
      "flos": 24645134448000.0,
      "grad_norm": 1.9386184671375328,
      "language_loss": 0.74892151,
      "learning_rate": 3.5008298497644555e-06,
      "loss": 0.77114898,
      "num_input_tokens_seen": 44937530,
      "step": 2104,
      "time_per_iteration": 2.7221508026123047
    },
    {
      "auxiliary_loss_clip": 0.01202118,
      "auxiliary_loss_mlp": 0.01042308,
      "balance_loss_clip": 1.06447911,
      "balance_loss_mlp": 1.03187692,
      "epoch": 0.2531112847952865,
      "flos": 23842135952640.0,
      "grad_norm": 2.850806556544392,
      "language_loss": 0.88246834,
      "learning_rate": 3.500314861348767e-06,
      "loss": 0.90491265,
      "num_input_tokens_seen": 44958165,
      "step": 2105,
      "time_per_iteration": 2.7351503372192383
    },
    {
      "auxiliary_loss_clip": 0.01176843,
      "auxiliary_loss_mlp": 0.01038995,
      "balance_loss_clip": 1.06231797,
      "balance_loss_mlp": 1.02913594,
      "epoch": 0.25323152768592555,
      "flos": 16143822207360.0,
      "grad_norm": 2.3422581849547566,
      "language_loss": 0.77532637,
      "learning_rate": 3.499799645336507e-06,
      "loss": 0.79748476,
      "num_input_tokens_seen": 44975060,
      "step": 2106,
      "time_per_iteration": 2.646540641784668
    },
    {
      "auxiliary_loss_clip": 0.01217482,
      "auxiliary_loss_mlp": 0.01038555,
      "balance_loss_clip": 1.06718659,
      "balance_loss_mlp": 1.0297811,
      "epoch": 0.25335177057656466,
      "flos": 28405161210240.0,
      "grad_norm": 1.55570543217678,
      "language_loss": 0.86936927,
      "learning_rate": 3.4992842018058336e-06,
      "loss": 0.89192963,
      "num_input_tokens_seen": 44997960,
      "step": 2107,
      "time_per_iteration": 2.7377612590789795
    },
    {
      "auxiliary_loss_clip": 0.01171004,
      "auxiliary_loss_mlp": 0.00764264,
      "balance_loss_clip": 1.06261826,
      "balance_loss_mlp": 1.00054812,
      "epoch": 0.25347201346720377,
      "flos": 18799666934400.0,
      "grad_norm": 2.3656712757947482,
      "language_loss": 0.88715541,
      "learning_rate": 3.4987685308349384e-06,
      "loss": 0.90650815,
      "num_input_tokens_seen": 45015690,
      "step": 2108,
      "time_per_iteration": 2.7370011806488037
    },
    {
      "auxiliary_loss_clip": 0.01167998,
      "auxiliary_loss_mlp": 0.01042578,
      "balance_loss_clip": 1.05910325,
      "balance_loss_mlp": 1.03236771,
      "epoch": 0.2535922563578428,
      "flos": 15815490963840.0,
      "grad_norm": 2.552824021001101,
      "language_loss": 0.61904395,
      "learning_rate": 3.4982526325020497e-06,
      "loss": 0.64114976,
      "num_input_tokens_seen": 45032660,
      "step": 2109,
      "time_per_iteration": 2.771719217300415
    },
    {
      "auxiliary_loss_clip": 0.01163108,
      "auxiliary_loss_mlp": 0.01046013,
      "balance_loss_clip": 1.06218636,
      "balance_loss_mlp": 1.03530192,
      "epoch": 0.25371249924848194,
      "flos": 16318922031360.0,
      "grad_norm": 2.6158848536325663,
      "language_loss": 0.82203281,
      "learning_rate": 3.4977365068854273e-06,
      "loss": 0.84412408,
      "num_input_tokens_seen": 45048280,
      "step": 2110,
      "time_per_iteration": 2.7350597381591797
    },
    {
      "auxiliary_loss_clip": 0.01190369,
      "auxiliary_loss_mlp": 0.01038976,
      "balance_loss_clip": 1.0668273,
      "balance_loss_mlp": 1.02813983,
      "epoch": 0.25383274213912105,
      "flos": 21761615364480.0,
      "grad_norm": 2.6672601160785057,
      "language_loss": 0.73708129,
      "learning_rate": 3.4972201540633676e-06,
      "loss": 0.75937468,
      "num_input_tokens_seen": 45067635,
      "step": 2111,
      "time_per_iteration": 2.717656373977661
    },
    {
      "auxiliary_loss_clip": 0.01152877,
      "auxiliary_loss_mlp": 0.00764478,
      "balance_loss_clip": 1.05632353,
      "balance_loss_mlp": 1.00048673,
      "epoch": 0.2539529850297601,
      "flos": 21396870708480.0,
      "grad_norm": 1.776691960116654,
      "language_loss": 0.8566398,
      "learning_rate": 3.4967035741142008e-06,
      "loss": 0.87581336,
      "num_input_tokens_seen": 45086455,
      "step": 2112,
      "time_per_iteration": 2.8756911754608154
    },
    {
      "auxiliary_loss_clip": 0.01169629,
      "auxiliary_loss_mlp": 0.01036577,
      "balance_loss_clip": 1.06469226,
      "balance_loss_mlp": 1.02782106,
      "epoch": 0.2540732279203992,
      "flos": 25228467319680.0,
      "grad_norm": 1.947402617442079,
      "language_loss": 0.82520509,
      "learning_rate": 3.4961867671162917e-06,
      "loss": 0.84726709,
      "num_input_tokens_seen": 45106385,
      "step": 2113,
      "time_per_iteration": 2.792965888977051
    },
    {
      "auxiliary_loss_clip": 0.0114635,
      "auxiliary_loss_mlp": 0.0104717,
      "balance_loss_clip": 1.05650353,
      "balance_loss_mlp": 1.03646469,
      "epoch": 0.2541934708110383,
      "flos": 19427386037760.0,
      "grad_norm": 2.465412450244462,
      "language_loss": 0.77364016,
      "learning_rate": 3.4956697331480402e-06,
      "loss": 0.79557538,
      "num_input_tokens_seen": 45124955,
      "step": 2114,
      "time_per_iteration": 2.866236686706543
    },
    {
      "auxiliary_loss_clip": 0.01131818,
      "auxiliary_loss_mlp": 0.01046498,
      "balance_loss_clip": 1.05223703,
      "balance_loss_mlp": 1.03610873,
      "epoch": 0.2543137137016774,
      "flos": 23949436855680.0,
      "grad_norm": 1.8376341758702461,
      "language_loss": 0.80012286,
      "learning_rate": 3.495152472287879e-06,
      "loss": 0.82190597,
      "num_input_tokens_seen": 45145665,
      "step": 2115,
      "time_per_iteration": 4.035470247268677
    },
    {
      "auxiliary_loss_clip": 0.01193752,
      "auxiliary_loss_mlp": 0.01041249,
      "balance_loss_clip": 1.06563342,
      "balance_loss_mlp": 1.03139043,
      "epoch": 0.2544339565923165,
      "flos": 25593283802880.0,
      "grad_norm": 2.199856051719243,
      "language_loss": 0.73793894,
      "learning_rate": 3.4946349846142766e-06,
      "loss": 0.76028895,
      "num_input_tokens_seen": 45164805,
      "step": 2116,
      "time_per_iteration": 3.736884117126465
    },
    {
      "auxiliary_loss_clip": 0.01148556,
      "auxiliary_loss_mlp": 0.01034867,
      "balance_loss_clip": 1.05581307,
      "balance_loss_mlp": 1.02513409,
      "epoch": 0.25455419948295555,
      "flos": 21689470897920.0,
      "grad_norm": 2.6820477971024985,
      "language_loss": 0.76561236,
      "learning_rate": 3.4941172702057353e-06,
      "loss": 0.7874465,
      "num_input_tokens_seen": 45184865,
      "step": 2117,
      "time_per_iteration": 2.857379198074341
    },
    {
      "auxiliary_loss_clip": 0.0121856,
      "auxiliary_loss_mlp": 0.01042036,
      "balance_loss_clip": 1.06843174,
      "balance_loss_mlp": 1.03208184,
      "epoch": 0.25467444237359466,
      "flos": 26250341339520.0,
      "grad_norm": 2.4267699878231483,
      "language_loss": 0.80976129,
      "learning_rate": 3.4935993291407924e-06,
      "loss": 0.8323673,
      "num_input_tokens_seen": 45203690,
      "step": 2118,
      "time_per_iteration": 3.725372076034546
    },
    {
      "auxiliary_loss_clip": 0.01151649,
      "auxiliary_loss_mlp": 0.01037696,
      "balance_loss_clip": 1.06008005,
      "balance_loss_mlp": 1.02743769,
      "epoch": 0.25479468526423377,
      "flos": 26979686997120.0,
      "grad_norm": 3.089610149761137,
      "language_loss": 0.71394783,
      "learning_rate": 3.4930811614980183e-06,
      "loss": 0.73584127,
      "num_input_tokens_seen": 45225385,
      "step": 2119,
      "time_per_iteration": 2.8679909706115723
    },
    {
      "auxiliary_loss_clip": 0.01163839,
      "auxiliary_loss_mlp": 0.0103785,
      "balance_loss_clip": 1.05807757,
      "balance_loss_mlp": 1.02697229,
      "epoch": 0.2549149281548728,
      "flos": 23475811098240.0,
      "grad_norm": 2.8417660549138595,
      "language_loss": 0.79270983,
      "learning_rate": 3.4925627673560198e-06,
      "loss": 0.81472671,
      "num_input_tokens_seen": 45246045,
      "step": 2120,
      "time_per_iteration": 2.8225183486938477
    },
    {
      "auxiliary_loss_clip": 0.01171961,
      "auxiliary_loss_mlp": 0.01041691,
      "balance_loss_clip": 1.06216967,
      "balance_loss_mlp": 1.03240466,
      "epoch": 0.25503517104551193,
      "flos": 25812302981760.0,
      "grad_norm": 2.9370076354305383,
      "language_loss": 0.88521123,
      "learning_rate": 3.4920441467934357e-06,
      "loss": 0.90734774,
      "num_input_tokens_seen": 45266560,
      "step": 2121,
      "time_per_iteration": 3.627880573272705
    },
    {
      "auxiliary_loss_clip": 0.01106285,
      "auxiliary_loss_mlp": 0.01036098,
      "balance_loss_clip": 1.04988122,
      "balance_loss_mlp": 1.0263114,
      "epoch": 0.25515541393615104,
      "flos": 26645106787200.0,
      "grad_norm": 2.212957031695775,
      "language_loss": 0.83094138,
      "learning_rate": 3.491525299888941e-06,
      "loss": 0.85236526,
      "num_input_tokens_seen": 45285405,
      "step": 2122,
      "time_per_iteration": 2.9612903594970703
    },
    {
      "auxiliary_loss_clip": 0.01099095,
      "auxiliary_loss_mlp": 0.01008814,
      "balance_loss_clip": 1.02376008,
      "balance_loss_mlp": 1.00671577,
      "epoch": 0.2552756568267901,
      "flos": 65955945847680.0,
      "grad_norm": 0.8690619006451205,
      "language_loss": 0.6255784,
      "learning_rate": 3.491006226721244e-06,
      "loss": 0.64665753,
      "num_input_tokens_seen": 45349615,
      "step": 2123,
      "time_per_iteration": 3.5568950176239014
    },
    {
      "auxiliary_loss_clip": 0.01208806,
      "auxiliary_loss_mlp": 0.01033824,
      "balance_loss_clip": 1.06894493,
      "balance_loss_mlp": 1.02437663,
      "epoch": 0.2553958997174292,
      "flos": 17931096161280.0,
      "grad_norm": 2.1704375919593026,
      "language_loss": 0.77734637,
      "learning_rate": 3.4904869273690882e-06,
      "loss": 0.79977274,
      "num_input_tokens_seen": 45367505,
      "step": 2124,
      "time_per_iteration": 2.7030317783355713
    },
    {
      "auxiliary_loss_clip": 0.01171388,
      "auxiliary_loss_mlp": 0.01034752,
      "balance_loss_clip": 1.05987871,
      "balance_loss_mlp": 1.02527475,
      "epoch": 0.2555161426080683,
      "flos": 23367791923200.0,
      "grad_norm": 3.2687408104922917,
      "language_loss": 0.89428377,
      "learning_rate": 3.489967401911251e-06,
      "loss": 0.91634512,
      "num_input_tokens_seen": 45386805,
      "step": 2125,
      "time_per_iteration": 2.818192958831787
    },
    {
      "auxiliary_loss_clip": 0.01174249,
      "auxiliary_loss_mlp": 0.01042726,
      "balance_loss_clip": 1.06314588,
      "balance_loss_mlp": 1.03208101,
      "epoch": 0.2556363854987074,
      "flos": 40625130723840.0,
      "grad_norm": 2.2990125406831527,
      "language_loss": 0.69367152,
      "learning_rate": 3.4894476504265428e-06,
      "loss": 0.71584129,
      "num_input_tokens_seen": 45411045,
      "step": 2126,
      "time_per_iteration": 2.8624155521392822
    },
    {
      "auxiliary_loss_clip": 0.01097188,
      "auxiliary_loss_mlp": 0.01007651,
      "balance_loss_clip": 1.02202904,
      "balance_loss_mlp": 1.005481,
      "epoch": 0.2557566283893465,
      "flos": 68019443389440.0,
      "grad_norm": 0.7475440133397014,
      "language_loss": 0.54415101,
      "learning_rate": 3.4889276729938104e-06,
      "loss": 0.56519943,
      "num_input_tokens_seen": 45469575,
      "step": 2127,
      "time_per_iteration": 3.1394946575164795
    },
    {
      "auxiliary_loss_clip": 0.01217133,
      "auxiliary_loss_mlp": 0.0103094,
      "balance_loss_clip": 1.06509256,
      "balance_loss_mlp": 1.02109313,
      "epoch": 0.2558768712799856,
      "flos": 22635645004800.0,
      "grad_norm": 2.458687198974695,
      "language_loss": 0.80642211,
      "learning_rate": 3.488407469691934e-06,
      "loss": 0.82890284,
      "num_input_tokens_seen": 45490270,
      "step": 2128,
      "time_per_iteration": 2.6605441570281982
    },
    {
      "auxiliary_loss_clip": 0.0113292,
      "auxiliary_loss_mlp": 0.01040697,
      "balance_loss_clip": 1.05417538,
      "balance_loss_mlp": 1.03082085,
      "epoch": 0.25599711417062465,
      "flos": 26396354125440.0,
      "grad_norm": 7.163394229765161,
      "language_loss": 0.80778205,
      "learning_rate": 3.487887040599828e-06,
      "loss": 0.82951826,
      "num_input_tokens_seen": 45510070,
      "step": 2129,
      "time_per_iteration": 2.8341712951660156
    },
    {
      "auxiliary_loss_clip": 0.01171063,
      "auxiliary_loss_mlp": 0.01044477,
      "balance_loss_clip": 1.06414676,
      "balance_loss_mlp": 1.03436852,
      "epoch": 0.25611735706126376,
      "flos": 22852042490880.0,
      "grad_norm": 3.16232396436872,
      "language_loss": 0.76222301,
      "learning_rate": 3.4873663857964407e-06,
      "loss": 0.78437841,
      "num_input_tokens_seen": 45527285,
      "step": 2130,
      "time_per_iteration": 2.818256139755249
    },
    {
      "auxiliary_loss_clip": 0.0117477,
      "auxiliary_loss_mlp": 0.00763561,
      "balance_loss_clip": 1.05805814,
      "balance_loss_mlp": 1.00073469,
      "epoch": 0.2562375999519028,
      "flos": 23367863750400.0,
      "grad_norm": 3.5274137495853934,
      "language_loss": 0.67042124,
      "learning_rate": 3.4868455053607556e-06,
      "loss": 0.68980455,
      "num_input_tokens_seen": 45546900,
      "step": 2131,
      "time_per_iteration": 2.793513774871826
    },
    {
      "auxiliary_loss_clip": 0.01172598,
      "auxiliary_loss_mlp": 0.00763568,
      "balance_loss_clip": 1.05857956,
      "balance_loss_mlp": 1.000664,
      "epoch": 0.2563578428425419,
      "flos": 22856962654080.0,
      "grad_norm": 2.716824387686832,
      "language_loss": 0.72213006,
      "learning_rate": 3.486324399371789e-06,
      "loss": 0.74149173,
      "num_input_tokens_seen": 45566200,
      "step": 2132,
      "time_per_iteration": 2.8634305000305176
    },
    {
      "auxiliary_loss_clip": 0.01213372,
      "auxiliary_loss_mlp": 0.01040846,
      "balance_loss_clip": 1.06545901,
      "balance_loss_mlp": 1.03153539,
      "epoch": 0.25647808573318104,
      "flos": 21653883498240.0,
      "grad_norm": 1.9979734262565385,
      "language_loss": 0.78668308,
      "learning_rate": 3.485803067908593e-06,
      "loss": 0.80922514,
      "num_input_tokens_seen": 45585710,
      "step": 2133,
      "time_per_iteration": 2.610114574432373
    },
    {
      "auxiliary_loss_clip": 0.01186013,
      "auxiliary_loss_mlp": 0.01039935,
      "balance_loss_clip": 1.06039357,
      "balance_loss_mlp": 1.03074384,
      "epoch": 0.2565983286238201,
      "flos": 33730569659520.0,
      "grad_norm": 4.098883895284645,
      "language_loss": 0.80026472,
      "learning_rate": 3.485281511050253e-06,
      "loss": 0.82252419,
      "num_input_tokens_seen": 45607845,
      "step": 2134,
      "time_per_iteration": 2.852858066558838
    },
    {
      "auxiliary_loss_clip": 0.01159281,
      "auxiliary_loss_mlp": 0.01030717,
      "balance_loss_clip": 1.0555625,
      "balance_loss_mlp": 1.02138305,
      "epoch": 0.2567185715144592,
      "flos": 16216002587520.0,
      "grad_norm": 3.678182917252236,
      "language_loss": 0.90428692,
      "learning_rate": 3.484759728875889e-06,
      "loss": 0.92618692,
      "num_input_tokens_seen": 45623210,
      "step": 2135,
      "time_per_iteration": 2.703345537185669
    },
    {
      "auxiliary_loss_clip": 0.01169369,
      "auxiliary_loss_mlp": 0.01037508,
      "balance_loss_clip": 1.06037259,
      "balance_loss_mlp": 1.02802491,
      "epoch": 0.2568388144050983,
      "flos": 17458475984640.0,
      "grad_norm": 2.339768725343681,
      "language_loss": 0.81232059,
      "learning_rate": 3.4842377214646543e-06,
      "loss": 0.83438933,
      "num_input_tokens_seen": 45641505,
      "step": 2136,
      "time_per_iteration": 2.713679075241089
    },
    {
      "auxiliary_loss_clip": 0.01148746,
      "auxiliary_loss_mlp": 0.01039239,
      "balance_loss_clip": 1.0590241,
      "balance_loss_mlp": 1.02983963,
      "epoch": 0.25695905729573737,
      "flos": 20887442069760.0,
      "grad_norm": 2.027584156402519,
      "language_loss": 0.6655556,
      "learning_rate": 3.483715488895737e-06,
      "loss": 0.68743551,
      "num_input_tokens_seen": 45661835,
      "step": 2137,
      "time_per_iteration": 2.7929911613464355
    },
    {
      "auxiliary_loss_clip": 0.01183669,
      "auxiliary_loss_mlp": 0.01033478,
      "balance_loss_clip": 1.06099737,
      "balance_loss_mlp": 1.02338648,
      "epoch": 0.2570793001863765,
      "flos": 24717278914560.0,
      "grad_norm": 6.804673119039798,
      "language_loss": 0.78865457,
      "learning_rate": 3.48319303124836e-06,
      "loss": 0.81082606,
      "num_input_tokens_seen": 45682215,
      "step": 2138,
      "time_per_iteration": 2.7363929748535156
    },
    {
      "auxiliary_loss_clip": 0.01202378,
      "auxiliary_loss_mlp": 0.01034235,
      "balance_loss_clip": 1.06331635,
      "balance_loss_mlp": 1.02471638,
      "epoch": 0.2571995430770156,
      "flos": 26906896085760.0,
      "grad_norm": 56.607351199071424,
      "language_loss": 0.67165053,
      "learning_rate": 3.4826703486017798e-06,
      "loss": 0.6940167,
      "num_input_tokens_seen": 45701840,
      "step": 2139,
      "time_per_iteration": 2.7523410320281982
    },
    {
      "auxiliary_loss_clip": 0.01195072,
      "auxiliary_loss_mlp": 0.01035049,
      "balance_loss_clip": 1.06372011,
      "balance_loss_mlp": 1.02561998,
      "epoch": 0.25731978596765465,
      "flos": 19792561656960.0,
      "grad_norm": 1.674964242578137,
      "language_loss": 0.76841551,
      "learning_rate": 3.4821474410352867e-06,
      "loss": 0.79071671,
      "num_input_tokens_seen": 45720500,
      "step": 2140,
      "time_per_iteration": 3.5970849990844727
    },
    {
      "auxiliary_loss_clip": 0.01058062,
      "auxiliary_loss_mlp": 0.01001868,
      "balance_loss_clip": 1.0217495,
      "balance_loss_mlp": 0.99978143,
      "epoch": 0.25744002885829376,
      "flos": 70564970471040.0,
      "grad_norm": 0.8965445470866952,
      "language_loss": 0.62680995,
      "learning_rate": 3.481624308628205e-06,
      "loss": 0.6474092,
      "num_input_tokens_seen": 45781870,
      "step": 2141,
      "time_per_iteration": 3.4448163509368896
    },
    {
      "auxiliary_loss_clip": 0.01200538,
      "auxiliary_loss_mlp": 0.01041957,
      "balance_loss_clip": 1.06402588,
      "balance_loss_mlp": 1.03218746,
      "epoch": 0.25756027174893287,
      "flos": 18038181582720.0,
      "grad_norm": 3.9273244505612643,
      "language_loss": 1.0040524,
      "learning_rate": 3.481100951459893e-06,
      "loss": 1.02647734,
      "num_input_tokens_seen": 45794890,
      "step": 2142,
      "time_per_iteration": 3.594195604324341
    },
    {
      "auxiliary_loss_clip": 0.01172107,
      "auxiliary_loss_mlp": 0.01044619,
      "balance_loss_clip": 1.05568194,
      "balance_loss_mlp": 1.03524923,
      "epoch": 0.2576805146395719,
      "flos": 22674069578880.0,
      "grad_norm": 3.424887832581487,
      "language_loss": 0.78405571,
      "learning_rate": 3.4805773696097453e-06,
      "loss": 0.80622298,
      "num_input_tokens_seen": 45815780,
      "step": 2143,
      "time_per_iteration": 2.7654712200164795
    },
    {
      "auxiliary_loss_clip": 0.01181544,
      "auxiliary_loss_mlp": 0.01042852,
      "balance_loss_clip": 1.06326044,
      "balance_loss_mlp": 1.03341663,
      "epoch": 0.25780075753021103,
      "flos": 16472225278080.0,
      "grad_norm": 3.3409515066124835,
      "language_loss": 0.87689865,
      "learning_rate": 3.4800535631571874e-06,
      "loss": 0.89914262,
      "num_input_tokens_seen": 45831310,
      "step": 2144,
      "time_per_iteration": 3.7322723865509033
    },
    {
      "auxiliary_loss_clip": 0.01204077,
      "auxiliary_loss_mlp": 0.01034083,
      "balance_loss_clip": 1.0663892,
      "balance_loss_mlp": 1.02421832,
      "epoch": 0.25792100042085014,
      "flos": 22820297846400.0,
      "grad_norm": 2.425774185290804,
      "language_loss": 0.76948333,
      "learning_rate": 3.4795295321816804e-06,
      "loss": 0.79186499,
      "num_input_tokens_seen": 45850135,
      "step": 2145,
      "time_per_iteration": 2.6789419651031494
    },
    {
      "auxiliary_loss_clip": 0.01200632,
      "auxiliary_loss_mlp": 0.01037723,
      "balance_loss_clip": 1.06579328,
      "balance_loss_mlp": 1.0283469,
      "epoch": 0.2580412433114892,
      "flos": 18697286194560.0,
      "grad_norm": 3.247046315199037,
      "language_loss": 0.91330683,
      "learning_rate": 3.47900527676272e-06,
      "loss": 0.9356904,
      "num_input_tokens_seen": 45868470,
      "step": 2146,
      "time_per_iteration": 2.6773908138275146
    },
    {
      "auxiliary_loss_clip": 0.01186857,
      "auxiliary_loss_mlp": 0.01048108,
      "balance_loss_clip": 1.06744635,
      "balance_loss_mlp": 1.03866708,
      "epoch": 0.2581614862021283,
      "flos": 14283146810880.0,
      "grad_norm": 2.5056137585026694,
      "language_loss": 0.89011979,
      "learning_rate": 3.478480796979835e-06,
      "loss": 0.91246945,
      "num_input_tokens_seen": 45886355,
      "step": 2147,
      "time_per_iteration": 3.704314947128296
    },
    {
      "auxiliary_loss_clip": 0.01123825,
      "auxiliary_loss_mlp": 0.0104291,
      "balance_loss_clip": 1.0525564,
      "balance_loss_mlp": 1.03318858,
      "epoch": 0.25828172909276736,
      "flos": 29498281856640.0,
      "grad_norm": 1.686843394190262,
      "language_loss": 0.78093463,
      "learning_rate": 3.4779560929125894e-06,
      "loss": 0.80260193,
      "num_input_tokens_seen": 45907900,
      "step": 2148,
      "time_per_iteration": 2.8718819618225098
    },
    {
      "auxiliary_loss_clip": 0.01079886,
      "auxiliary_loss_mlp": 0.01000373,
      "balance_loss_clip": 1.02282548,
      "balance_loss_mlp": 0.99835825,
      "epoch": 0.2584019719834065,
      "flos": 67114387376640.0,
      "grad_norm": 0.6640306725739326,
      "language_loss": 0.56851143,
      "learning_rate": 3.4774311646405783e-06,
      "loss": 0.58931404,
      "num_input_tokens_seen": 45977805,
      "step": 2149,
      "time_per_iteration": 3.415915012359619
    },
    {
      "auxiliary_loss_clip": 0.01126692,
      "auxiliary_loss_mlp": 0.01036397,
      "balance_loss_clip": 1.05334175,
      "balance_loss_mlp": 1.02672267,
      "epoch": 0.2585222148740456,
      "flos": 22893555634560.0,
      "grad_norm": 1.9559720191759487,
      "language_loss": 0.83642137,
      "learning_rate": 3.476906012243435e-06,
      "loss": 0.85805225,
      "num_input_tokens_seen": 45996715,
      "step": 2150,
      "time_per_iteration": 2.8389928340911865
    },
    {
      "auxiliary_loss_clip": 0.01202773,
      "auxiliary_loss_mlp": 0.01039857,
      "balance_loss_clip": 1.06632686,
      "balance_loss_mlp": 1.03040981,
      "epoch": 0.25864245776468464,
      "flos": 28909202808960.0,
      "grad_norm": 2.9040796118171452,
      "language_loss": 0.81285304,
      "learning_rate": 3.476380635800824e-06,
      "loss": 0.83527935,
      "num_input_tokens_seen": 46017915,
      "step": 2151,
      "time_per_iteration": 2.7044870853424072
    },
    {
      "auxiliary_loss_clip": 0.01170151,
      "auxiliary_loss_mlp": 0.01033303,
      "balance_loss_clip": 1.05851746,
      "balance_loss_mlp": 1.0233252,
      "epoch": 0.25876270065532375,
      "flos": 14793185980800.0,
      "grad_norm": 3.2011107541394908,
      "language_loss": 0.85599291,
      "learning_rate": 3.475855035392444e-06,
      "loss": 0.8780275,
      "num_input_tokens_seen": 46033235,
      "step": 2152,
      "time_per_iteration": 2.7226014137268066
    },
    {
      "auxiliary_loss_clip": 0.0116784,
      "auxiliary_loss_mlp": 0.0104438,
      "balance_loss_clip": 1.06196654,
      "balance_loss_mlp": 1.03476048,
      "epoch": 0.25888294354596286,
      "flos": 60467821810560.0,
      "grad_norm": 1.7645138006948156,
      "language_loss": 0.71330225,
      "learning_rate": 3.475329211098029e-06,
      "loss": 0.7354244,
      "num_input_tokens_seen": 46056390,
      "step": 2153,
      "time_per_iteration": 3.0694692134857178
    },
    {
      "auxiliary_loss_clip": 0.01176194,
      "auxiliary_loss_mlp": 0.01045696,
      "balance_loss_clip": 1.06154358,
      "balance_loss_mlp": 1.03610003,
      "epoch": 0.2590031864366019,
      "flos": 27851166771840.0,
      "grad_norm": 1.9068643610217673,
      "language_loss": 0.82269263,
      "learning_rate": 3.4748031629973453e-06,
      "loss": 0.84491146,
      "num_input_tokens_seen": 46077120,
      "step": 2154,
      "time_per_iteration": 2.7647876739501953
    },
    {
      "auxiliary_loss_clip": 0.0106918,
      "auxiliary_loss_mlp": 0.01007292,
      "balance_loss_clip": 1.02839422,
      "balance_loss_mlp": 1.00534916,
      "epoch": 0.25912342932724103,
      "flos": 62422444206720.0,
      "grad_norm": 0.9161780746811833,
      "language_loss": 0.56491143,
      "learning_rate": 3.4742768911701944e-06,
      "loss": 0.58567613,
      "num_input_tokens_seen": 46139815,
      "step": 2155,
      "time_per_iteration": 3.3609023094177246
    },
    {
      "auxiliary_loss_clip": 0.0119204,
      "auxiliary_loss_mlp": 0.00764442,
      "balance_loss_clip": 1.06630921,
      "balance_loss_mlp": 1.00061107,
      "epoch": 0.25924367221788014,
      "flos": 12378839368320.0,
      "grad_norm": 4.0002479995439275,
      "language_loss": 0.7079134,
      "learning_rate": 3.4737503956964113e-06,
      "loss": 0.72747827,
      "num_input_tokens_seen": 46152120,
      "step": 2156,
      "time_per_iteration": 2.7124788761138916
    },
    {
      "auxiliary_loss_clip": 0.01153124,
      "auxiliary_loss_mlp": 0.01036683,
      "balance_loss_clip": 1.05952775,
      "balance_loss_mlp": 1.02674079,
      "epoch": 0.2593639151085192,
      "flos": 14575208296320.0,
      "grad_norm": 2.2699515506770354,
      "language_loss": 0.67309761,
      "learning_rate": 3.473223676655865e-06,
      "loss": 0.69499564,
      "num_input_tokens_seen": 46170120,
      "step": 2157,
      "time_per_iteration": 2.8310272693634033
    },
    {
      "auxiliary_loss_clip": 0.01144572,
      "auxiliary_loss_mlp": 0.00764353,
      "balance_loss_clip": 1.0566957,
      "balance_loss_mlp": 1.00064898,
      "epoch": 0.2594841579991583,
      "flos": 15230937029760.0,
      "grad_norm": 1.7338256947121296,
      "language_loss": 0.79753441,
      "learning_rate": 3.472696734128459e-06,
      "loss": 0.81662369,
      "num_input_tokens_seen": 46187985,
      "step": 2158,
      "time_per_iteration": 2.8043453693389893
    },
    {
      "auxiliary_loss_clip": 0.01131319,
      "auxiliary_loss_mlp": 0.01034019,
      "balance_loss_clip": 1.05647731,
      "balance_loss_mlp": 1.02511394,
      "epoch": 0.2596044008897974,
      "flos": 23623583650560.0,
      "grad_norm": 1.8214913939036783,
      "language_loss": 0.76007485,
      "learning_rate": 3.4721695681941286e-06,
      "loss": 0.78172827,
      "num_input_tokens_seen": 46207025,
      "step": 2159,
      "time_per_iteration": 2.904837131500244
    },
    {
      "auxiliary_loss_clip": 0.01176772,
      "auxiliary_loss_mlp": 0.01035877,
      "balance_loss_clip": 1.05852389,
      "balance_loss_mlp": 1.02639401,
      "epoch": 0.25972464378043647,
      "flos": 13772281628160.0,
      "grad_norm": 2.0347172286738173,
      "language_loss": 0.82694077,
      "learning_rate": 3.471642178932845e-06,
      "loss": 0.84906721,
      "num_input_tokens_seen": 46225670,
      "step": 2160,
      "time_per_iteration": 2.70458984375
    },
    {
      "auxiliary_loss_clip": 0.01147703,
      "auxiliary_loss_mlp": 0.01037546,
      "balance_loss_clip": 1.05787456,
      "balance_loss_mlp": 1.02776456,
      "epoch": 0.2598448866710756,
      "flos": 19573578391680.0,
      "grad_norm": 3.8251332560938867,
      "language_loss": 0.90016782,
      "learning_rate": 3.471114566424613e-06,
      "loss": 0.92202038,
      "num_input_tokens_seen": 46244130,
      "step": 2161,
      "time_per_iteration": 2.825284957885742
    },
    {
      "auxiliary_loss_clip": 0.01199626,
      "auxiliary_loss_mlp": 0.01039938,
      "balance_loss_clip": 1.06696391,
      "balance_loss_mlp": 1.02979982,
      "epoch": 0.25996512956171464,
      "flos": 21653237053440.0,
      "grad_norm": 2.1797404049683737,
      "language_loss": 0.75811863,
      "learning_rate": 3.4705867307494715e-06,
      "loss": 0.78051424,
      "num_input_tokens_seen": 46263200,
      "step": 2162,
      "time_per_iteration": 2.722607374191284
    },
    {
      "auxiliary_loss_clip": 0.01187219,
      "auxiliary_loss_mlp": 0.01041922,
      "balance_loss_clip": 1.06673157,
      "balance_loss_mlp": 1.03242111,
      "epoch": 0.26008537245235375,
      "flos": 18223480869120.0,
      "grad_norm": 2.245696660740515,
      "language_loss": 0.84811717,
      "learning_rate": 3.470058671987492e-06,
      "loss": 0.87040854,
      "num_input_tokens_seen": 46281465,
      "step": 2163,
      "time_per_iteration": 2.868891477584839
    },
    {
      "auxiliary_loss_clip": 0.01192929,
      "auxiliary_loss_mlp": 0.01044627,
      "balance_loss_clip": 1.06324863,
      "balance_loss_mlp": 1.03389239,
      "epoch": 0.26020561534299286,
      "flos": 24645385843200.0,
      "grad_norm": 4.675363603370533,
      "language_loss": 0.84383082,
      "learning_rate": 3.4695303902187805e-06,
      "loss": 0.86620641,
      "num_input_tokens_seen": 46301020,
      "step": 2164,
      "time_per_iteration": 2.768124580383301
    },
    {
      "auxiliary_loss_clip": 0.01218771,
      "auxiliary_loss_mlp": 0.01035695,
      "balance_loss_clip": 1.06869388,
      "balance_loss_mlp": 1.02584231,
      "epoch": 0.2603258582336319,
      "flos": 25773662926080.0,
      "grad_norm": 2.7010069187594565,
      "language_loss": 0.78872812,
      "learning_rate": 3.469001885523478e-06,
      "loss": 0.8112728,
      "num_input_tokens_seen": 46321740,
      "step": 2165,
      "time_per_iteration": 2.6653692722320557
    },
    {
      "auxiliary_loss_clip": 0.01112609,
      "auxiliary_loss_mlp": 0.01050802,
      "balance_loss_clip": 1.04984879,
      "balance_loss_mlp": 1.04078269,
      "epoch": 0.260446101124271,
      "flos": 28766314506240.0,
      "grad_norm": 1.8996136922456368,
      "language_loss": 0.81233126,
      "learning_rate": 3.4684731579817568e-06,
      "loss": 0.8339653,
      "num_input_tokens_seen": 46342730,
      "step": 2166,
      "time_per_iteration": 3.895745277404785
    },
    {
      "auxiliary_loss_clip": 0.01189807,
      "auxiliary_loss_mlp": 0.01039751,
      "balance_loss_clip": 1.06502986,
      "balance_loss_mlp": 1.03024364,
      "epoch": 0.26056634401491013,
      "flos": 25666757072640.0,
      "grad_norm": 2.3894118601707084,
      "language_loss": 0.77061391,
      "learning_rate": 3.4679442076738247e-06,
      "loss": 0.7929095,
      "num_input_tokens_seen": 46362445,
      "step": 2167,
      "time_per_iteration": 3.775010585784912
    },
    {
      "auxiliary_loss_clip": 0.01202225,
      "auxiliary_loss_mlp": 0.0104007,
      "balance_loss_clip": 1.06567121,
      "balance_loss_mlp": 1.0293057,
      "epoch": 0.2606865869055492,
      "flos": 27052765217280.0,
      "grad_norm": 3.695960255188547,
      "language_loss": 0.83913285,
      "learning_rate": 3.4674150346799245e-06,
      "loss": 0.86155581,
      "num_input_tokens_seen": 46382145,
      "step": 2168,
      "time_per_iteration": 2.7116572856903076
    },
    {
      "auxiliary_loss_clip": 0.01176751,
      "auxiliary_loss_mlp": 0.01035838,
      "balance_loss_clip": 1.06069207,
      "balance_loss_mlp": 1.02488244,
      "epoch": 0.2608068297961883,
      "flos": 17712615686400.0,
      "grad_norm": 13.906444538924841,
      "language_loss": 0.80647457,
      "learning_rate": 3.4668856390803295e-06,
      "loss": 0.82860041,
      "num_input_tokens_seen": 46400025,
      "step": 2169,
      "time_per_iteration": 2.778482675552368
    },
    {
      "auxiliary_loss_clip": 0.01181229,
      "auxiliary_loss_mlp": 0.0103453,
      "balance_loss_clip": 1.06283963,
      "balance_loss_mlp": 1.02463531,
      "epoch": 0.2609270726868274,
      "flos": 18551632544640.0,
      "grad_norm": 2.5261748592347892,
      "language_loss": 0.89834481,
      "learning_rate": 3.4663560209553495e-06,
      "loss": 0.92050242,
      "num_input_tokens_seen": 46418090,
      "step": 2170,
      "time_per_iteration": 3.658853530883789
    },
    {
      "auxiliary_loss_clip": 0.01200586,
      "auxiliary_loss_mlp": 0.01044234,
      "balance_loss_clip": 1.06749201,
      "balance_loss_mlp": 1.03500104,
      "epoch": 0.26104731557746647,
      "flos": 21835699165440.0,
      "grad_norm": 2.343870782464588,
      "language_loss": 0.79488665,
      "learning_rate": 3.4658261803853267e-06,
      "loss": 0.81733483,
      "num_input_tokens_seen": 46436015,
      "step": 2171,
      "time_per_iteration": 2.6214873790740967
    },
    {
      "auxiliary_loss_clip": 0.01188382,
      "auxiliary_loss_mlp": 0.01031054,
      "balance_loss_clip": 1.06374681,
      "balance_loss_mlp": 1.02101684,
      "epoch": 0.2611675584681056,
      "flos": 21689650465920.0,
      "grad_norm": 3.961469108041387,
      "language_loss": 0.80852824,
      "learning_rate": 3.4652961174506383e-06,
      "loss": 0.83072257,
      "num_input_tokens_seen": 46455885,
      "step": 2172,
      "time_per_iteration": 2.7630062103271484
    },
    {
      "auxiliary_loss_clip": 0.01067468,
      "auxiliary_loss_mlp": 0.01019941,
      "balance_loss_clip": 1.01952624,
      "balance_loss_mlp": 1.01781893,
      "epoch": 0.2612878013587447,
      "flos": 71862101389440.0,
      "grad_norm": 0.9654582625253781,
      "language_loss": 0.58020103,
      "learning_rate": 3.464765832231694e-06,
      "loss": 0.60107511,
      "num_input_tokens_seen": 46510050,
      "step": 2173,
      "time_per_iteration": 4.3216118812561035
    },
    {
      "auxiliary_loss_clip": 0.01196987,
      "auxiliary_loss_mlp": 0.010416,
      "balance_loss_clip": 1.06528866,
      "balance_loss_mlp": 1.03166425,
      "epoch": 0.26140804424938374,
      "flos": 20227511445120.0,
      "grad_norm": 43.12523696691856,
      "language_loss": 0.71064425,
      "learning_rate": 3.4642353248089373e-06,
      "loss": 0.7330302,
      "num_input_tokens_seen": 46528810,
      "step": 2174,
      "time_per_iteration": 2.6950583457946777
    },
    {
      "auxiliary_loss_clip": 0.01218194,
      "auxiliary_loss_mlp": 0.01041183,
      "balance_loss_clip": 1.06625926,
      "balance_loss_mlp": 1.0304122,
      "epoch": 0.26152828714002285,
      "flos": 25557085872000.0,
      "grad_norm": 2.173838349000386,
      "language_loss": 0.8061887,
      "learning_rate": 3.463704595262846e-06,
      "loss": 0.82878244,
      "num_input_tokens_seen": 46549690,
      "step": 2175,
      "time_per_iteration": 2.712907075881958
    },
    {
      "auxiliary_loss_clip": 0.01203119,
      "auxiliary_loss_mlp": 0.01036923,
      "balance_loss_clip": 1.06685078,
      "balance_loss_mlp": 1.02795863,
      "epoch": 0.26164853003066196,
      "flos": 25446516831360.0,
      "grad_norm": 2.219014339618736,
      "language_loss": 0.70889616,
      "learning_rate": 3.463173643673931e-06,
      "loss": 0.7312966,
      "num_input_tokens_seen": 46572215,
      "step": 2176,
      "time_per_iteration": 2.839968204498291
    },
    {
      "auxiliary_loss_clip": 0.01080064,
      "auxiliary_loss_mlp": 0.01004521,
      "balance_loss_clip": 1.02085519,
      "balance_loss_mlp": 1.00263715,
      "epoch": 0.261768772921301,
      "flos": 53944580568960.0,
      "grad_norm": 0.9016035220484413,
      "language_loss": 0.6350981,
      "learning_rate": 3.4626424701227387e-06,
      "loss": 0.65594399,
      "num_input_tokens_seen": 46627275,
      "step": 2177,
      "time_per_iteration": 3.2302322387695312
    },
    {
      "auxiliary_loss_clip": 0.01094329,
      "auxiliary_loss_mlp": 0.01005472,
      "balance_loss_clip": 1.01979494,
      "balance_loss_mlp": 1.00366044,
      "epoch": 0.26188901581194013,
      "flos": 70687606481280.0,
      "grad_norm": 0.8324192681932012,
      "language_loss": 0.55744493,
      "learning_rate": 3.4621110746898452e-06,
      "loss": 0.57844293,
      "num_input_tokens_seen": 46695135,
      "step": 2178,
      "time_per_iteration": 3.3529155254364014
    },
    {
      "auxiliary_loss_clip": 0.01200718,
      "auxiliary_loss_mlp": 0.01035168,
      "balance_loss_clip": 1.06664872,
      "balance_loss_mlp": 1.02598262,
      "epoch": 0.2620092587025792,
      "flos": 21069580959360.0,
      "grad_norm": 2.706342292886628,
      "language_loss": 0.74763322,
      "learning_rate": 3.4615794574558654e-06,
      "loss": 0.76999211,
      "num_input_tokens_seen": 46714145,
      "step": 2179,
      "time_per_iteration": 2.662538528442383
    },
    {
      "auxiliary_loss_clip": 0.01191826,
      "auxiliary_loss_mlp": 0.01043979,
      "balance_loss_clip": 1.06183255,
      "balance_loss_mlp": 1.03363168,
      "epoch": 0.2621295015932183,
      "flos": 18369601395840.0,
      "grad_norm": 2.597703473260973,
      "language_loss": 0.84212738,
      "learning_rate": 3.4610476185014436e-06,
      "loss": 0.86448538,
      "num_input_tokens_seen": 46731405,
      "step": 2180,
      "time_per_iteration": 2.730377197265625
    },
    {
      "auxiliary_loss_clip": 0.01190869,
      "auxiliary_loss_mlp": 0.00764339,
      "balance_loss_clip": 1.06412649,
      "balance_loss_mlp": 1.00074792,
      "epoch": 0.2622497444838574,
      "flos": 23659997063040.0,
      "grad_norm": 2.286619188509542,
      "language_loss": 0.79674208,
      "learning_rate": 3.4605155579072597e-06,
      "loss": 0.81629419,
      "num_input_tokens_seen": 46751260,
      "step": 2181,
      "time_per_iteration": 2.770840883255005
    },
    {
      "auxiliary_loss_clip": 0.01191931,
      "auxiliary_loss_mlp": 0.01037128,
      "balance_loss_clip": 1.06279492,
      "balance_loss_mlp": 1.02735901,
      "epoch": 0.26236998737449646,
      "flos": 22123810154880.0,
      "grad_norm": 3.2986077469198216,
      "language_loss": 0.71279311,
      "learning_rate": 3.459983275754027e-06,
      "loss": 0.7350837,
      "num_input_tokens_seen": 46770155,
      "step": 2182,
      "time_per_iteration": 2.73930025100708
    },
    {
      "auxiliary_loss_clip": 0.01191062,
      "auxiliary_loss_mlp": 0.01035062,
      "balance_loss_clip": 1.06178904,
      "balance_loss_mlp": 1.02536488,
      "epoch": 0.26249023026513557,
      "flos": 17895185539200.0,
      "grad_norm": 2.8543374859566173,
      "language_loss": 0.79627949,
      "learning_rate": 3.4594507721224918e-06,
      "loss": 0.81854075,
      "num_input_tokens_seen": 46788805,
      "step": 2183,
      "time_per_iteration": 2.6863415241241455
    },
    {
      "auxiliary_loss_clip": 0.01161874,
      "auxiliary_loss_mlp": 0.01040047,
      "balance_loss_clip": 1.0606432,
      "balance_loss_mlp": 1.02973557,
      "epoch": 0.2626104731557747,
      "flos": 18332936588160.0,
      "grad_norm": 1.9074120930351544,
      "language_loss": 0.8207376,
      "learning_rate": 3.4589180470934353e-06,
      "loss": 0.84275687,
      "num_input_tokens_seen": 46808670,
      "step": 2184,
      "time_per_iteration": 2.840564012527466
    },
    {
      "auxiliary_loss_clip": 0.01162985,
      "auxiliary_loss_mlp": 0.01038195,
      "balance_loss_clip": 1.05521321,
      "balance_loss_mlp": 1.02755523,
      "epoch": 0.26273071604641374,
      "flos": 19317714837120.0,
      "grad_norm": 2.202783062420899,
      "language_loss": 0.76567757,
      "learning_rate": 3.4583851007476713e-06,
      "loss": 0.78768939,
      "num_input_tokens_seen": 46827140,
      "step": 2185,
      "time_per_iteration": 2.7348780632019043
    },
    {
      "auxiliary_loss_clip": 0.01171517,
      "auxiliary_loss_mlp": 0.01042421,
      "balance_loss_clip": 1.05971599,
      "balance_loss_mlp": 1.03187704,
      "epoch": 0.26285095893705285,
      "flos": 18327477720960.0,
      "grad_norm": 3.4795602940841226,
      "language_loss": 0.68344814,
      "learning_rate": 3.4578519331660464e-06,
      "loss": 0.70558751,
      "num_input_tokens_seen": 46844135,
      "step": 2186,
      "time_per_iteration": 2.7538299560546875
    },
    {
      "auxiliary_loss_clip": 0.01195681,
      "auxiliary_loss_mlp": 0.00763745,
      "balance_loss_clip": 1.06442678,
      "balance_loss_mlp": 1.00081897,
      "epoch": 0.26297120182769196,
      "flos": 20193827466240.0,
      "grad_norm": 1.9272759736478744,
      "language_loss": 0.82690418,
      "learning_rate": 3.4573185444294426e-06,
      "loss": 0.84649837,
      "num_input_tokens_seen": 46862500,
      "step": 2187,
      "time_per_iteration": 2.769282817840576
    },
    {
      "auxiliary_loss_clip": 0.01169023,
      "auxiliary_loss_mlp": 0.01040688,
      "balance_loss_clip": 1.05914783,
      "balance_loss_mlp": 1.03089523,
      "epoch": 0.263091444718331,
      "flos": 22418421505920.0,
      "grad_norm": 1.6358159326619797,
      "language_loss": 0.78720468,
      "learning_rate": 3.456784934618774e-06,
      "loss": 0.80930179,
      "num_input_tokens_seen": 46883665,
      "step": 2188,
      "time_per_iteration": 2.765294313430786
    },
    {
      "auxiliary_loss_clip": 0.01172001,
      "auxiliary_loss_mlp": 0.00763812,
      "balance_loss_clip": 1.05858636,
      "balance_loss_mlp": 1.00079179,
      "epoch": 0.2632116876089701,
      "flos": 19024827338880.0,
      "grad_norm": 2.4058883133206908,
      "language_loss": 0.79722506,
      "learning_rate": 3.4562511038149897e-06,
      "loss": 0.81658322,
      "num_input_tokens_seen": 46899160,
      "step": 2189,
      "time_per_iteration": 2.970916271209717
    },
    {
      "auxiliary_loss_clip": 0.01096041,
      "auxiliary_loss_mlp": 0.01000978,
      "balance_loss_clip": 1.02209008,
      "balance_loss_mlp": 0.99907082,
      "epoch": 0.26333193049960923,
      "flos": 67308054531840.0,
      "grad_norm": 0.8589782288790538,
      "language_loss": 0.57718331,
      "learning_rate": 3.4557170520990705e-06,
      "loss": 0.59815347,
      "num_input_tokens_seen": 46959835,
      "step": 2190,
      "time_per_iteration": 3.322808027267456
    },
    {
      "auxiliary_loss_clip": 0.01185699,
      "auxiliary_loss_mlp": 0.00764295,
      "balance_loss_clip": 1.06350684,
      "balance_loss_mlp": 1.00080597,
      "epoch": 0.2634521733902483,
      "flos": 25048806468480.0,
      "grad_norm": 1.670017147513658,
      "language_loss": 0.86749303,
      "learning_rate": 3.4551827795520324e-06,
      "loss": 0.88699299,
      "num_input_tokens_seen": 46982720,
      "step": 2191,
      "time_per_iteration": 2.8913843631744385
    },
    {
      "auxiliary_loss_clip": 0.01200977,
      "auxiliary_loss_mlp": 0.01035148,
      "balance_loss_clip": 1.06255007,
      "balance_loss_mlp": 1.02567124,
      "epoch": 0.2635724162808874,
      "flos": 20594985534720.0,
      "grad_norm": 1.835596192037433,
      "language_loss": 0.84778428,
      "learning_rate": 3.4546482862549226e-06,
      "loss": 0.87014556,
      "num_input_tokens_seen": 47003035,
      "step": 2192,
      "time_per_iteration": 2.6645355224609375
    },
    {
      "auxiliary_loss_clip": 0.01200894,
      "auxiliary_loss_mlp": 0.0104463,
      "balance_loss_clip": 1.06543088,
      "balance_loss_mlp": 1.03441978,
      "epoch": 0.2636926591715265,
      "flos": 19244636616960.0,
      "grad_norm": 2.223015172403824,
      "language_loss": 0.78600776,
      "learning_rate": 3.4541135722888253e-06,
      "loss": 0.80846298,
      "num_input_tokens_seen": 47019625,
      "step": 2193,
      "time_per_iteration": 3.604745864868164
    },
    {
      "auxiliary_loss_clip": 0.01159397,
      "auxiliary_loss_mlp": 0.01030776,
      "balance_loss_clip": 1.05852985,
      "balance_loss_mlp": 1.02102518,
      "epoch": 0.26381290206216557,
      "flos": 28804882734720.0,
      "grad_norm": 2.2770709951375143,
      "language_loss": 0.80419785,
      "learning_rate": 3.453578637734854e-06,
      "loss": 0.82609963,
      "num_input_tokens_seen": 47040815,
      "step": 2194,
      "time_per_iteration": 2.8184139728546143
    },
    {
      "auxiliary_loss_clip": 0.01205142,
      "auxiliary_loss_mlp": 0.01039649,
      "balance_loss_clip": 1.06827998,
      "balance_loss_mlp": 1.03016615,
      "epoch": 0.2639331449528047,
      "flos": 25008909436800.0,
      "grad_norm": 1.9558531382448596,
      "language_loss": 0.78653693,
      "learning_rate": 3.4530434826741605e-06,
      "loss": 0.80898488,
      "num_input_tokens_seen": 47061755,
      "step": 2195,
      "time_per_iteration": 2.6884541511535645
    },
    {
      "auxiliary_loss_clip": 0.01127301,
      "auxiliary_loss_mlp": 0.01036304,
      "balance_loss_clip": 1.04892039,
      "balance_loss_mlp": 1.02699947,
      "epoch": 0.26405338784344373,
      "flos": 46535775465600.0,
      "grad_norm": 1.7929170530801086,
      "language_loss": 0.69450587,
      "learning_rate": 3.452508107187926e-06,
      "loss": 0.71614188,
      "num_input_tokens_seen": 47085130,
      "step": 2196,
      "time_per_iteration": 3.9307446479797363
    },
    {
      "auxiliary_loss_clip": 0.01144951,
      "auxiliary_loss_mlp": 0.0103859,
      "balance_loss_clip": 1.05637908,
      "balance_loss_mlp": 1.02776039,
      "epoch": 0.26417363073408284,
      "flos": 21179467641600.0,
      "grad_norm": 6.25218361068542,
      "language_loss": 0.77201307,
      "learning_rate": 3.451972511357366e-06,
      "loss": 0.7938484,
      "num_input_tokens_seen": 47104675,
      "step": 2197,
      "time_per_iteration": 2.79244065284729
    },
    {
      "auxiliary_loss_clip": 0.01170793,
      "auxiliary_loss_mlp": 0.01034313,
      "balance_loss_clip": 1.05646253,
      "balance_loss_mlp": 1.02445996,
      "epoch": 0.26429387362472195,
      "flos": 22674751937280.0,
      "grad_norm": 1.8235244015242995,
      "language_loss": 0.85214913,
      "learning_rate": 3.45143669526373e-06,
      "loss": 0.87420017,
      "num_input_tokens_seen": 47124435,
      "step": 2198,
      "time_per_iteration": 3.7250139713287354
    },
    {
      "auxiliary_loss_clip": 0.01086512,
      "auxiliary_loss_mlp": 0.00999869,
      "balance_loss_clip": 1.01879311,
      "balance_loss_mlp": 0.99812907,
      "epoch": 0.264414116515361,
      "flos": 67180534272000.0,
      "grad_norm": 0.8043091225669439,
      "language_loss": 0.63192862,
      "learning_rate": 3.450900658988302e-06,
      "loss": 0.65279245,
      "num_input_tokens_seen": 47185985,
      "step": 2199,
      "time_per_iteration": 3.2070014476776123
    },
    {
      "auxiliary_loss_clip": 0.01202427,
      "auxiliary_loss_mlp": 0.01036637,
      "balance_loss_clip": 1.06649005,
      "balance_loss_mlp": 1.02643251,
      "epoch": 0.2645343594060001,
      "flos": 25664709997440.0,
      "grad_norm": 3.34436861694104,
      "language_loss": 0.77680981,
      "learning_rate": 3.450364402612397e-06,
      "loss": 0.79920048,
      "num_input_tokens_seen": 47203140,
      "step": 2200,
      "time_per_iteration": 2.6984376907348633
    },
    {
      "auxiliary_loss_clip": 0.0117614,
      "auxiliary_loss_mlp": 0.01037143,
      "balance_loss_clip": 1.05910051,
      "balance_loss_mlp": 1.02706981,
      "epoch": 0.26465460229663923,
      "flos": 22491822948480.0,
      "grad_norm": 2.2349890968269865,
      "language_loss": 0.83888716,
      "learning_rate": 3.449827926217366e-06,
      "loss": 0.86101997,
      "num_input_tokens_seen": 47222575,
      "step": 2201,
      "time_per_iteration": 2.8197500705718994
    },
    {
      "auxiliary_loss_clip": 0.01138661,
      "auxiliary_loss_mlp": 0.01035633,
      "balance_loss_clip": 1.05272722,
      "balance_loss_mlp": 1.02587557,
      "epoch": 0.2647748451872783,
      "flos": 29388036038400.0,
      "grad_norm": 3.265718964876891,
      "language_loss": 0.80870306,
      "learning_rate": 3.449291229884591e-06,
      "loss": 0.830446,
      "num_input_tokens_seen": 47243815,
      "step": 2202,
      "time_per_iteration": 2.8674840927124023
    },
    {
      "auxiliary_loss_clip": 0.01154072,
      "auxiliary_loss_mlp": 0.01032452,
      "balance_loss_clip": 1.05624461,
      "balance_loss_mlp": 1.02284408,
      "epoch": 0.2648950880779174,
      "flos": 26797799502720.0,
      "grad_norm": 2.1508152755969956,
      "language_loss": 0.86851251,
      "learning_rate": 3.4487543136954887e-06,
      "loss": 0.89037776,
      "num_input_tokens_seen": 47263435,
      "step": 2203,
      "time_per_iteration": 2.8138482570648193
    },
    {
      "auxiliary_loss_clip": 0.0115827,
      "auxiliary_loss_mlp": 0.01036842,
      "balance_loss_clip": 1.05705619,
      "balance_loss_mlp": 1.02696538,
      "epoch": 0.2650153309685565,
      "flos": 28841008838400.0,
      "grad_norm": 1.7218062580010962,
      "language_loss": 0.91605872,
      "learning_rate": 3.448217177731509e-06,
      "loss": 0.93800986,
      "num_input_tokens_seen": 47283920,
      "step": 2204,
      "time_per_iteration": 2.870962619781494
    },
    {
      "auxiliary_loss_clip": 0.01158058,
      "auxiliary_loss_mlp": 0.01042362,
      "balance_loss_clip": 1.06033111,
      "balance_loss_mlp": 1.03246212,
      "epoch": 0.26513557385919556,
      "flos": 20303247271680.0,
      "grad_norm": 2.7079531756175257,
      "language_loss": 0.78206646,
      "learning_rate": 3.4476798220741348e-06,
      "loss": 0.80407059,
      "num_input_tokens_seen": 47302800,
      "step": 2205,
      "time_per_iteration": 2.7819812297821045
    },
    {
      "auxiliary_loss_clip": 0.01203215,
      "auxiliary_loss_mlp": 0.01042242,
      "balance_loss_clip": 1.06571531,
      "balance_loss_mlp": 1.03270555,
      "epoch": 0.26525581674983467,
      "flos": 17676274101120.0,
      "grad_norm": 1.9118049242495514,
      "language_loss": 0.7867189,
      "learning_rate": 3.4471422468048826e-06,
      "loss": 0.80917346,
      "num_input_tokens_seen": 47321525,
      "step": 2206,
      "time_per_iteration": 2.7203638553619385
    },
    {
      "auxiliary_loss_clip": 0.01192734,
      "auxiliary_loss_mlp": 0.01043218,
      "balance_loss_clip": 1.06772447,
      "balance_loss_mlp": 1.03314483,
      "epoch": 0.2653760596404738,
      "flos": 26833746038400.0,
      "grad_norm": 4.815748327440133,
      "language_loss": 0.73365676,
      "learning_rate": 3.4466044520053022e-06,
      "loss": 0.75601637,
      "num_input_tokens_seen": 47340530,
      "step": 2207,
      "time_per_iteration": 2.711885690689087
    },
    {
      "auxiliary_loss_clip": 0.01197708,
      "auxiliary_loss_mlp": 0.00763836,
      "balance_loss_clip": 1.06426096,
      "balance_loss_mlp": 1.00061202,
      "epoch": 0.26549630253111284,
      "flos": 22782160581120.0,
      "grad_norm": 2.020115995638419,
      "language_loss": 0.6029557,
      "learning_rate": 3.446066437756977e-06,
      "loss": 0.62257111,
      "num_input_tokens_seen": 47359735,
      "step": 2208,
      "time_per_iteration": 2.7348577976226807
    },
    {
      "auxiliary_loss_clip": 0.01162943,
      "auxiliary_loss_mlp": 0.01035363,
      "balance_loss_clip": 1.05226231,
      "balance_loss_mlp": 1.02537966,
      "epoch": 0.26561654542175195,
      "flos": 23550002640000.0,
      "grad_norm": 2.402225554004261,
      "language_loss": 0.75433898,
      "learning_rate": 3.4455282041415224e-06,
      "loss": 0.77632201,
      "num_input_tokens_seen": 47378945,
      "step": 2209,
      "time_per_iteration": 2.816978693008423
    },
    {
      "auxiliary_loss_clip": 0.01177323,
      "auxiliary_loss_mlp": 0.01033133,
      "balance_loss_clip": 1.05668902,
      "balance_loss_mlp": 1.0235424,
      "epoch": 0.265736788312391,
      "flos": 26906680604160.0,
      "grad_norm": 2.715441050585948,
      "language_loss": 0.87370181,
      "learning_rate": 3.4449897512405894e-06,
      "loss": 0.89580637,
      "num_input_tokens_seen": 47398095,
      "step": 2210,
      "time_per_iteration": 2.745206356048584
    },
    {
      "auxiliary_loss_clip": 0.01198079,
      "auxiliary_loss_mlp": 0.01037299,
      "balance_loss_clip": 1.06170034,
      "balance_loss_mlp": 1.0273931,
      "epoch": 0.2658570312030301,
      "flos": 23477139901440.0,
      "grad_norm": 2.0971332561222833,
      "language_loss": 0.75904226,
      "learning_rate": 3.444451079135859e-06,
      "loss": 0.78139603,
      "num_input_tokens_seen": 47417605,
      "step": 2211,
      "time_per_iteration": 2.7445194721221924
    },
    {
      "auxiliary_loss_clip": 0.01111272,
      "auxiliary_loss_mlp": 0.01038852,
      "balance_loss_clip": 1.04919791,
      "balance_loss_mlp": 1.02777183,
      "epoch": 0.2659772740936692,
      "flos": 21866402315520.0,
      "grad_norm": 2.6163216999789864,
      "language_loss": 0.73966187,
      "learning_rate": 3.4439121879090493e-06,
      "loss": 0.76116312,
      "num_input_tokens_seen": 47435385,
      "step": 2212,
      "time_per_iteration": 2.7784578800201416
    },
    {
      "auxiliary_loss_clip": 0.0120278,
      "auxiliary_loss_mlp": 0.01035802,
      "balance_loss_clip": 1.06509507,
      "balance_loss_mlp": 1.02510333,
      "epoch": 0.2660975169843083,
      "flos": 19793100360960.0,
      "grad_norm": 2.749726879678324,
      "language_loss": 0.83360302,
      "learning_rate": 3.4433730776419082e-06,
      "loss": 0.85598886,
      "num_input_tokens_seen": 47454310,
      "step": 2213,
      "time_per_iteration": 2.767334222793579
    },
    {
      "auxiliary_loss_clip": 0.01204985,
      "auxiliary_loss_mlp": 0.01035285,
      "balance_loss_clip": 1.06478262,
      "balance_loss_mlp": 1.02471066,
      "epoch": 0.2662177598749474,
      "flos": 29018981750400.0,
      "grad_norm": 6.94926053846536,
      "language_loss": 0.80571628,
      "learning_rate": 3.4428337484162183e-06,
      "loss": 0.82811898,
      "num_input_tokens_seen": 47475120,
      "step": 2214,
      "time_per_iteration": 2.697922468185425
    },
    {
      "auxiliary_loss_clip": 0.01134959,
      "auxiliary_loss_mlp": 0.01045602,
      "balance_loss_clip": 1.05379891,
      "balance_loss_mlp": 1.03574967,
      "epoch": 0.2663380027655865,
      "flos": 21762549118080.0,
      "grad_norm": 1.918097153284572,
      "language_loss": 0.84454727,
      "learning_rate": 3.442294200313797e-06,
      "loss": 0.86635292,
      "num_input_tokens_seen": 47493150,
      "step": 2215,
      "time_per_iteration": 2.7796437740325928
    },
    {
      "auxiliary_loss_clip": 0.01076822,
      "auxiliary_loss_mlp": 0.01000973,
      "balance_loss_clip": 1.01917219,
      "balance_loss_mlp": 0.99916089,
      "epoch": 0.26645824565622556,
      "flos": 66980333819520.0,
      "grad_norm": 0.8018452230714114,
      "language_loss": 0.52702361,
      "learning_rate": 3.4417544334164916e-06,
      "loss": 0.54780155,
      "num_input_tokens_seen": 47557295,
      "step": 2216,
      "time_per_iteration": 3.299875259399414
    },
    {
      "auxiliary_loss_clip": 0.01150974,
      "auxiliary_loss_mlp": 0.01037258,
      "balance_loss_clip": 1.05858016,
      "balance_loss_mlp": 1.0280962,
      "epoch": 0.26657848854686467,
      "flos": 25264198373760.0,
      "grad_norm": 2.4316450380531385,
      "language_loss": 0.77581066,
      "learning_rate": 3.4412144478061854e-06,
      "loss": 0.79769295,
      "num_input_tokens_seen": 47579705,
      "step": 2217,
      "time_per_iteration": 2.8266446590423584
    },
    {
      "auxiliary_loss_clip": 0.01189176,
      "auxiliary_loss_mlp": 0.01041132,
      "balance_loss_clip": 1.06490755,
      "balance_loss_mlp": 1.03085017,
      "epoch": 0.2666987314375038,
      "flos": 23696769611520.0,
      "grad_norm": 2.6225027124140565,
      "language_loss": 0.75748134,
      "learning_rate": 3.4406742435647925e-06,
      "loss": 0.77978444,
      "num_input_tokens_seen": 47599770,
      "step": 2218,
      "time_per_iteration": 3.628828763961792
    },
    {
      "auxiliary_loss_clip": 0.01183071,
      "auxiliary_loss_mlp": 0.01035941,
      "balance_loss_clip": 1.06428051,
      "balance_loss_mlp": 1.02546215,
      "epoch": 0.26681897432814283,
      "flos": 27048958375680.0,
      "grad_norm": 3.5213041444952076,
      "language_loss": 0.79303056,
      "learning_rate": 3.440133820774263e-06,
      "loss": 0.81522065,
      "num_input_tokens_seen": 47619580,
      "step": 2219,
      "time_per_iteration": 3.817589044570923
    },
    {
      "auxiliary_loss_clip": 0.01200248,
      "auxiliary_loss_mlp": 0.01042191,
      "balance_loss_clip": 1.06532431,
      "balance_loss_mlp": 1.03195143,
      "epoch": 0.26693921721878194,
      "flos": 28985944216320.0,
      "grad_norm": 11.472891942806312,
      "language_loss": 0.82275027,
      "learning_rate": 3.439593179516578e-06,
      "loss": 0.84517467,
      "num_input_tokens_seen": 47639490,
      "step": 2220,
      "time_per_iteration": 2.7588114738464355
    },
    {
      "auxiliary_loss_clip": 0.01171728,
      "auxiliary_loss_mlp": 0.00763882,
      "balance_loss_clip": 1.06302583,
      "balance_loss_mlp": 1.00061989,
      "epoch": 0.26705946010942105,
      "flos": 21507834798720.0,
      "grad_norm": 2.505129374647731,
      "language_loss": 0.81501162,
      "learning_rate": 3.4390523198737524e-06,
      "loss": 0.83436775,
      "num_input_tokens_seen": 47658650,
      "step": 2221,
      "time_per_iteration": 3.7605557441711426
    },
    {
      "auxiliary_loss_clip": 0.01176105,
      "auxiliary_loss_mlp": 0.01038294,
      "balance_loss_clip": 1.06052566,
      "balance_loss_mlp": 1.02797067,
      "epoch": 0.2671797030000601,
      "flos": 21471277731840.0,
      "grad_norm": 1.9855614608378378,
      "language_loss": 0.73600721,
      "learning_rate": 3.4385112419278333e-06,
      "loss": 0.75815117,
      "num_input_tokens_seen": 47679875,
      "step": 2222,
      "time_per_iteration": 2.7093613147735596
    },
    {
      "auxiliary_loss_clip": 0.01042599,
      "auxiliary_loss_mlp": 0.00754712,
      "balance_loss_clip": 1.01858866,
      "balance_loss_mlp": 1.00017452,
      "epoch": 0.2672999458906992,
      "flos": 64189929767040.0,
      "grad_norm": 0.7898968252035977,
      "language_loss": 0.64736116,
      "learning_rate": 3.4379699457609033e-06,
      "loss": 0.66533422,
      "num_input_tokens_seen": 47737700,
      "step": 2223,
      "time_per_iteration": 3.268874168395996
    },
    {
      "auxiliary_loss_clip": 0.01135328,
      "auxiliary_loss_mlp": 0.01042273,
      "balance_loss_clip": 1.04828489,
      "balance_loss_mlp": 1.03115082,
      "epoch": 0.26742018878133833,
      "flos": 16909042573440.0,
      "grad_norm": 5.012341585317213,
      "language_loss": 0.90469086,
      "learning_rate": 3.4374284314550755e-06,
      "loss": 0.92646682,
      "num_input_tokens_seen": 47756740,
      "step": 2224,
      "time_per_iteration": 4.5016257762908936
    },
    {
      "auxiliary_loss_clip": 0.01181085,
      "auxiliary_loss_mlp": 0.01043402,
      "balance_loss_clip": 1.05950427,
      "balance_loss_mlp": 1.03276241,
      "epoch": 0.2675404316719774,
      "flos": 20667560964480.0,
      "grad_norm": 2.6286421862963145,
      "language_loss": 0.81146067,
      "learning_rate": 3.436886699092498e-06,
      "loss": 0.83370554,
      "num_input_tokens_seen": 47775255,
      "step": 2225,
      "time_per_iteration": 2.6691231727600098
    },
    {
      "auxiliary_loss_clip": 0.01084018,
      "auxiliary_loss_mlp": 0.01033446,
      "balance_loss_clip": 1.04229152,
      "balance_loss_mlp": 1.02306259,
      "epoch": 0.2676606745626165,
      "flos": 17485013157120.0,
      "grad_norm": 2.677316231712705,
      "language_loss": 0.71410799,
      "learning_rate": 3.4363447487553502e-06,
      "loss": 0.73528266,
      "num_input_tokens_seen": 47788570,
      "step": 2226,
      "time_per_iteration": 3.1174421310424805
    },
    {
      "auxiliary_loss_clip": 0.01197679,
      "auxiliary_loss_mlp": 0.01036847,
      "balance_loss_clip": 1.06513762,
      "balance_loss_mlp": 1.02695298,
      "epoch": 0.26778091745325555,
      "flos": 27852675143040.0,
      "grad_norm": 2.1369071979804,
      "language_loss": 0.77963388,
      "learning_rate": 3.4358025805258455e-06,
      "loss": 0.80197912,
      "num_input_tokens_seen": 47808275,
      "step": 2227,
      "time_per_iteration": 2.916728973388672
    },
    {
      "auxiliary_loss_clip": 0.01172553,
      "auxiliary_loss_mlp": 0.01043759,
      "balance_loss_clip": 1.06224704,
      "balance_loss_mlp": 1.03300059,
      "epoch": 0.26790116034389466,
      "flos": 20955995176320.0,
      "grad_norm": 2.0085251540423603,
      "language_loss": 0.8354305,
      "learning_rate": 3.435260194486232e-06,
      "loss": 0.8575936,
      "num_input_tokens_seen": 47826245,
      "step": 2228,
      "time_per_iteration": 2.7377333641052246
    },
    {
      "auxiliary_loss_clip": 0.01187464,
      "auxiliary_loss_mlp": 0.01035364,
      "balance_loss_clip": 1.06205666,
      "balance_loss_mlp": 1.02480197,
      "epoch": 0.2680214032345338,
      "flos": 18040659621120.0,
      "grad_norm": 2.6349091873117616,
      "language_loss": 0.82203197,
      "learning_rate": 3.4347175907187875e-06,
      "loss": 0.84426022,
      "num_input_tokens_seen": 47843235,
      "step": 2229,
      "time_per_iteration": 2.7249832153320312
    },
    {
      "auxiliary_loss_clip": 0.01175155,
      "auxiliary_loss_mlp": 0.01033177,
      "balance_loss_clip": 1.05663013,
      "balance_loss_mlp": 1.02356839,
      "epoch": 0.26814164612517283,
      "flos": 22419427086720.0,
      "grad_norm": 2.989005776792406,
      "language_loss": 0.87940979,
      "learning_rate": 3.4341747693058254e-06,
      "loss": 0.90149319,
      "num_input_tokens_seen": 47861710,
      "step": 2230,
      "time_per_iteration": 2.7510011196136475
    },
    {
      "auxiliary_loss_clip": 0.01131186,
      "auxiliary_loss_mlp": 0.01046293,
      "balance_loss_clip": 1.05528188,
      "balance_loss_mlp": 1.03677392,
      "epoch": 0.26826188901581194,
      "flos": 35627371159680.0,
      "grad_norm": 1.9301207386844268,
      "language_loss": 0.77568877,
      "learning_rate": 3.4336317303296916e-06,
      "loss": 0.79746366,
      "num_input_tokens_seen": 47882685,
      "step": 2231,
      "time_per_iteration": 2.9688894748687744
    },
    {
      "auxiliary_loss_clip": 0.01140906,
      "auxiliary_loss_mlp": 0.01033179,
      "balance_loss_clip": 1.050035,
      "balance_loss_mlp": 1.02358222,
      "epoch": 0.26838213190645105,
      "flos": 17639788861440.0,
      "grad_norm": 2.7632748635888205,
      "language_loss": 0.7593708,
      "learning_rate": 3.4330884738727635e-06,
      "loss": 0.78111166,
      "num_input_tokens_seen": 47900860,
      "step": 2232,
      "time_per_iteration": 2.7151923179626465
    },
    {
      "auxiliary_loss_clip": 0.0117976,
      "auxiliary_loss_mlp": 0.01044151,
      "balance_loss_clip": 1.05864811,
      "balance_loss_mlp": 1.03383899,
      "epoch": 0.2685023747970901,
      "flos": 22674823764480.0,
      "grad_norm": 3.603120796863301,
      "language_loss": 0.71081519,
      "learning_rate": 3.4325450000174535e-06,
      "loss": 0.73305428,
      "num_input_tokens_seen": 47917500,
      "step": 2233,
      "time_per_iteration": 2.7851791381835938
    },
    {
      "auxiliary_loss_clip": 0.01157997,
      "auxiliary_loss_mlp": 0.01034106,
      "balance_loss_clip": 1.0575211,
      "balance_loss_mlp": 1.02501583,
      "epoch": 0.2686226176877292,
      "flos": 20120533764480.0,
      "grad_norm": 1.8228631885537196,
      "language_loss": 0.7490055,
      "learning_rate": 3.4320013088462067e-06,
      "loss": 0.77092654,
      "num_input_tokens_seen": 47934860,
      "step": 2234,
      "time_per_iteration": 2.7473199367523193
    },
    {
      "auxiliary_loss_clip": 0.01165414,
      "auxiliary_loss_mlp": 0.01033811,
      "balance_loss_clip": 1.06028867,
      "balance_loss_mlp": 1.0249176,
      "epoch": 0.2687428605783683,
      "flos": 21872040750720.0,
      "grad_norm": 1.643397062333048,
      "language_loss": 0.8149972,
      "learning_rate": 3.431457400441499e-06,
      "loss": 0.83698946,
      "num_input_tokens_seen": 47955255,
      "step": 2235,
      "time_per_iteration": 2.7227320671081543
    },
    {
      "auxiliary_loss_clip": 0.01054977,
      "auxiliary_loss_mlp": 0.01012925,
      "balance_loss_clip": 1.01987052,
      "balance_loss_mlp": 1.01093423,
      "epoch": 0.2688631034690074,
      "flos": 69943320766080.0,
      "grad_norm": 0.9221265162482798,
      "language_loss": 0.6081028,
      "learning_rate": 3.4309132748858424e-06,
      "loss": 0.6287818,
      "num_input_tokens_seen": 48016245,
      "step": 2236,
      "time_per_iteration": 3.359496831893921
    },
    {
      "auxiliary_loss_clip": 0.01178048,
      "auxiliary_loss_mlp": 0.01030051,
      "balance_loss_clip": 1.06287861,
      "balance_loss_mlp": 1.02102733,
      "epoch": 0.2689833463596465,
      "flos": 22856639431680.0,
      "grad_norm": 2.080941262920627,
      "language_loss": 0.83962685,
      "learning_rate": 3.430368932261779e-06,
      "loss": 0.86170793,
      "num_input_tokens_seen": 48036600,
      "step": 2237,
      "time_per_iteration": 2.706033229827881
    },
    {
      "auxiliary_loss_clip": 0.01125687,
      "auxiliary_loss_mlp": 0.00763129,
      "balance_loss_clip": 1.04835343,
      "balance_loss_mlp": 1.00052524,
      "epoch": 0.2691035892502856,
      "flos": 17200242132480.0,
      "grad_norm": 2.0539206924816407,
      "language_loss": 0.75189322,
      "learning_rate": 3.429824372651886e-06,
      "loss": 0.7707814,
      "num_input_tokens_seen": 48054750,
      "step": 2238,
      "time_per_iteration": 2.8274788856506348
    },
    {
      "auxiliary_loss_clip": 0.01172017,
      "auxiliary_loss_mlp": 0.01037565,
      "balance_loss_clip": 1.06249928,
      "balance_loss_mlp": 1.02778411,
      "epoch": 0.26922383214092466,
      "flos": 17747484814080.0,
      "grad_norm": 3.967934488950613,
      "language_loss": 0.84328556,
      "learning_rate": 3.4292795961387732e-06,
      "loss": 0.86538136,
      "num_input_tokens_seen": 48072650,
      "step": 2239,
      "time_per_iteration": 2.6998181343078613
    },
    {
      "auxiliary_loss_clip": 0.01202462,
      "auxiliary_loss_mlp": 0.01045389,
      "balance_loss_clip": 1.06224394,
      "balance_loss_mlp": 1.03593588,
      "epoch": 0.26934407503156377,
      "flos": 16173376122240.0,
      "grad_norm": 2.3982540567022967,
      "language_loss": 0.87823772,
      "learning_rate": 3.4287346028050818e-06,
      "loss": 0.90071625,
      "num_input_tokens_seen": 48088720,
      "step": 2240,
      "time_per_iteration": 2.674835205078125
    },
    {
      "auxiliary_loss_clip": 0.01167654,
      "auxiliary_loss_mlp": 0.01030163,
      "balance_loss_clip": 1.06055784,
      "balance_loss_mlp": 1.02188993,
      "epoch": 0.2694643179222028,
      "flos": 23732895715200.0,
      "grad_norm": 1.6399828893613655,
      "language_loss": 0.80136871,
      "learning_rate": 3.4281893927334866e-06,
      "loss": 0.82334691,
      "num_input_tokens_seen": 48108630,
      "step": 2241,
      "time_per_iteration": 2.863368034362793
    },
    {
      "auxiliary_loss_clip": 0.01213428,
      "auxiliary_loss_mlp": 0.00763051,
      "balance_loss_clip": 1.06675363,
      "balance_loss_mlp": 1.00059175,
      "epoch": 0.26958456081284193,
      "flos": 24718140840960.0,
      "grad_norm": 3.3629183456788847,
      "language_loss": 0.75556791,
      "learning_rate": 3.4276439660066963e-06,
      "loss": 0.77533269,
      "num_input_tokens_seen": 48128330,
      "step": 2242,
      "time_per_iteration": 2.6660587787628174
    },
    {
      "auxiliary_loss_clip": 0.01151261,
      "auxiliary_loss_mlp": 0.01033983,
      "balance_loss_clip": 1.05992949,
      "balance_loss_mlp": 1.02528644,
      "epoch": 0.26970480370348104,
      "flos": 18112588606080.0,
      "grad_norm": 2.2987148789811167,
      "language_loss": 0.8496483,
      "learning_rate": 3.427098322707452e-06,
      "loss": 0.87150073,
      "num_input_tokens_seen": 48144295,
      "step": 2243,
      "time_per_iteration": 2.7221875190734863
    },
    {
      "auxiliary_loss_clip": 0.0111278,
      "auxiliary_loss_mlp": 0.01045818,
      "balance_loss_clip": 1.05841672,
      "balance_loss_mlp": 1.03606629,
      "epoch": 0.2698250465941201,
      "flos": 10816546250880.0,
      "grad_norm": 3.057258821745124,
      "language_loss": 0.9000544,
      "learning_rate": 3.426552462918526e-06,
      "loss": 0.9216404,
      "num_input_tokens_seen": 48162230,
      "step": 2244,
      "time_per_iteration": 3.931208848953247
    },
    {
      "auxiliary_loss_clip": 0.01195661,
      "auxiliary_loss_mlp": 0.01031997,
      "balance_loss_clip": 1.06538033,
      "balance_loss_mlp": 1.02273464,
      "epoch": 0.2699452894847592,
      "flos": 17308117653120.0,
      "grad_norm": 2.621665978933489,
      "language_loss": 0.72672725,
      "learning_rate": 3.426006386722726e-06,
      "loss": 0.74900383,
      "num_input_tokens_seen": 48180290,
      "step": 2245,
      "time_per_iteration": 4.230551242828369
    },
    {
      "auxiliary_loss_clip": 0.01221424,
      "auxiliary_loss_mlp": 0.01040488,
      "balance_loss_clip": 1.07087946,
      "balance_loss_mlp": 1.03109419,
      "epoch": 0.2700655323753983,
      "flos": 18078150441600.0,
      "grad_norm": 2.1010143573682347,
      "language_loss": 0.92990118,
      "learning_rate": 3.4254600942028914e-06,
      "loss": 0.95252025,
      "num_input_tokens_seen": 48198165,
      "step": 2246,
      "time_per_iteration": 2.684384822845459
    },
    {
      "auxiliary_loss_clip": 0.01173454,
      "auxiliary_loss_mlp": 0.01036564,
      "balance_loss_clip": 1.05889535,
      "balance_loss_mlp": 1.02750981,
      "epoch": 0.2701857752660374,
      "flos": 18186636493440.0,
      "grad_norm": 2.6326122264842233,
      "language_loss": 0.8320716,
      "learning_rate": 3.424913585441893e-06,
      "loss": 0.85417181,
      "num_input_tokens_seen": 48216000,
      "step": 2247,
      "time_per_iteration": 3.567511558532715
    },
    {
      "auxiliary_loss_clip": 0.0118341,
      "auxiliary_loss_mlp": 0.01038912,
      "balance_loss_clip": 1.06196713,
      "balance_loss_mlp": 1.02934575,
      "epoch": 0.2703060181566765,
      "flos": 16319496648960.0,
      "grad_norm": 2.406369944378646,
      "language_loss": 0.87626004,
      "learning_rate": 3.4243668605226374e-06,
      "loss": 0.89848322,
      "num_input_tokens_seen": 48233025,
      "step": 2248,
      "time_per_iteration": 2.7726235389709473
    },
    {
      "auxiliary_loss_clip": 0.01180549,
      "auxiliary_loss_mlp": 0.01040148,
      "balance_loss_clip": 1.05891633,
      "balance_loss_mlp": 1.0301826,
      "epoch": 0.2704262610473156,
      "flos": 19572357329280.0,
      "grad_norm": 2.207548574026253,
      "language_loss": 0.83148295,
      "learning_rate": 3.423819919528061e-06,
      "loss": 0.85368991,
      "num_input_tokens_seen": 48251110,
      "step": 2249,
      "time_per_iteration": 2.7848362922668457
    },
    {
      "auxiliary_loss_clip": 0.01134327,
      "auxiliary_loss_mlp": 0.01035797,
      "balance_loss_clip": 1.05121732,
      "balance_loss_mlp": 1.02588511,
      "epoch": 0.27054650393795465,
      "flos": 20740746925440.0,
      "grad_norm": 1.9164034159681498,
      "language_loss": 0.78278536,
      "learning_rate": 3.4232727625411355e-06,
      "loss": 0.80448663,
      "num_input_tokens_seen": 48270215,
      "step": 2250,
      "time_per_iteration": 3.6705305576324463
    },
    {
      "auxiliary_loss_clip": 0.01178327,
      "auxiliary_loss_mlp": 0.01038902,
      "balance_loss_clip": 1.05912912,
      "balance_loss_mlp": 1.0296874,
      "epoch": 0.27066674682859376,
      "flos": 18658322916480.0,
      "grad_norm": 3.518612102734593,
      "language_loss": 0.86436027,
      "learning_rate": 3.4227253896448626e-06,
      "loss": 0.88653255,
      "num_input_tokens_seen": 48288075,
      "step": 2251,
      "time_per_iteration": 2.7748453617095947
    },
    {
      "auxiliary_loss_clip": 0.01163769,
      "auxiliary_loss_mlp": 0.01036458,
      "balance_loss_clip": 1.05993366,
      "balance_loss_mlp": 1.02696872,
      "epoch": 0.2707869897192329,
      "flos": 23002759958400.0,
      "grad_norm": 3.066394477330713,
      "language_loss": 0.82117534,
      "learning_rate": 3.42217780092228e-06,
      "loss": 0.84317756,
      "num_input_tokens_seen": 48306415,
      "step": 2252,
      "time_per_iteration": 2.7248735427856445
    },
    {
      "auxiliary_loss_clip": 0.01094278,
      "auxiliary_loss_mlp": 0.00754776,
      "balance_loss_clip": 1.03056931,
      "balance_loss_mlp": 1.00036955,
      "epoch": 0.27090723260987193,
      "flos": 58323240293760.0,
      "grad_norm": 0.7958355799872049,
      "language_loss": 0.60269165,
      "learning_rate": 3.421629996456456e-06,
      "loss": 0.62118214,
      "num_input_tokens_seen": 48365035,
      "step": 2253,
      "time_per_iteration": 3.1966304779052734
    },
    {
      "auxiliary_loss_clip": 0.0116469,
      "auxiliary_loss_mlp": 0.01045022,
      "balance_loss_clip": 1.06372249,
      "balance_loss_mlp": 1.03514528,
      "epoch": 0.27102747550051104,
      "flos": 11984540797440.0,
      "grad_norm": 3.466614075040941,
      "language_loss": 0.8258667,
      "learning_rate": 3.421081976330491e-06,
      "loss": 0.84796381,
      "num_input_tokens_seen": 48383550,
      "step": 2254,
      "time_per_iteration": 2.746033191680908
    },
    {
      "auxiliary_loss_clip": 0.0121177,
      "auxiliary_loss_mlp": 0.01038492,
      "balance_loss_clip": 1.06523824,
      "balance_loss_mlp": 1.0294621,
      "epoch": 0.27114771839115015,
      "flos": 19900401264000.0,
      "grad_norm": 2.118454934374008,
      "language_loss": 0.88371998,
      "learning_rate": 3.4205337406275207e-06,
      "loss": 0.90622258,
      "num_input_tokens_seen": 48403670,
      "step": 2255,
      "time_per_iteration": 2.667325258255005
    },
    {
      "auxiliary_loss_clip": 0.01136475,
      "auxiliary_loss_mlp": 0.01037481,
      "balance_loss_clip": 1.0564158,
      "balance_loss_mlp": 1.02799237,
      "epoch": 0.2712679612817892,
      "flos": 18331966920960.0,
      "grad_norm": 5.796488298731521,
      "language_loss": 0.75869322,
      "learning_rate": 3.4199852894307114e-06,
      "loss": 0.78043282,
      "num_input_tokens_seen": 48420420,
      "step": 2256,
      "time_per_iteration": 2.826124429702759
    },
    {
      "auxiliary_loss_clip": 0.01150641,
      "auxiliary_loss_mlp": 0.01039294,
      "balance_loss_clip": 1.05882883,
      "balance_loss_mlp": 1.0294534,
      "epoch": 0.2713882041724283,
      "flos": 24460302038400.0,
      "grad_norm": 2.3116560062099234,
      "language_loss": 0.79411137,
      "learning_rate": 3.419436622823262e-06,
      "loss": 0.81601071,
      "num_input_tokens_seen": 48441140,
      "step": 2257,
      "time_per_iteration": 2.8176586627960205
    },
    {
      "auxiliary_loss_clip": 0.01194286,
      "auxiliary_loss_mlp": 0.01037008,
      "balance_loss_clip": 1.06212962,
      "balance_loss_mlp": 1.02778733,
      "epoch": 0.27150844706306737,
      "flos": 23039317025280.0,
      "grad_norm": 1.7631864742651648,
      "language_loss": 0.74417353,
      "learning_rate": 3.4188877408884063e-06,
      "loss": 0.76648653,
      "num_input_tokens_seen": 48461845,
      "step": 2258,
      "time_per_iteration": 2.712289571762085
    },
    {
      "auxiliary_loss_clip": 0.01167581,
      "auxiliary_loss_mlp": 0.01036406,
      "balance_loss_clip": 1.06102228,
      "balance_loss_mlp": 1.02657127,
      "epoch": 0.2716286899537065,
      "flos": 22563644192640.0,
      "grad_norm": 2.329125258167846,
      "language_loss": 0.65019071,
      "learning_rate": 3.4183386437094088e-06,
      "loss": 0.6722306,
      "num_input_tokens_seen": 48478510,
      "step": 2259,
      "time_per_iteration": 2.7160651683807373
    },
    {
      "auxiliary_loss_clip": 0.01195504,
      "auxiliary_loss_mlp": 0.01034277,
      "balance_loss_clip": 1.06205595,
      "balance_loss_mlp": 1.02420998,
      "epoch": 0.2717489328443456,
      "flos": 13115044523520.0,
      "grad_norm": 3.2729870455428944,
      "language_loss": 0.82629514,
      "learning_rate": 3.417789331369565e-06,
      "loss": 0.848593,
      "num_input_tokens_seen": 48494300,
      "step": 2260,
      "time_per_iteration": 2.694321632385254
    },
    {
      "auxiliary_loss_clip": 0.01145969,
      "auxiliary_loss_mlp": 0.01039705,
      "balance_loss_clip": 1.05293489,
      "balance_loss_mlp": 1.02973938,
      "epoch": 0.27186917573498465,
      "flos": 29278688060160.0,
      "grad_norm": 2.075916035966282,
      "language_loss": 0.91371077,
      "learning_rate": 3.4172398039522088e-06,
      "loss": 0.9355675,
      "num_input_tokens_seen": 48515585,
      "step": 2261,
      "time_per_iteration": 2.7973694801330566
    },
    {
      "auxiliary_loss_clip": 0.01207144,
      "auxiliary_loss_mlp": 0.01036225,
      "balance_loss_clip": 1.06531584,
      "balance_loss_mlp": 1.02619946,
      "epoch": 0.27198941862562376,
      "flos": 26032220000640.0,
      "grad_norm": 2.23310990646002,
      "language_loss": 0.80069923,
      "learning_rate": 3.4166900615407e-06,
      "loss": 0.82313287,
      "num_input_tokens_seen": 48533500,
      "step": 2262,
      "time_per_iteration": 2.782731294631958
    },
    {
      "auxiliary_loss_clip": 0.01197071,
      "auxiliary_loss_mlp": 0.01035545,
      "balance_loss_clip": 1.06029844,
      "balance_loss_mlp": 1.02650905,
      "epoch": 0.27210966151626287,
      "flos": 32780983760640.0,
      "grad_norm": 2.270529000790642,
      "language_loss": 0.744699,
      "learning_rate": 3.416140104218436e-06,
      "loss": 0.76702511,
      "num_input_tokens_seen": 48552865,
      "step": 2263,
      "time_per_iteration": 2.8029074668884277
    },
    {
      "auxiliary_loss_clip": 0.01085453,
      "auxiliary_loss_mlp": 0.01007737,
      "balance_loss_clip": 1.01793289,
      "balance_loss_mlp": 1.00591278,
      "epoch": 0.2722299044069019,
      "flos": 65471043219840.0,
      "grad_norm": 0.8449162084142388,
      "language_loss": 0.69627917,
      "learning_rate": 3.4155899320688437e-06,
      "loss": 0.71721107,
      "num_input_tokens_seen": 48618940,
      "step": 2264,
      "time_per_iteration": 3.2607102394104004
    },
    {
      "auxiliary_loss_clip": 0.01181003,
      "auxiliary_loss_mlp": 0.01035529,
      "balance_loss_clip": 1.06372178,
      "balance_loss_mlp": 1.02601004,
      "epoch": 0.27235014729754103,
      "flos": 15334143782400.0,
      "grad_norm": 2.712557746882823,
      "language_loss": 0.74572951,
      "learning_rate": 3.415039545175384e-06,
      "loss": 0.7678948,
      "num_input_tokens_seen": 48634665,
      "step": 2265,
      "time_per_iteration": 2.7032392024993896
    },
    {
      "auxiliary_loss_clip": 0.01149284,
      "auxiliary_loss_mlp": 0.01028405,
      "balance_loss_clip": 1.05347741,
      "balance_loss_mlp": 1.01933956,
      "epoch": 0.27247039018818014,
      "flos": 21872363973120.0,
      "grad_norm": 2.374131242663476,
      "language_loss": 0.65554774,
      "learning_rate": 3.414488943621551e-06,
      "loss": 0.67732471,
      "num_input_tokens_seen": 48653330,
      "step": 2266,
      "time_per_iteration": 2.88737154006958
    },
    {
      "auxiliary_loss_clip": 0.01177273,
      "auxiliary_loss_mlp": 0.01032621,
      "balance_loss_clip": 1.05875921,
      "balance_loss_mlp": 1.02319765,
      "epoch": 0.2725906330788192,
      "flos": 18695490514560.0,
      "grad_norm": 6.8400429599159445,
      "language_loss": 0.74111784,
      "learning_rate": 3.41393812749087e-06,
      "loss": 0.76321673,
      "num_input_tokens_seen": 48671375,
      "step": 2267,
      "time_per_iteration": 2.768723487854004
    },
    {
      "auxiliary_loss_clip": 0.01183646,
      "auxiliary_loss_mlp": 0.01037175,
      "balance_loss_clip": 1.06323326,
      "balance_loss_mlp": 1.0279541,
      "epoch": 0.2727108759694583,
      "flos": 17886099398400.0,
      "grad_norm": 3.6584717693207014,
      "language_loss": 0.72264278,
      "learning_rate": 3.4133870968668984e-06,
      "loss": 0.74485099,
      "num_input_tokens_seen": 48686175,
      "step": 2268,
      "time_per_iteration": 2.6093509197235107
    },
    {
      "auxiliary_loss_clip": 0.01195489,
      "auxiliary_loss_mlp": 0.01038287,
      "balance_loss_clip": 1.06311035,
      "balance_loss_mlp": 1.02801168,
      "epoch": 0.2728311188600974,
      "flos": 24461666755200.0,
      "grad_norm": 1.9898034949972492,
      "language_loss": 0.79002249,
      "learning_rate": 3.412835851833229e-06,
      "loss": 0.81236029,
      "num_input_tokens_seen": 48708370,
      "step": 2269,
      "time_per_iteration": 2.760483980178833
    },
    {
      "auxiliary_loss_clip": 0.01181286,
      "auxiliary_loss_mlp": 0.01040417,
      "balance_loss_clip": 1.05938125,
      "balance_loss_mlp": 1.03080904,
      "epoch": 0.2729513617507365,
      "flos": 30993314757120.0,
      "grad_norm": 3.2551380530678258,
      "language_loss": 0.78047872,
      "learning_rate": 3.4122843924734834e-06,
      "loss": 0.80269575,
      "num_input_tokens_seen": 48730670,
      "step": 2270,
      "time_per_iteration": 3.6941728591918945
    },
    {
      "auxiliary_loss_clip": 0.01195195,
      "auxiliary_loss_mlp": 0.01035774,
      "balance_loss_clip": 1.06026363,
      "balance_loss_mlp": 1.02629662,
      "epoch": 0.2730716046413756,
      "flos": 19094637421440.0,
      "grad_norm": 3.5641130218942405,
      "language_loss": 0.87994504,
      "learning_rate": 3.411732718871319e-06,
      "loss": 0.9022547,
      "num_input_tokens_seen": 48746510,
      "step": 2271,
      "time_per_iteration": 3.582289457321167
    },
    {
      "auxiliary_loss_clip": 0.01128156,
      "auxiliary_loss_mlp": 0.01040511,
      "balance_loss_clip": 1.0522846,
      "balance_loss_mlp": 1.03171384,
      "epoch": 0.27319184753201464,
      "flos": 26944566474240.0,
      "grad_norm": 1.8502190511319418,
      "language_loss": 0.78857565,
      "learning_rate": 3.4111808311104227e-06,
      "loss": 0.81026232,
      "num_input_tokens_seen": 48768825,
      "step": 2272,
      "time_per_iteration": 2.9093244075775146
    },
    {
      "auxiliary_loss_clip": 0.01168426,
      "auxiliary_loss_mlp": 0.01033027,
      "balance_loss_clip": 1.0593518,
      "balance_loss_mlp": 1.02343082,
      "epoch": 0.27331209042265375,
      "flos": 31759828012800.0,
      "grad_norm": 1.9764621849270305,
      "language_loss": 0.69675279,
      "learning_rate": 3.410628729274517e-06,
      "loss": 0.71876734,
      "num_input_tokens_seen": 48790345,
      "step": 2273,
      "time_per_iteration": 3.750950336456299
    },
    {
      "auxiliary_loss_clip": 0.01214099,
      "auxiliary_loss_mlp": 0.01036293,
      "balance_loss_clip": 1.06498861,
      "balance_loss_mlp": 1.02705395,
      "epoch": 0.27343233331329286,
      "flos": 25739081107200.0,
      "grad_norm": 3.5441792446656177,
      "language_loss": 0.83162749,
      "learning_rate": 3.4100764134473546e-06,
      "loss": 0.85413146,
      "num_input_tokens_seen": 48809630,
      "step": 2274,
      "time_per_iteration": 2.671018362045288
    },
    {
      "auxiliary_loss_clip": 0.01157615,
      "auxiliary_loss_mlp": 0.01037453,
      "balance_loss_clip": 1.0559752,
      "balance_loss_mlp": 1.02855968,
      "epoch": 0.2735525762039319,
      "flos": 24389414547840.0,
      "grad_norm": 2.541919992767346,
      "language_loss": 0.85083151,
      "learning_rate": 3.4095238837127215e-06,
      "loss": 0.87278223,
      "num_input_tokens_seen": 48828770,
      "step": 2275,
      "time_per_iteration": 3.7418723106384277
    },
    {
      "auxiliary_loss_clip": 0.01187011,
      "auxiliary_loss_mlp": 0.0076372,
      "balance_loss_clip": 1.05751467,
      "balance_loss_mlp": 1.00060904,
      "epoch": 0.27367281909457103,
      "flos": 14465357527680.0,
      "grad_norm": 2.4350443396006725,
      "language_loss": 0.79741716,
      "learning_rate": 3.4089711401544355e-06,
      "loss": 0.81692445,
      "num_input_tokens_seen": 48846365,
      "step": 2276,
      "time_per_iteration": 2.6859352588653564
    },
    {
      "auxiliary_loss_clip": 0.0117498,
      "auxiliary_loss_mlp": 0.01033627,
      "balance_loss_clip": 1.05631757,
      "balance_loss_mlp": 1.0236907,
      "epoch": 0.27379306198521014,
      "flos": 23476996247040.0,
      "grad_norm": 2.297285063640222,
      "language_loss": 0.6789068,
      "learning_rate": 3.4084181828563486e-06,
      "loss": 0.70099288,
      "num_input_tokens_seen": 48863085,
      "step": 2277,
      "time_per_iteration": 2.7263309955596924
    },
    {
      "auxiliary_loss_clip": 0.01173673,
      "auxiliary_loss_mlp": 0.01041964,
      "balance_loss_clip": 1.05708027,
      "balance_loss_mlp": 1.03239202,
      "epoch": 0.2739133048758492,
      "flos": 17458152762240.0,
      "grad_norm": 1.76417460977482,
      "language_loss": 0.70827806,
      "learning_rate": 3.4078650119023428e-06,
      "loss": 0.73043442,
      "num_input_tokens_seen": 48881400,
      "step": 2278,
      "time_per_iteration": 2.7257893085479736
    },
    {
      "auxiliary_loss_clip": 0.01155455,
      "auxiliary_loss_mlp": 0.01043184,
      "balance_loss_clip": 1.05861604,
      "balance_loss_mlp": 1.03300405,
      "epoch": 0.2740335477664883,
      "flos": 19273113123840.0,
      "grad_norm": 2.6178642657797013,
      "language_loss": 0.74140626,
      "learning_rate": 3.4073116273763337e-06,
      "loss": 0.76339263,
      "num_input_tokens_seen": 48895845,
      "step": 2279,
      "time_per_iteration": 2.727224349975586
    },
    {
      "auxiliary_loss_clip": 0.01170964,
      "auxiliary_loss_mlp": 0.00764168,
      "balance_loss_clip": 1.05632365,
      "balance_loss_mlp": 1.00058246,
      "epoch": 0.2741537906571274,
      "flos": 26104723603200.0,
      "grad_norm": 2.489387623882007,
      "language_loss": 0.81534207,
      "learning_rate": 3.40675802936227e-06,
      "loss": 0.83469337,
      "num_input_tokens_seen": 48916630,
      "step": 2280,
      "time_per_iteration": 2.809309720993042
    },
    {
      "auxiliary_loss_clip": 0.01168506,
      "auxiliary_loss_mlp": 0.01037418,
      "balance_loss_clip": 1.0604825,
      "balance_loss_mlp": 1.02838826,
      "epoch": 0.27427403354776647,
      "flos": 34164190644480.0,
      "grad_norm": 2.424692836422709,
      "language_loss": 0.72135675,
      "learning_rate": 3.4062042179441318e-06,
      "loss": 0.74341601,
      "num_input_tokens_seen": 48937100,
      "step": 2281,
      "time_per_iteration": 2.822106122970581
    },
    {
      "auxiliary_loss_clip": 0.01198462,
      "auxiliary_loss_mlp": 0.01038823,
      "balance_loss_clip": 1.06051993,
      "balance_loss_mlp": 1.02993011,
      "epoch": 0.2743942764384056,
      "flos": 18766988536320.0,
      "grad_norm": 1.8246843365362728,
      "language_loss": 0.80565023,
      "learning_rate": 3.4056501932059314e-06,
      "loss": 0.82802308,
      "num_input_tokens_seen": 48955175,
      "step": 2282,
      "time_per_iteration": 2.698129892349243
    },
    {
      "auxiliary_loss_clip": 0.01087995,
      "auxiliary_loss_mlp": 0.01001952,
      "balance_loss_clip": 1.02142787,
      "balance_loss_mlp": 1.00009251,
      "epoch": 0.2745145193290447,
      "flos": 64904048058240.0,
      "grad_norm": 0.7696992968414491,
      "language_loss": 0.58043802,
      "learning_rate": 3.405095955231715e-06,
      "loss": 0.60133755,
      "num_input_tokens_seen": 49006830,
      "step": 2283,
      "time_per_iteration": 3.198923349380493
    },
    {
      "auxiliary_loss_clip": 0.0115283,
      "auxiliary_loss_mlp": 0.0103221,
      "balance_loss_clip": 1.05401731,
      "balance_loss_mlp": 1.02291131,
      "epoch": 0.27463476221968375,
      "flos": 16136926796160.0,
      "grad_norm": 2.5838927913588496,
      "language_loss": 0.94632828,
      "learning_rate": 3.4045415041055585e-06,
      "loss": 0.96817869,
      "num_input_tokens_seen": 49022470,
      "step": 2284,
      "time_per_iteration": 2.7554092407226562
    },
    {
      "auxiliary_loss_clip": 0.0114431,
      "auxiliary_loss_mlp": 0.01035492,
      "balance_loss_clip": 1.04961181,
      "balance_loss_mlp": 1.02543044,
      "epoch": 0.27475500511032286,
      "flos": 10376712213120.0,
      "grad_norm": 4.637400612256782,
      "language_loss": 0.78430384,
      "learning_rate": 3.4039868399115728e-06,
      "loss": 0.8061018,
      "num_input_tokens_seen": 49037110,
      "step": 2285,
      "time_per_iteration": 2.7860512733459473
    },
    {
      "auxiliary_loss_clip": 0.01193126,
      "auxiliary_loss_mlp": 0.01037286,
      "balance_loss_clip": 1.06568456,
      "balance_loss_mlp": 1.02785015,
      "epoch": 0.27487524800096197,
      "flos": 17311062568320.0,
      "grad_norm": 24.05733795966888,
      "language_loss": 0.80395502,
      "learning_rate": 3.4034319627339003e-06,
      "loss": 0.82625914,
      "num_input_tokens_seen": 49053975,
      "step": 2286,
      "time_per_iteration": 2.691906690597534
    },
    {
      "auxiliary_loss_clip": 0.01205489,
      "auxiliary_loss_mlp": 0.01033963,
      "balance_loss_clip": 1.06407213,
      "balance_loss_mlp": 1.02454591,
      "epoch": 0.274995490891601,
      "flos": 27120205002240.0,
      "grad_norm": 2.8886117250623653,
      "language_loss": 0.69481146,
      "learning_rate": 3.402876872656715e-06,
      "loss": 0.717206,
      "num_input_tokens_seen": 49072295,
      "step": 2287,
      "time_per_iteration": 2.749523162841797
    },
    {
      "auxiliary_loss_clip": 0.01141763,
      "auxiliary_loss_mlp": 0.01031814,
      "balance_loss_clip": 1.05144215,
      "balance_loss_mlp": 1.02213979,
      "epoch": 0.27511573378224013,
      "flos": 23436093634560.0,
      "grad_norm": 1.8384877271077271,
      "language_loss": 0.89310396,
      "learning_rate": 3.402321569764223e-06,
      "loss": 0.91483974,
      "num_input_tokens_seen": 49091600,
      "step": 2288,
      "time_per_iteration": 2.781170129776001
    },
    {
      "auxiliary_loss_clip": 0.01172647,
      "auxiliary_loss_mlp": 0.01038066,
      "balance_loss_clip": 1.05668068,
      "balance_loss_mlp": 1.02795076,
      "epoch": 0.2752359766728792,
      "flos": 16722019434240.0,
      "grad_norm": 5.284177548984342,
      "language_loss": 0.83610904,
      "learning_rate": 3.4017660541406635e-06,
      "loss": 0.85821611,
      "num_input_tokens_seen": 49107665,
      "step": 2289,
      "time_per_iteration": 2.775087833404541
    },
    {
      "auxiliary_loss_clip": 0.01184906,
      "auxiliary_loss_mlp": 0.0103415,
      "balance_loss_clip": 1.0565536,
      "balance_loss_mlp": 1.02411819,
      "epoch": 0.2753562195635183,
      "flos": 25297738698240.0,
      "grad_norm": 2.3657679035473724,
      "language_loss": 0.74625665,
      "learning_rate": 3.4012103258703092e-06,
      "loss": 0.76844722,
      "num_input_tokens_seen": 49126420,
      "step": 2290,
      "time_per_iteration": 2.8033041954040527
    },
    {
      "auxiliary_loss_clip": 0.01148004,
      "auxiliary_loss_mlp": 0.01042645,
      "balance_loss_clip": 1.05542088,
      "balance_loss_mlp": 1.0319761,
      "epoch": 0.2754764624541574,
      "flos": 27338972785920.0,
      "grad_norm": 4.905645039442069,
      "language_loss": 0.83219981,
      "learning_rate": 3.4006543850374616e-06,
      "loss": 0.85410631,
      "num_input_tokens_seen": 49141470,
      "step": 2291,
      "time_per_iteration": 2.900672435760498
    },
    {
      "auxiliary_loss_clip": 0.01191031,
      "auxiliary_loss_mlp": 0.01036985,
      "balance_loss_clip": 1.05993342,
      "balance_loss_mlp": 1.02730489,
      "epoch": 0.27559670534479647,
      "flos": 17238379397760.0,
      "grad_norm": 2.48384613160417,
      "language_loss": 0.75325096,
      "learning_rate": 3.400098231726458e-06,
      "loss": 0.77553111,
      "num_input_tokens_seen": 49158570,
      "step": 2292,
      "time_per_iteration": 2.6904966831207275
    },
    {
      "auxiliary_loss_clip": 0.01186434,
      "auxiliary_loss_mlp": 0.01040123,
      "balance_loss_clip": 1.05893016,
      "balance_loss_mlp": 1.02906656,
      "epoch": 0.2757169482354356,
      "flos": 21939085486080.0,
      "grad_norm": 2.3215965493548496,
      "language_loss": 0.87312233,
      "learning_rate": 3.3995418660216657e-06,
      "loss": 0.89538789,
      "num_input_tokens_seen": 49176025,
      "step": 2293,
      "time_per_iteration": 2.678966999053955
    },
    {
      "auxiliary_loss_clip": 0.01186161,
      "auxiliary_loss_mlp": 0.0104204,
      "balance_loss_clip": 1.05918002,
      "balance_loss_mlp": 1.03126943,
      "epoch": 0.2758371911260747,
      "flos": 20850669521280.0,
      "grad_norm": 2.9778654390773953,
      "language_loss": 0.80364901,
      "learning_rate": 3.3989852880074848e-06,
      "loss": 0.82593107,
      "num_input_tokens_seen": 49197455,
      "step": 2294,
      "time_per_iteration": 2.8082799911499023
    },
    {
      "auxiliary_loss_clip": 0.01095056,
      "auxiliary_loss_mlp": 0.01004475,
      "balance_loss_clip": 1.02285218,
      "balance_loss_mlp": 1.00267458,
      "epoch": 0.27595743401671374,
      "flos": 69269063592960.0,
      "grad_norm": 0.7497776549249359,
      "language_loss": 0.60571539,
      "learning_rate": 3.398428497768348e-06,
      "loss": 0.62671077,
      "num_input_tokens_seen": 49262625,
      "step": 2295,
      "time_per_iteration": 4.260572195053101
    },
    {
      "auxiliary_loss_clip": 0.01199553,
      "auxiliary_loss_mlp": 0.01039961,
      "balance_loss_clip": 1.06436181,
      "balance_loss_mlp": 1.03006637,
      "epoch": 0.27607767690735285,
      "flos": 21215019127680.0,
      "grad_norm": 1.9938371323675126,
      "language_loss": 0.71945786,
      "learning_rate": 3.3978714953887205e-06,
      "loss": 0.74185294,
      "num_input_tokens_seen": 49282380,
      "step": 2296,
      "time_per_iteration": 2.704340934753418
    },
    {
      "auxiliary_loss_clip": 0.01166355,
      "auxiliary_loss_mlp": 0.01039126,
      "balance_loss_clip": 1.05668616,
      "balance_loss_mlp": 1.0291301,
      "epoch": 0.27619791979799196,
      "flos": 24825334003200.0,
      "grad_norm": 9.973746231644517,
      "language_loss": 0.86134499,
      "learning_rate": 3.397314280953098e-06,
      "loss": 0.88339978,
      "num_input_tokens_seen": 49303205,
      "step": 2297,
      "time_per_iteration": 3.8198866844177246
    },
    {
      "auxiliary_loss_clip": 0.01169726,
      "auxiliary_loss_mlp": 0.01042781,
      "balance_loss_clip": 1.06180859,
      "balance_loss_mlp": 1.03323865,
      "epoch": 0.276318162688631,
      "flos": 24753548672640.0,
      "grad_norm": 2.143107263627781,
      "language_loss": 0.80873287,
      "learning_rate": 3.3967568545460108e-06,
      "loss": 0.83085787,
      "num_input_tokens_seen": 49322745,
      "step": 2298,
      "time_per_iteration": 2.756283760070801
    },
    {
      "auxiliary_loss_clip": 0.01166664,
      "auxiliary_loss_mlp": 0.00764297,
      "balance_loss_clip": 1.06094432,
      "balance_loss_mlp": 1.00053668,
      "epoch": 0.27643840557927013,
      "flos": 18150007599360.0,
      "grad_norm": 2.3196960134061166,
      "language_loss": 0.80860096,
      "learning_rate": 3.3961992162520185e-06,
      "loss": 0.8279106,
      "num_input_tokens_seen": 49341370,
      "step": 2299,
      "time_per_iteration": 3.6854240894317627
    },
    {
      "auxiliary_loss_clip": 0.0118551,
      "auxiliary_loss_mlp": 0.01032013,
      "balance_loss_clip": 1.05993927,
      "balance_loss_mlp": 1.02106404,
      "epoch": 0.27655864846990924,
      "flos": 24823933372800.0,
      "grad_norm": 3.261249666034426,
      "language_loss": 0.71959019,
      "learning_rate": 3.3956413661557156e-06,
      "loss": 0.74176538,
      "num_input_tokens_seen": 49361545,
      "step": 2300,
      "time_per_iteration": 3.7246859073638916
    },
    {
      "auxiliary_loss_clip": 0.01190298,
      "auxiliary_loss_mlp": 0.01040326,
      "balance_loss_clip": 1.0621767,
      "balance_loss_mlp": 1.0300622,
      "epoch": 0.2766788913605483,
      "flos": 20266582464000.0,
      "grad_norm": 4.029160780742891,
      "language_loss": 0.66215688,
      "learning_rate": 3.3950833043417273e-06,
      "loss": 0.68446314,
      "num_input_tokens_seen": 49379690,
      "step": 2301,
      "time_per_iteration": 2.6826260089874268
    },
    {
      "auxiliary_loss_clip": 0.01146172,
      "auxiliary_loss_mlp": 0.0103693,
      "balance_loss_clip": 1.05727696,
      "balance_loss_mlp": 1.02592707,
      "epoch": 0.2767991342511874,
      "flos": 21470272151040.0,
      "grad_norm": 2.1241254670849914,
      "language_loss": 0.72924918,
      "learning_rate": 3.3945250308947105e-06,
      "loss": 0.75108021,
      "num_input_tokens_seen": 49395995,
      "step": 2302,
      "time_per_iteration": 2.8288252353668213
    },
    {
      "auxiliary_loss_clip": 0.0107278,
      "auxiliary_loss_mlp": 0.01007644,
      "balance_loss_clip": 1.01840818,
      "balance_loss_mlp": 1.00572479,
      "epoch": 0.2769193771418265,
      "flos": 66002627571840.0,
      "grad_norm": 1.3303824068855479,
      "language_loss": 0.68287516,
      "learning_rate": 3.3939665458993556e-06,
      "loss": 0.70367944,
      "num_input_tokens_seen": 49450415,
      "step": 2303,
      "time_per_iteration": 3.2818496227264404
    },
    {
      "auxiliary_loss_clip": 0.01169053,
      "auxiliary_loss_mlp": 0.01035235,
      "balance_loss_clip": 1.05983186,
      "balance_loss_mlp": 1.02475595,
      "epoch": 0.27703962003246557,
      "flos": 20704441253760.0,
      "grad_norm": 3.385737284357854,
      "language_loss": 0.76892805,
      "learning_rate": 3.3934078494403843e-06,
      "loss": 0.79097098,
      "num_input_tokens_seen": 49469990,
      "step": 2304,
      "time_per_iteration": 2.789461851119995
    },
    {
      "auxiliary_loss_clip": 0.0113691,
      "auxiliary_loss_mlp": 0.01039556,
      "balance_loss_clip": 1.05953145,
      "balance_loss_mlp": 1.02896452,
      "epoch": 0.2771598629231047,
      "flos": 22929897219840.0,
      "grad_norm": 2.6813957009183373,
      "language_loss": 0.81216383,
      "learning_rate": 3.3928489416025495e-06,
      "loss": 0.83392847,
      "num_input_tokens_seen": 49490835,
      "step": 2305,
      "time_per_iteration": 2.8582403659820557
    },
    {
      "auxiliary_loss_clip": 0.01179269,
      "auxiliary_loss_mlp": 0.01037807,
      "balance_loss_clip": 1.06124115,
      "balance_loss_mlp": 1.0269649,
      "epoch": 0.27728010581374374,
      "flos": 18369457741440.0,
      "grad_norm": 3.0947285070087283,
      "language_loss": 0.79370135,
      "learning_rate": 3.392289822470638e-06,
      "loss": 0.81587213,
      "num_input_tokens_seen": 49508815,
      "step": 2306,
      "time_per_iteration": 2.714648962020874
    },
    {
      "auxiliary_loss_clip": 0.01162652,
      "auxiliary_loss_mlp": 0.01036305,
      "balance_loss_clip": 1.05396402,
      "balance_loss_mlp": 1.02621984,
      "epoch": 0.27740034870438285,
      "flos": 19427637432960.0,
      "grad_norm": 2.4909341599351422,
      "language_loss": 0.76189303,
      "learning_rate": 3.3917304921294674e-06,
      "loss": 0.78388262,
      "num_input_tokens_seen": 49526980,
      "step": 2307,
      "time_per_iteration": 2.7973992824554443
    },
    {
      "auxiliary_loss_clip": 0.01185943,
      "auxiliary_loss_mlp": 0.00764645,
      "balance_loss_clip": 1.05998945,
      "balance_loss_mlp": 1.00057435,
      "epoch": 0.27752059159502196,
      "flos": 21614776565760.0,
      "grad_norm": 2.016097658006954,
      "language_loss": 0.80940497,
      "learning_rate": 3.3911709506638876e-06,
      "loss": 0.82891083,
      "num_input_tokens_seen": 49546290,
      "step": 2308,
      "time_per_iteration": 2.739560127258301
    },
    {
      "auxiliary_loss_clip": 0.01200847,
      "auxiliary_loss_mlp": 0.01046441,
      "balance_loss_clip": 1.06370842,
      "balance_loss_mlp": 1.03469849,
      "epoch": 0.277640834485661,
      "flos": 26608011016320.0,
      "grad_norm": 2.3035372028793555,
      "language_loss": 0.81432861,
      "learning_rate": 3.390611198158781e-06,
      "loss": 0.83680147,
      "num_input_tokens_seen": 49564165,
      "step": 2309,
      "time_per_iteration": 2.647118091583252
    },
    {
      "auxiliary_loss_clip": 0.01206942,
      "auxiliary_loss_mlp": 0.01041467,
      "balance_loss_clip": 1.06689286,
      "balance_loss_mlp": 1.03117907,
      "epoch": 0.2777610773763001,
      "flos": 19492814661120.0,
      "grad_norm": 2.512589923078179,
      "language_loss": 0.90282661,
      "learning_rate": 3.3900512346990612e-06,
      "loss": 0.92531073,
      "num_input_tokens_seen": 49580155,
      "step": 2310,
      "time_per_iteration": 2.6176388263702393
    },
    {
      "auxiliary_loss_clip": 0.0115668,
      "auxiliary_loss_mlp": 0.0103569,
      "balance_loss_clip": 1.05947554,
      "balance_loss_mlp": 1.02496696,
      "epoch": 0.27788132026693924,
      "flos": 38290650001920.0,
      "grad_norm": 1.7921695227737868,
      "language_loss": 0.660662,
      "learning_rate": 3.389491060369674e-06,
      "loss": 0.68258572,
      "num_input_tokens_seen": 49605830,
      "step": 2311,
      "time_per_iteration": 2.964177131652832
    },
    {
      "auxiliary_loss_clip": 0.01163617,
      "auxiliary_loss_mlp": 0.0103863,
      "balance_loss_clip": 1.06035352,
      "balance_loss_mlp": 1.02944541,
      "epoch": 0.2780015631575783,
      "flos": 22382546797440.0,
      "grad_norm": 2.3145734477697815,
      "language_loss": 0.89722794,
      "learning_rate": 3.388930675255598e-06,
      "loss": 0.91925037,
      "num_input_tokens_seen": 49625680,
      "step": 2312,
      "time_per_iteration": 2.7402420043945312
    },
    {
      "auxiliary_loss_clip": 0.01160751,
      "auxiliary_loss_mlp": 0.01039629,
      "balance_loss_clip": 1.05640268,
      "balance_loss_mlp": 1.02839947,
      "epoch": 0.2781218060482174,
      "flos": 12203200840320.0,
      "grad_norm": 9.800874720691732,
      "language_loss": 0.79484546,
      "learning_rate": 3.388370079441843e-06,
      "loss": 0.81684929,
      "num_input_tokens_seen": 49641195,
      "step": 2313,
      "time_per_iteration": 2.740818977355957
    },
    {
      "auxiliary_loss_clip": 0.01149432,
      "auxiliary_loss_mlp": 0.01044724,
      "balance_loss_clip": 1.06082058,
      "balance_loss_mlp": 1.03394699,
      "epoch": 0.2782420489388565,
      "flos": 18107632529280.0,
      "grad_norm": 4.758173854506446,
      "language_loss": 0.93044323,
      "learning_rate": 3.3878092730134505e-06,
      "loss": 0.95238477,
      "num_input_tokens_seen": 49659180,
      "step": 2314,
      "time_per_iteration": 2.768597364425659
    },
    {
      "auxiliary_loss_clip": 0.01189489,
      "auxiliary_loss_mlp": 0.00764241,
      "balance_loss_clip": 1.06310582,
      "balance_loss_mlp": 1.00054598,
      "epoch": 0.27836229182949557,
      "flos": 18514752255360.0,
      "grad_norm": 2.7097717039810574,
      "language_loss": 0.81065714,
      "learning_rate": 3.3872482560554947e-06,
      "loss": 0.83019447,
      "num_input_tokens_seen": 49677955,
      "step": 2315,
      "time_per_iteration": 2.6806745529174805
    },
    {
      "auxiliary_loss_clip": 0.0105741,
      "auxiliary_loss_mlp": 0.01001975,
      "balance_loss_clip": 1.0204761,
      "balance_loss_mlp": 0.99987715,
      "epoch": 0.2784825347201347,
      "flos": 67079230940160.0,
      "grad_norm": 0.7945360489915108,
      "language_loss": 0.56921661,
      "learning_rate": 3.386687028653082e-06,
      "loss": 0.58981049,
      "num_input_tokens_seen": 49740800,
      "step": 2316,
      "time_per_iteration": 3.427621841430664
    },
    {
      "auxiliary_loss_clip": 0.01179475,
      "auxiliary_loss_mlp": 0.01044244,
      "balance_loss_clip": 1.06276298,
      "balance_loss_mlp": 1.03329504,
      "epoch": 0.2786027776107738,
      "flos": 22631119891200.0,
      "grad_norm": 1.9105013188203708,
      "language_loss": 0.85366058,
      "learning_rate": 3.386125590891349e-06,
      "loss": 0.87589782,
      "num_input_tokens_seen": 49757675,
      "step": 2317,
      "time_per_iteration": 2.826881170272827
    },
    {
      "auxiliary_loss_clip": 0.01143361,
      "auxiliary_loss_mlp": 0.01040768,
      "balance_loss_clip": 1.05462503,
      "balance_loss_mlp": 1.03050959,
      "epoch": 0.27872302050141284,
      "flos": 15778826156160.0,
      "grad_norm": 5.14800980255508,
      "language_loss": 0.83333278,
      "learning_rate": 3.3855639428554657e-06,
      "loss": 0.85517406,
      "num_input_tokens_seen": 49775205,
      "step": 2318,
      "time_per_iteration": 2.7620906829833984
    },
    {
      "auxiliary_loss_clip": 0.01126678,
      "auxiliary_loss_mlp": 0.01040474,
      "balance_loss_clip": 1.05554533,
      "balance_loss_mlp": 1.02986383,
      "epoch": 0.27884326339205195,
      "flos": 22126970551680.0,
      "grad_norm": 3.1766728696467523,
      "language_loss": 0.80683184,
      "learning_rate": 3.385002084630635e-06,
      "loss": 0.82850331,
      "num_input_tokens_seen": 49794175,
      "step": 2319,
      "time_per_iteration": 2.839728593826294
    },
    {
      "auxiliary_loss_clip": 0.01172785,
      "auxiliary_loss_mlp": 0.01033711,
      "balance_loss_clip": 1.06236959,
      "balance_loss_mlp": 1.02325654,
      "epoch": 0.278963506282691,
      "flos": 20558715776640.0,
      "grad_norm": 2.4311962278737655,
      "language_loss": 0.85084593,
      "learning_rate": 3.384440016302088e-06,
      "loss": 0.87291086,
      "num_input_tokens_seen": 49812850,
      "step": 2320,
      "time_per_iteration": 2.737504720687866
    },
    {
      "auxiliary_loss_clip": 0.0112721,
      "auxiliary_loss_mlp": 0.01037637,
      "balance_loss_clip": 1.04925013,
      "balance_loss_mlp": 1.02668166,
      "epoch": 0.2790837491733301,
      "flos": 21942928241280.0,
      "grad_norm": 2.5760098132841702,
      "language_loss": 0.61598563,
      "learning_rate": 3.3838777379550923e-06,
      "loss": 0.63763416,
      "num_input_tokens_seen": 49832295,
      "step": 2321,
      "time_per_iteration": 3.694931745529175
    },
    {
      "auxiliary_loss_clip": 0.01141159,
      "auxiliary_loss_mlp": 0.01043722,
      "balance_loss_clip": 1.04963493,
      "balance_loss_mlp": 1.03395307,
      "epoch": 0.27920399206396923,
      "flos": 26286790665600.0,
      "grad_norm": 2.378928352586079,
      "language_loss": 0.78953975,
      "learning_rate": 3.383315249674944e-06,
      "loss": 0.81138861,
      "num_input_tokens_seen": 49850860,
      "step": 2322,
      "time_per_iteration": 3.8887076377868652
    },
    {
      "auxiliary_loss_clip": 0.01140508,
      "auxiliary_loss_mlp": 0.01033955,
      "balance_loss_clip": 1.05745482,
      "balance_loss_mlp": 1.02354145,
      "epoch": 0.2793242349546083,
      "flos": 25400981364480.0,
      "grad_norm": 2.3325308651696175,
      "language_loss": 0.86355293,
      "learning_rate": 3.3827525515469715e-06,
      "loss": 0.8852976,
      "num_input_tokens_seen": 49865765,
      "step": 2323,
      "time_per_iteration": 2.775977373123169
    },
    {
      "auxiliary_loss_clip": 0.01132608,
      "auxiliary_loss_mlp": 0.01037807,
      "balance_loss_clip": 1.05489206,
      "balance_loss_mlp": 1.02666104,
      "epoch": 0.2794444778452474,
      "flos": 20850346298880.0,
      "grad_norm": 2.644699170116803,
      "language_loss": 0.71475607,
      "learning_rate": 3.3821896436565367e-06,
      "loss": 0.73646021,
      "num_input_tokens_seen": 49885425,
      "step": 2324,
      "time_per_iteration": 2.8516042232513428
    },
    {
      "auxiliary_loss_clip": 0.01197797,
      "auxiliary_loss_mlp": 0.01034536,
      "balance_loss_clip": 1.06494331,
      "balance_loss_mlp": 1.02477252,
      "epoch": 0.2795647207358865,
      "flos": 21576244250880.0,
      "grad_norm": 2.068130483396194,
      "language_loss": 0.70393884,
      "learning_rate": 3.381626526089032e-06,
      "loss": 0.72626215,
      "num_input_tokens_seen": 49904990,
      "step": 2325,
      "time_per_iteration": 3.6731081008911133
    },
    {
      "auxiliary_loss_clip": 0.01199444,
      "auxiliary_loss_mlp": 0.01041279,
      "balance_loss_clip": 1.06430936,
      "balance_loss_mlp": 1.03044283,
      "epoch": 0.27968496362652556,
      "flos": 21471744608640.0,
      "grad_norm": 3.9346741956292117,
      "language_loss": 0.7948463,
      "learning_rate": 3.3810631989298815e-06,
      "loss": 0.81725353,
      "num_input_tokens_seen": 49924600,
      "step": 2326,
      "time_per_iteration": 3.769294261932373
    },
    {
      "auxiliary_loss_clip": 0.01207407,
      "auxiliary_loss_mlp": 0.01040584,
      "balance_loss_clip": 1.06709051,
      "balance_loss_mlp": 1.02813864,
      "epoch": 0.2798052065171647,
      "flos": 23258695340160.0,
      "grad_norm": 2.322840407727973,
      "language_loss": 0.84454179,
      "learning_rate": 3.3804996622645423e-06,
      "loss": 0.86702168,
      "num_input_tokens_seen": 49942600,
      "step": 2327,
      "time_per_iteration": 2.7106575965881348
    },
    {
      "auxiliary_loss_clip": 0.01158289,
      "auxiliary_loss_mlp": 0.01041287,
      "balance_loss_clip": 1.05973458,
      "balance_loss_mlp": 1.03096306,
      "epoch": 0.2799254494078038,
      "flos": 21539328048000.0,
      "grad_norm": 6.371466379929556,
      "language_loss": 0.89578205,
      "learning_rate": 3.3799359161785015e-06,
      "loss": 0.91777784,
      "num_input_tokens_seen": 49962250,
      "step": 2328,
      "time_per_iteration": 2.7599103450775146
    },
    {
      "auxiliary_loss_clip": 0.01156492,
      "auxiliary_loss_mlp": 0.00764828,
      "balance_loss_clip": 1.05449581,
      "balance_loss_mlp": 1.00061369,
      "epoch": 0.28004569229844284,
      "flos": 26393912000640.0,
      "grad_norm": 1.798590805798552,
      "language_loss": 0.86171615,
      "learning_rate": 3.3793719607572798e-06,
      "loss": 0.88092935,
      "num_input_tokens_seen": 49983215,
      "step": 2329,
      "time_per_iteration": 2.8186702728271484
    },
    {
      "auxiliary_loss_clip": 0.01115063,
      "auxiliary_loss_mlp": 0.01042682,
      "balance_loss_clip": 1.05291414,
      "balance_loss_mlp": 1.03244174,
      "epoch": 0.28016593518908195,
      "flos": 33547676584320.0,
      "grad_norm": 10.975679015684927,
      "language_loss": 0.77616239,
      "learning_rate": 3.378807796086428e-06,
      "loss": 0.79773986,
      "num_input_tokens_seen": 50006075,
      "step": 2330,
      "time_per_iteration": 2.9353551864624023
    },
    {
      "auxiliary_loss_clip": 0.01186815,
      "auxiliary_loss_mlp": 0.01032651,
      "balance_loss_clip": 1.06184244,
      "balance_loss_mlp": 1.02154636,
      "epoch": 0.28028617807972106,
      "flos": 15340823712000.0,
      "grad_norm": 2.628101796302286,
      "language_loss": 0.77272445,
      "learning_rate": 3.37824342225153e-06,
      "loss": 0.79491913,
      "num_input_tokens_seen": 50022495,
      "step": 2331,
      "time_per_iteration": 2.8049094676971436
    },
    {
      "auxiliary_loss_clip": 0.01204729,
      "auxiliary_loss_mlp": 0.01038464,
      "balance_loss_clip": 1.06677723,
      "balance_loss_mlp": 1.02790833,
      "epoch": 0.2804064209703601,
      "flos": 25520277409920.0,
      "grad_norm": 2.553109062749407,
      "language_loss": 0.77803212,
      "learning_rate": 3.3776788393382006e-06,
      "loss": 0.80046409,
      "num_input_tokens_seen": 50041975,
      "step": 2332,
      "time_per_iteration": 2.725667715072632
    },
    {
      "auxiliary_loss_clip": 0.01203982,
      "auxiliary_loss_mlp": 0.01040931,
      "balance_loss_clip": 1.06548905,
      "balance_loss_mlp": 1.03070903,
      "epoch": 0.2805266638609992,
      "flos": 29351766280320.0,
      "grad_norm": 2.843308488847672,
      "language_loss": 0.77065122,
      "learning_rate": 3.3771140474320872e-06,
      "loss": 0.79310036,
      "num_input_tokens_seen": 50061925,
      "step": 2333,
      "time_per_iteration": 2.7624354362487793
    },
    {
      "auxiliary_loss_clip": 0.01166104,
      "auxiliary_loss_mlp": 0.0104289,
      "balance_loss_clip": 1.06071413,
      "balance_loss_mlp": 1.03185129,
      "epoch": 0.28064690675163834,
      "flos": 21463735875840.0,
      "grad_norm": 2.23317150618117,
      "language_loss": 0.79496241,
      "learning_rate": 3.3765490466188664e-06,
      "loss": 0.81705236,
      "num_input_tokens_seen": 50079325,
      "step": 2334,
      "time_per_iteration": 2.749140739440918
    },
    {
      "auxiliary_loss_clip": 0.01145916,
      "auxiliary_loss_mlp": 0.01031425,
      "balance_loss_clip": 1.05672026,
      "balance_loss_mlp": 1.02094615,
      "epoch": 0.2807671496422774,
      "flos": 20995640812800.0,
      "grad_norm": 2.6814912305802254,
      "language_loss": 0.74159193,
      "learning_rate": 3.3759838369842508e-06,
      "loss": 0.76336533,
      "num_input_tokens_seen": 50097400,
      "step": 2335,
      "time_per_iteration": 2.8409605026245117
    },
    {
      "auxiliary_loss_clip": 0.01124943,
      "auxiliary_loss_mlp": 0.0103438,
      "balance_loss_clip": 1.05289054,
      "balance_loss_mlp": 1.02357411,
      "epoch": 0.2808873925329165,
      "flos": 21506577822720.0,
      "grad_norm": 2.3069572161326595,
      "language_loss": 0.73319376,
      "learning_rate": 3.375418418613981e-06,
      "loss": 0.75478697,
      "num_input_tokens_seen": 50116425,
      "step": 2336,
      "time_per_iteration": 2.8556759357452393
    },
    {
      "auxiliary_loss_clip": 0.01219357,
      "auxiliary_loss_mlp": 0.01039127,
      "balance_loss_clip": 1.06659508,
      "balance_loss_mlp": 1.02875006,
      "epoch": 0.28100763542355556,
      "flos": 16070815814400.0,
      "grad_norm": 6.017842430248333,
      "language_loss": 0.83965421,
      "learning_rate": 3.374852791593831e-06,
      "loss": 0.86223906,
      "num_input_tokens_seen": 50132625,
      "step": 2337,
      "time_per_iteration": 2.6484858989715576
    },
    {
      "auxiliary_loss_clip": 0.01152925,
      "auxiliary_loss_mlp": 0.01049912,
      "balance_loss_clip": 1.05835879,
      "balance_loss_mlp": 1.03836668,
      "epoch": 0.28112787831419467,
      "flos": 19062605468160.0,
      "grad_norm": 2.6532901372522164,
      "language_loss": 0.53869945,
      "learning_rate": 3.374286956009605e-06,
      "loss": 0.56072783,
      "num_input_tokens_seen": 50151190,
      "step": 2338,
      "time_per_iteration": 2.8542425632476807
    },
    {
      "auxiliary_loss_clip": 0.01204784,
      "auxiliary_loss_mlp": 0.01036778,
      "balance_loss_clip": 1.06666327,
      "balance_loss_mlp": 1.0263772,
      "epoch": 0.2812481212048338,
      "flos": 12823629482880.0,
      "grad_norm": 2.4268106952963087,
      "language_loss": 0.75189149,
      "learning_rate": 3.3737209119471405e-06,
      "loss": 0.77430707,
      "num_input_tokens_seen": 50167700,
      "step": 2339,
      "time_per_iteration": 2.6444811820983887
    },
    {
      "auxiliary_loss_clip": 0.01181524,
      "auxiliary_loss_mlp": 0.00765813,
      "balance_loss_clip": 1.06117904,
      "balance_loss_mlp": 1.00063658,
      "epoch": 0.28136836409547283,
      "flos": 15633064765440.0,
      "grad_norm": 3.828314698926483,
      "language_loss": 0.63662535,
      "learning_rate": 3.373154659492306e-06,
      "loss": 0.65609872,
      "num_input_tokens_seen": 50185840,
      "step": 2340,
      "time_per_iteration": 2.7731270790100098
    },
    {
      "auxiliary_loss_clip": 0.01179899,
      "auxiliary_loss_mlp": 0.01039605,
      "balance_loss_clip": 1.06041968,
      "balance_loss_mlp": 1.03009796,
      "epoch": 0.28148860698611194,
      "flos": 19933726106880.0,
      "grad_norm": 2.473587111787679,
      "language_loss": 0.85164857,
      "learning_rate": 3.3725881987310016e-06,
      "loss": 0.87384361,
      "num_input_tokens_seen": 50203375,
      "step": 2341,
      "time_per_iteration": 2.742778778076172
    },
    {
      "auxiliary_loss_clip": 0.01120616,
      "auxiliary_loss_mlp": 0.0103429,
      "balance_loss_clip": 1.05080283,
      "balance_loss_mlp": 1.02471769,
      "epoch": 0.28160884987675106,
      "flos": 17457219008640.0,
      "grad_norm": 4.674385190471725,
      "language_loss": 0.87559181,
      "learning_rate": 3.372021529749159e-06,
      "loss": 0.89714086,
      "num_input_tokens_seen": 50222435,
      "step": 2342,
      "time_per_iteration": 2.773362159729004
    },
    {
      "auxiliary_loss_clip": 0.01169127,
      "auxiliary_loss_mlp": 0.01041286,
      "balance_loss_clip": 1.05953407,
      "balance_loss_mlp": 1.03077173,
      "epoch": 0.2817290927673901,
      "flos": 16834743290880.0,
      "grad_norm": 2.2014422950148167,
      "language_loss": 0.92287815,
      "learning_rate": 3.3714546526327405e-06,
      "loss": 0.94498229,
      "num_input_tokens_seen": 50240435,
      "step": 2343,
      "time_per_iteration": 2.780642509460449
    },
    {
      "auxiliary_loss_clip": 0.01182744,
      "auxiliary_loss_mlp": 0.01037706,
      "balance_loss_clip": 1.06009614,
      "balance_loss_mlp": 1.02678084,
      "epoch": 0.2818493356580292,
      "flos": 15414081500160.0,
      "grad_norm": 2.4760936957589346,
      "language_loss": 0.87695169,
      "learning_rate": 3.3708875674677423e-06,
      "loss": 0.89915621,
      "num_input_tokens_seen": 50258410,
      "step": 2344,
      "time_per_iteration": 2.7950093746185303
    },
    {
      "auxiliary_loss_clip": 0.01205568,
      "auxiliary_loss_mlp": 0.01041365,
      "balance_loss_clip": 1.06719708,
      "balance_loss_mlp": 1.02914047,
      "epoch": 0.28196957854866833,
      "flos": 20412451595520.0,
      "grad_norm": 7.410337532196014,
      "language_loss": 0.83689392,
      "learning_rate": 3.37032027434019e-06,
      "loss": 0.85936326,
      "num_input_tokens_seen": 50277930,
      "step": 2345,
      "time_per_iteration": 2.736111879348755
    },
    {
      "auxiliary_loss_clip": 0.01163713,
      "auxiliary_loss_mlp": 0.01049012,
      "balance_loss_clip": 1.06004465,
      "balance_loss_mlp": 1.03654838,
      "epoch": 0.2820898214393074,
      "flos": 19973120348160.0,
      "grad_norm": 1.9907207111227183,
      "language_loss": 0.82712889,
      "learning_rate": 3.369752773336141e-06,
      "loss": 0.84925616,
      "num_input_tokens_seen": 50297410,
      "step": 2346,
      "time_per_iteration": 2.7509829998016357
    },
    {
      "auxiliary_loss_clip": 0.01189472,
      "auxiliary_loss_mlp": 0.01043745,
      "balance_loss_clip": 1.06439376,
      "balance_loss_mlp": 1.03290915,
      "epoch": 0.2822100643299465,
      "flos": 22528308188160.0,
      "grad_norm": 2.1511153534784095,
      "language_loss": 0.78010929,
      "learning_rate": 3.3691850645416864e-06,
      "loss": 0.80244148,
      "num_input_tokens_seen": 50317120,
      "step": 2347,
      "time_per_iteration": 3.678972005844116
    },
    {
      "auxiliary_loss_clip": 0.01199854,
      "auxiliary_loss_mlp": 0.00764758,
      "balance_loss_clip": 1.06279731,
      "balance_loss_mlp": 1.00067973,
      "epoch": 0.2823303072205856,
      "flos": 11546682007680.0,
      "grad_norm": 3.177355855738976,
      "language_loss": 0.83405173,
      "learning_rate": 3.368617148042945e-06,
      "loss": 0.8536979,
      "num_input_tokens_seen": 50334790,
      "step": 2348,
      "time_per_iteration": 3.684384346008301
    },
    {
      "auxiliary_loss_clip": 0.01179689,
      "auxiliary_loss_mlp": 0.01042617,
      "balance_loss_clip": 1.06196678,
      "balance_loss_mlp": 1.0320369,
      "epoch": 0.28245055011122466,
      "flos": 18259894281600.0,
      "grad_norm": 6.045214083792943,
      "language_loss": 0.84479392,
      "learning_rate": 3.368049023926071e-06,
      "loss": 0.86701691,
      "num_input_tokens_seen": 50353785,
      "step": 2349,
      "time_per_iteration": 2.640165090560913
    },
    {
      "auxiliary_loss_clip": 0.01206017,
      "auxiliary_loss_mlp": 0.010369,
      "balance_loss_clip": 1.06827712,
      "balance_loss_mlp": 1.02650499,
      "epoch": 0.2825707930018638,
      "flos": 24608110504320.0,
      "grad_norm": 1.7520980799483967,
      "language_loss": 0.83946824,
      "learning_rate": 3.3674806922772476e-06,
      "loss": 0.86189735,
      "num_input_tokens_seen": 50374670,
      "step": 2350,
      "time_per_iteration": 3.7082626819610596
    },
    {
      "auxiliary_loss_clip": 0.01207181,
      "auxiliary_loss_mlp": 0.01043201,
      "balance_loss_clip": 1.06419277,
      "balance_loss_mlp": 1.03241253,
      "epoch": 0.28269103589250283,
      "flos": 25226994862080.0,
      "grad_norm": 2.289742895008636,
      "language_loss": 0.74991727,
      "learning_rate": 3.3669121531826904e-06,
      "loss": 0.77242112,
      "num_input_tokens_seen": 50395650,
      "step": 2351,
      "time_per_iteration": 2.7436556816101074
    },
    {
      "auxiliary_loss_clip": 0.01131987,
      "auxiliary_loss_mlp": 0.01035971,
      "balance_loss_clip": 1.05612278,
      "balance_loss_mlp": 1.02654719,
      "epoch": 0.28281127878314194,
      "flos": 19281552819840.0,
      "grad_norm": 5.167373646044148,
      "language_loss": 0.83850592,
      "learning_rate": 3.366343406728647e-06,
      "loss": 0.8601855,
      "num_input_tokens_seen": 50415100,
      "step": 2352,
      "time_per_iteration": 3.7686827182769775
    },
    {
      "auxiliary_loss_clip": 0.01199717,
      "auxiliary_loss_mlp": 0.01034979,
      "balance_loss_clip": 1.06133747,
      "balance_loss_mlp": 1.02458429,
      "epoch": 0.28293152167378105,
      "flos": 23878405710720.0,
      "grad_norm": 2.394590185967648,
      "language_loss": 0.6919291,
      "learning_rate": 3.3657744530013946e-06,
      "loss": 0.71427608,
      "num_input_tokens_seen": 50434335,
      "step": 2353,
      "time_per_iteration": 2.7310309410095215
    },
    {
      "auxiliary_loss_clip": 0.01128384,
      "auxiliary_loss_mlp": 0.01052916,
      "balance_loss_clip": 1.05475831,
      "balance_loss_mlp": 1.04235375,
      "epoch": 0.2830517645644201,
      "flos": 43866965928960.0,
      "grad_norm": 3.426735875354676,
      "language_loss": 0.71520603,
      "learning_rate": 3.3652052920872437e-06,
      "loss": 0.737019,
      "num_input_tokens_seen": 50457200,
      "step": 2354,
      "time_per_iteration": 2.999281644821167
    },
    {
      "auxiliary_loss_clip": 0.01150955,
      "auxiliary_loss_mlp": 0.0103773,
      "balance_loss_clip": 1.05925941,
      "balance_loss_mlp": 1.02772212,
      "epoch": 0.2831720074550592,
      "flos": 26651750803200.0,
      "grad_norm": 6.648734725847468,
      "language_loss": 0.85725683,
      "learning_rate": 3.3646359240725355e-06,
      "loss": 0.87914371,
      "num_input_tokens_seen": 50476390,
      "step": 2355,
      "time_per_iteration": 2.823272943496704
    },
    {
      "auxiliary_loss_clip": 0.01174044,
      "auxiliary_loss_mlp": 0.01044048,
      "balance_loss_clip": 1.06168294,
      "balance_loss_mlp": 1.0336889,
      "epoch": 0.2832922503456983,
      "flos": 31029979564800.0,
      "grad_norm": 2.8484539646635008,
      "language_loss": 0.68002915,
      "learning_rate": 3.364066349043643e-06,
      "loss": 0.70221007,
      "num_input_tokens_seen": 50497595,
      "step": 2356,
      "time_per_iteration": 2.8218793869018555
    },
    {
      "auxiliary_loss_clip": 0.01180471,
      "auxiliary_loss_mlp": 0.01038713,
      "balance_loss_clip": 1.06134796,
      "balance_loss_mlp": 1.02965868,
      "epoch": 0.2834124932363374,
      "flos": 20405699838720.0,
      "grad_norm": 1.8763573058014986,
      "language_loss": 0.82145876,
      "learning_rate": 3.363496567086969e-06,
      "loss": 0.84365058,
      "num_input_tokens_seen": 50514690,
      "step": 2357,
      "time_per_iteration": 2.6742608547210693
    },
    {
      "auxiliary_loss_clip": 0.01193018,
      "auxiliary_loss_mlp": 0.01041088,
      "balance_loss_clip": 1.06376421,
      "balance_loss_mlp": 1.03125882,
      "epoch": 0.2835327361269765,
      "flos": 39384848056320.0,
      "grad_norm": 2.381263788651342,
      "language_loss": 0.7531364,
      "learning_rate": 3.3629265782889506e-06,
      "loss": 0.77547753,
      "num_input_tokens_seen": 50536515,
      "step": 2358,
      "time_per_iteration": 2.9051530361175537
    },
    {
      "auxiliary_loss_clip": 0.01167026,
      "auxiliary_loss_mlp": 0.01045067,
      "balance_loss_clip": 1.05652952,
      "balance_loss_mlp": 1.03409338,
      "epoch": 0.2836529790176156,
      "flos": 30261598801920.0,
      "grad_norm": 5.170748684045738,
      "language_loss": 0.7227211,
      "learning_rate": 3.362356382736054e-06,
      "loss": 0.74484205,
      "num_input_tokens_seen": 50557120,
      "step": 2359,
      "time_per_iteration": 2.888170003890991
    },
    {
      "auxiliary_loss_clip": 0.01186806,
      "auxiliary_loss_mlp": 0.01036451,
      "balance_loss_clip": 1.0609436,
      "balance_loss_mlp": 1.02685475,
      "epoch": 0.28377322190825466,
      "flos": 12677796264960.0,
      "grad_norm": 2.7760986420192606,
      "language_loss": 0.91086364,
      "learning_rate": 3.361785980514777e-06,
      "loss": 0.93309629,
      "num_input_tokens_seen": 50573320,
      "step": 2360,
      "time_per_iteration": 2.712092638015747
    },
    {
      "auxiliary_loss_clip": 0.01163826,
      "auxiliary_loss_mlp": 0.01036449,
      "balance_loss_clip": 1.05788279,
      "balance_loss_mlp": 1.02548814,
      "epoch": 0.28389346479889377,
      "flos": 18296666830080.0,
      "grad_norm": 2.3542979912573507,
      "language_loss": 0.76409978,
      "learning_rate": 3.361215371711649e-06,
      "loss": 0.78610253,
      "num_input_tokens_seen": 50592415,
      "step": 2361,
      "time_per_iteration": 2.6958401203155518
    },
    {
      "auxiliary_loss_clip": 0.01185052,
      "auxiliary_loss_mlp": 0.01034891,
      "balance_loss_clip": 1.06135094,
      "balance_loss_mlp": 1.02477646,
      "epoch": 0.2840137076895329,
      "flos": 20406992728320.0,
      "grad_norm": 4.360268497983524,
      "language_loss": 0.83414996,
      "learning_rate": 3.3606445564132326e-06,
      "loss": 0.85634935,
      "num_input_tokens_seen": 50609710,
      "step": 2362,
      "time_per_iteration": 2.843278408050537
    },
    {
      "auxiliary_loss_clip": 0.01192003,
      "auxiliary_loss_mlp": 0.01038169,
      "balance_loss_clip": 1.06390667,
      "balance_loss_mlp": 1.0281558,
      "epoch": 0.28413395058017193,
      "flos": 20048030161920.0,
      "grad_norm": 2.358911378991741,
      "language_loss": 0.81854618,
      "learning_rate": 3.360073534706118e-06,
      "loss": 0.84084791,
      "num_input_tokens_seen": 50626865,
      "step": 2363,
      "time_per_iteration": 2.742037296295166
    },
    {
      "auxiliary_loss_clip": 0.01187915,
      "auxiliary_loss_mlp": 0.01038516,
      "balance_loss_clip": 1.06570494,
      "balance_loss_mlp": 1.02773404,
      "epoch": 0.28425419347081105,
      "flos": 37663613256960.0,
      "grad_norm": 3.010304099894135,
      "language_loss": 0.76445079,
      "learning_rate": 3.35950230667693e-06,
      "loss": 0.78671509,
      "num_input_tokens_seen": 50648560,
      "step": 2364,
      "time_per_iteration": 2.9320950508117676
    },
    {
      "auxiliary_loss_clip": 0.01170554,
      "auxiliary_loss_mlp": 0.01037409,
      "balance_loss_clip": 1.05671048,
      "balance_loss_mlp": 1.02811098,
      "epoch": 0.28437443636145016,
      "flos": 13845072539520.0,
      "grad_norm": 2.4161293354297575,
      "language_loss": 0.85918516,
      "learning_rate": 3.358930872412323e-06,
      "loss": 0.88126481,
      "num_input_tokens_seen": 50665725,
      "step": 2365,
      "time_per_iteration": 2.7272987365722656
    },
    {
      "auxiliary_loss_clip": 0.01177084,
      "auxiliary_loss_mlp": 0.00764169,
      "balance_loss_clip": 1.06381524,
      "balance_loss_mlp": 1.00065637,
      "epoch": 0.2844946792520892,
      "flos": 22747794243840.0,
      "grad_norm": 6.372438897665931,
      "language_loss": 0.81262964,
      "learning_rate": 3.3583592319989825e-06,
      "loss": 0.83204222,
      "num_input_tokens_seen": 50685095,
      "step": 2366,
      "time_per_iteration": 2.805391788482666
    },
    {
      "auxiliary_loss_clip": 0.01200007,
      "auxiliary_loss_mlp": 0.01048269,
      "balance_loss_clip": 1.06623554,
      "balance_loss_mlp": 1.03658104,
      "epoch": 0.2846149221427283,
      "flos": 32415987709440.0,
      "grad_norm": 3.500352227533982,
      "language_loss": 0.69614685,
      "learning_rate": 3.357787385523627e-06,
      "loss": 0.71862966,
      "num_input_tokens_seen": 50706500,
      "step": 2367,
      "time_per_iteration": 2.8605129718780518
    },
    {
      "auxiliary_loss_clip": 0.01203627,
      "auxiliary_loss_mlp": 0.01042074,
      "balance_loss_clip": 1.06595433,
      "balance_loss_mlp": 1.03117263,
      "epoch": 0.2847351650333674,
      "flos": 28475976873600.0,
      "grad_norm": 2.010411651641877,
      "language_loss": 0.83242238,
      "learning_rate": 3.3572153330730048e-06,
      "loss": 0.85487938,
      "num_input_tokens_seen": 50727595,
      "step": 2368,
      "time_per_iteration": 2.72371244430542
    },
    {
      "auxiliary_loss_clip": 0.010965,
      "auxiliary_loss_mlp": 0.01005338,
      "balance_loss_clip": 1.02460194,
      "balance_loss_mlp": 1.00347805,
      "epoch": 0.2848554079240065,
      "flos": 55753399704960.0,
      "grad_norm": 0.9232455046527505,
      "language_loss": 0.64678061,
      "learning_rate": 3.3566430747338956e-06,
      "loss": 0.667799,
      "num_input_tokens_seen": 50782800,
      "step": 2369,
      "time_per_iteration": 3.112905979156494
    },
    {
      "auxiliary_loss_clip": 0.01177839,
      "auxiliary_loss_mlp": 0.01026344,
      "balance_loss_clip": 1.05999947,
      "balance_loss_mlp": 1.01610947,
      "epoch": 0.2849756508146456,
      "flos": 11836875985920.0,
      "grad_norm": 2.1906508441018686,
      "language_loss": 0.86796105,
      "learning_rate": 3.35607061059311e-06,
      "loss": 0.89000285,
      "num_input_tokens_seen": 50797730,
      "step": 2370,
      "time_per_iteration": 2.714517116546631
    },
    {
      "auxiliary_loss_clip": 0.01185748,
      "auxiliary_loss_mlp": 0.01041768,
      "balance_loss_clip": 1.06597805,
      "balance_loss_mlp": 1.03171277,
      "epoch": 0.28509589370528465,
      "flos": 25155209531520.0,
      "grad_norm": 4.5844589028468015,
      "language_loss": 0.74954641,
      "learning_rate": 3.3554979407374917e-06,
      "loss": 0.77182162,
      "num_input_tokens_seen": 50819840,
      "step": 2371,
      "time_per_iteration": 2.8106040954589844
    },
    {
      "auxiliary_loss_clip": 0.0112506,
      "auxiliary_loss_mlp": 0.01040696,
      "balance_loss_clip": 1.05223727,
      "balance_loss_mlp": 1.03043842,
      "epoch": 0.28521613659592376,
      "flos": 19974808287360.0,
      "grad_norm": 2.1872596373919992,
      "language_loss": 0.73992938,
      "learning_rate": 3.3549250652539134e-06,
      "loss": 0.7615869,
      "num_input_tokens_seen": 50838935,
      "step": 2372,
      "time_per_iteration": 2.8572394847869873
    },
    {
      "auxiliary_loss_clip": 0.01200569,
      "auxiliary_loss_mlp": 0.01043326,
      "balance_loss_clip": 1.0636754,
      "balance_loss_mlp": 1.03200126,
      "epoch": 0.2853363794865629,
      "flos": 23367971491200.0,
      "grad_norm": 2.997741044622496,
      "language_loss": 0.817173,
      "learning_rate": 3.3543519842292794e-06,
      "loss": 0.83961189,
      "num_input_tokens_seen": 50858590,
      "step": 2373,
      "time_per_iteration": 3.6427736282348633
    },
    {
      "auxiliary_loss_clip": 0.01200476,
      "auxiliary_loss_mlp": 0.01042421,
      "balance_loss_clip": 1.06407309,
      "balance_loss_mlp": 1.03215098,
      "epoch": 0.28545662237720193,
      "flos": 19861940776320.0,
      "grad_norm": 6.716385596821076,
      "language_loss": 0.8348158,
      "learning_rate": 3.353778697750527e-06,
      "loss": 0.85724479,
      "num_input_tokens_seen": 50876995,
      "step": 2374,
      "time_per_iteration": 2.719449758529663
    },
    {
      "auxiliary_loss_clip": 0.01189254,
      "auxiliary_loss_mlp": 0.01035353,
      "balance_loss_clip": 1.06225967,
      "balance_loss_mlp": 1.02550626,
      "epoch": 0.28557686526784104,
      "flos": 23879016241920.0,
      "grad_norm": 3.4536292071403305,
      "language_loss": 0.89479512,
      "learning_rate": 3.353205205904622e-06,
      "loss": 0.91704118,
      "num_input_tokens_seen": 50896105,
      "step": 2375,
      "time_per_iteration": 2.7376482486724854
    },
    {
      "auxiliary_loss_clip": 0.01180675,
      "auxiliary_loss_mlp": 0.0103585,
      "balance_loss_clip": 1.05898952,
      "balance_loss_mlp": 1.0258956,
      "epoch": 0.28569710815848015,
      "flos": 44890384233600.0,
      "grad_norm": 2.2925553050480985,
      "language_loss": 0.72177958,
      "learning_rate": 3.3526315087785637e-06,
      "loss": 0.74394476,
      "num_input_tokens_seen": 50917220,
      "step": 2376,
      "time_per_iteration": 2.974684238433838
    },
    {
      "auxiliary_loss_clip": 0.01205376,
      "auxiliary_loss_mlp": 0.01045809,
      "balance_loss_clip": 1.06605875,
      "balance_loss_mlp": 1.0351274,
      "epoch": 0.2858173510491192,
      "flos": 26829759628800.0,
      "grad_norm": 1.8267399147589005,
      "language_loss": 0.81236994,
      "learning_rate": 3.3520576064593805e-06,
      "loss": 0.83488178,
      "num_input_tokens_seen": 50937175,
      "step": 2377,
      "time_per_iteration": 4.624175548553467
    },
    {
      "auxiliary_loss_clip": 0.01152246,
      "auxiliary_loss_mlp": 0.01031941,
      "balance_loss_clip": 1.05834162,
      "balance_loss_mlp": 1.02166533,
      "epoch": 0.2859375939397583,
      "flos": 23148916398720.0,
      "grad_norm": 1.6193503366040995,
      "language_loss": 0.81983602,
      "learning_rate": 3.3514834990341337e-06,
      "loss": 0.8416779,
      "num_input_tokens_seen": 50957500,
      "step": 2378,
      "time_per_iteration": 2.8290011882781982
    },
    {
      "auxiliary_loss_clip": 0.01176819,
      "auxiliary_loss_mlp": 0.01049474,
      "balance_loss_clip": 1.06109619,
      "balance_loss_mlp": 1.03914428,
      "epoch": 0.2860578368303974,
      "flos": 12129799397760.0,
      "grad_norm": 3.0003835554917493,
      "language_loss": 0.92980301,
      "learning_rate": 3.3509091865899144e-06,
      "loss": 0.95206594,
      "num_input_tokens_seen": 50972690,
      "step": 2379,
      "time_per_iteration": 2.740726947784424
    },
    {
      "auxiliary_loss_clip": 0.01144824,
      "auxiliary_loss_mlp": 0.00765234,
      "balance_loss_clip": 1.0566076,
      "balance_loss_mlp": 1.00060153,
      "epoch": 0.2861780797210365,
      "flos": 19938035738880.0,
      "grad_norm": 1.9546884876392738,
      "language_loss": 0.7045927,
      "learning_rate": 3.350334669213846e-06,
      "loss": 0.72369331,
      "num_input_tokens_seen": 50990095,
      "step": 2380,
      "time_per_iteration": 2.87402081489563
    },
    {
      "auxiliary_loss_clip": 0.01194481,
      "auxiliary_loss_mlp": 0.01039607,
      "balance_loss_clip": 1.0665884,
      "balance_loss_mlp": 1.02957535,
      "epoch": 0.2862983226116756,
      "flos": 27563127609600.0,
      "grad_norm": 1.9701342476283505,
      "language_loss": 0.75887442,
      "learning_rate": 3.3497599469930816e-06,
      "loss": 0.78121531,
      "num_input_tokens_seen": 51008305,
      "step": 2381,
      "time_per_iteration": 2.7387959957122803
    },
    {
      "auxiliary_loss_clip": 0.01208879,
      "auxiliary_loss_mlp": 0.010367,
      "balance_loss_clip": 1.06473875,
      "balance_loss_mlp": 1.02598906,
      "epoch": 0.28641856550231465,
      "flos": 22053964158720.0,
      "grad_norm": 2.3005897820843715,
      "language_loss": 0.83311093,
      "learning_rate": 3.349185020014807e-06,
      "loss": 0.85556674,
      "num_input_tokens_seen": 51025570,
      "step": 2382,
      "time_per_iteration": 2.7539174556732178
    },
    {
      "auxiliary_loss_clip": 0.01173796,
      "auxiliary_loss_mlp": 0.01040042,
      "balance_loss_clip": 1.0586822,
      "balance_loss_mlp": 1.02984917,
      "epoch": 0.28653880839295376,
      "flos": 22378775869440.0,
      "grad_norm": 2.457498676755163,
      "language_loss": 0.74729908,
      "learning_rate": 3.348609888366237e-06,
      "loss": 0.76943743,
      "num_input_tokens_seen": 51044585,
      "step": 2383,
      "time_per_iteration": 2.7616801261901855
    },
    {
      "auxiliary_loss_clip": 0.01191317,
      "auxiliary_loss_mlp": 0.01037408,
      "balance_loss_clip": 1.06372988,
      "balance_loss_mlp": 1.02828884,
      "epoch": 0.28665905128359287,
      "flos": 23367971491200.0,
      "grad_norm": 2.3048178215471693,
      "language_loss": 0.6277529,
      "learning_rate": 3.348034552134619e-06,
      "loss": 0.65004015,
      "num_input_tokens_seen": 51063990,
      "step": 2384,
      "time_per_iteration": 2.8028039932250977
    },
    {
      "auxiliary_loss_clip": 0.01133266,
      "auxiliary_loss_mlp": 0.01042614,
      "balance_loss_clip": 1.0501591,
      "balance_loss_mlp": 1.03333998,
      "epoch": 0.2867792941742319,
      "flos": 20881695893760.0,
      "grad_norm": 2.2598743111929607,
      "language_loss": 0.84517777,
      "learning_rate": 3.3474590114072316e-06,
      "loss": 0.86693662,
      "num_input_tokens_seen": 51081990,
      "step": 2385,
      "time_per_iteration": 2.846290349960327
    },
    {
      "auxiliary_loss_clip": 0.01179885,
      "auxiliary_loss_mlp": 0.01035545,
      "balance_loss_clip": 1.0615226,
      "balance_loss_mlp": 1.025388,
      "epoch": 0.28689953706487104,
      "flos": 20664005518080.0,
      "grad_norm": 2.4474990880400695,
      "language_loss": 0.83183801,
      "learning_rate": 3.3468832662713836e-06,
      "loss": 0.85399234,
      "num_input_tokens_seen": 51100235,
      "step": 2386,
      "time_per_iteration": 2.7622427940368652
    },
    {
      "auxiliary_loss_clip": 0.01179087,
      "auxiliary_loss_mlp": 0.01040085,
      "balance_loss_clip": 1.05876648,
      "balance_loss_mlp": 1.03005981,
      "epoch": 0.28701977995551015,
      "flos": 12675533708160.0,
      "grad_norm": 2.6117587004976865,
      "language_loss": 0.83918464,
      "learning_rate": 3.346307316814415e-06,
      "loss": 0.8613764,
      "num_input_tokens_seen": 51115405,
      "step": 2387,
      "time_per_iteration": 2.745777130126953
    },
    {
      "auxiliary_loss_clip": 0.01221839,
      "auxiliary_loss_mlp": 0.01043985,
      "balance_loss_clip": 1.0690155,
      "balance_loss_mlp": 1.03299367,
      "epoch": 0.2871400228461492,
      "flos": 21252366293760.0,
      "grad_norm": 5.43317677905549,
      "language_loss": 0.75663412,
      "learning_rate": 3.3457311631236965e-06,
      "loss": 0.77929235,
      "num_input_tokens_seen": 51136390,
      "step": 2388,
      "time_per_iteration": 2.6852216720581055
    },
    {
      "auxiliary_loss_clip": 0.01164889,
      "auxiliary_loss_mlp": 0.01036942,
      "balance_loss_clip": 1.0591104,
      "balance_loss_mlp": 1.02571237,
      "epoch": 0.2872602657367883,
      "flos": 25119262995840.0,
      "grad_norm": 2.3018523947256546,
      "language_loss": 0.848665,
      "learning_rate": 3.345154805286631e-06,
      "loss": 0.87068331,
      "num_input_tokens_seen": 51156650,
      "step": 2389,
      "time_per_iteration": 2.891362428665161
    },
    {
      "auxiliary_loss_clip": 0.0120644,
      "auxiliary_loss_mlp": 0.01035647,
      "balance_loss_clip": 1.06444502,
      "balance_loss_mlp": 1.02605045,
      "epoch": 0.2873805086274274,
      "flos": 16646606830080.0,
      "grad_norm": 2.932566635598745,
      "language_loss": 0.7617349,
      "learning_rate": 3.344578243390651e-06,
      "loss": 0.78415573,
      "num_input_tokens_seen": 51172210,
      "step": 2390,
      "time_per_iteration": 2.649172067642212
    },
    {
      "auxiliary_loss_clip": 0.01163364,
      "auxiliary_loss_mlp": 0.01034612,
      "balance_loss_clip": 1.06124437,
      "balance_loss_mlp": 1.02444375,
      "epoch": 0.2875007515180665,
      "flos": 17420123237760.0,
      "grad_norm": 2.83692802687859,
      "language_loss": 0.78719187,
      "learning_rate": 3.3440014775232206e-06,
      "loss": 0.80917162,
      "num_input_tokens_seen": 51190265,
      "step": 2391,
      "time_per_iteration": 2.935469388961792
    },
    {
      "auxiliary_loss_clip": 0.01201654,
      "auxiliary_loss_mlp": 0.01039133,
      "balance_loss_clip": 1.06614339,
      "balance_loss_mlp": 1.03059721,
      "epoch": 0.2876209944087056,
      "flos": 23434190213760.0,
      "grad_norm": 3.8429756984429844,
      "language_loss": 0.71711659,
      "learning_rate": 3.343424507771834e-06,
      "loss": 0.73952448,
      "num_input_tokens_seen": 51208475,
      "step": 2392,
      "time_per_iteration": 2.741016387939453
    },
    {
      "auxiliary_loss_clip": 0.01204882,
      "auxiliary_loss_mlp": 0.01034858,
      "balance_loss_clip": 1.06488752,
      "balance_loss_mlp": 1.0247606,
      "epoch": 0.2877412372993447,
      "flos": 13735509079680.0,
      "grad_norm": 2.0719879404673844,
      "language_loss": 0.87069571,
      "learning_rate": 3.342847334224018e-06,
      "loss": 0.89309311,
      "num_input_tokens_seen": 51225875,
      "step": 2393,
      "time_per_iteration": 2.717022657394409
    },
    {
      "auxiliary_loss_clip": 0.01081217,
      "auxiliary_loss_mlp": 0.01003729,
      "balance_loss_clip": 1.03247833,
      "balance_loss_mlp": 1.00191689,
      "epoch": 0.28786148018998375,
      "flos": 58079695104000.0,
      "grad_norm": 0.9541313730589216,
      "language_loss": 0.62333703,
      "learning_rate": 3.342269956967329e-06,
      "loss": 0.6441865,
      "num_input_tokens_seen": 51287780,
      "step": 2394,
      "time_per_iteration": 3.3808224201202393
    },
    {
      "auxiliary_loss_clip": 0.0117801,
      "auxiliary_loss_mlp": 0.0103818,
      "balance_loss_clip": 1.06406951,
      "balance_loss_mlp": 1.02733839,
      "epoch": 0.28798172308062286,
      "flos": 23435052140160.0,
      "grad_norm": 2.7907549212102656,
      "language_loss": 0.71844113,
      "learning_rate": 3.341692376089355e-06,
      "loss": 0.74060297,
      "num_input_tokens_seen": 51303335,
      "step": 2395,
      "time_per_iteration": 2.7472095489501953
    },
    {
      "auxiliary_loss_clip": 0.01169392,
      "auxiliary_loss_mlp": 0.0103668,
      "balance_loss_clip": 1.06096935,
      "balance_loss_mlp": 1.02669001,
      "epoch": 0.288101965971262,
      "flos": 25110033200640.0,
      "grad_norm": 8.765143055589126,
      "language_loss": 0.84601182,
      "learning_rate": 3.3411145916777146e-06,
      "loss": 0.86807257,
      "num_input_tokens_seen": 51317495,
      "step": 2396,
      "time_per_iteration": 2.768209457397461
    },
    {
      "auxiliary_loss_clip": 0.01209704,
      "auxiliary_loss_mlp": 0.01039098,
      "balance_loss_clip": 1.0684458,
      "balance_loss_mlp": 1.02861381,
      "epoch": 0.28822220886190103,
      "flos": 16252559654400.0,
      "grad_norm": 3.7072133060418047,
      "language_loss": 0.91090858,
      "learning_rate": 3.3405366038200566e-06,
      "loss": 0.93339664,
      "num_input_tokens_seen": 51336430,
      "step": 2397,
      "time_per_iteration": 2.6511199474334717
    },
    {
      "auxiliary_loss_clip": 0.01193783,
      "auxiliary_loss_mlp": 0.01036582,
      "balance_loss_clip": 1.06727219,
      "balance_loss_mlp": 1.02553177,
      "epoch": 0.28834245175254014,
      "flos": 24535642815360.0,
      "grad_norm": 2.7639098886227638,
      "language_loss": 0.850438,
      "learning_rate": 3.3399584126040617e-06,
      "loss": 0.8727417,
      "num_input_tokens_seen": 51355930,
      "step": 2398,
      "time_per_iteration": 2.733652353286743
    },
    {
      "auxiliary_loss_clip": 0.01147374,
      "auxiliary_loss_mlp": 0.01036267,
      "balance_loss_clip": 1.05535483,
      "balance_loss_mlp": 1.02677226,
      "epoch": 0.2884626946431792,
      "flos": 24571445696640.0,
      "grad_norm": 2.199583840842096,
      "language_loss": 0.91147488,
      "learning_rate": 3.339380018117441e-06,
      "loss": 0.93331128,
      "num_input_tokens_seen": 51376765,
      "step": 2399,
      "time_per_iteration": 3.8399150371551514
    },
    {
      "auxiliary_loss_clip": 0.01154958,
      "auxiliary_loss_mlp": 0.01042388,
      "balance_loss_clip": 1.05473971,
      "balance_loss_mlp": 1.03244615,
      "epoch": 0.2885829375338183,
      "flos": 16544657053440.0,
      "grad_norm": 3.3843663893312113,
      "language_loss": 0.79029435,
      "learning_rate": 3.3388014204479366e-06,
      "loss": 0.81226778,
      "num_input_tokens_seen": 51394570,
      "step": 2400,
      "time_per_iteration": 2.8397653102874756
    },
    {
      "auxiliary_loss_clip": 0.01163773,
      "auxiliary_loss_mlp": 0.01035424,
      "balance_loss_clip": 1.06517935,
      "balance_loss_mlp": 1.02572656,
      "epoch": 0.2887031804244574,
      "flos": 24061226958720.0,
      "grad_norm": 2.340740246236037,
      "language_loss": 0.91462678,
      "learning_rate": 3.338222619683321e-06,
      "loss": 0.93661875,
      "num_input_tokens_seen": 51414535,
      "step": 2401,
      "time_per_iteration": 2.922966241836548
    },
    {
      "auxiliary_loss_clip": 0.0115756,
      "auxiliary_loss_mlp": 0.00764807,
      "balance_loss_clip": 1.05929959,
      "balance_loss_mlp": 1.00071013,
      "epoch": 0.2888234233150965,
      "flos": 23330696152320.0,
      "grad_norm": 5.52528284776393,
      "language_loss": 0.74180639,
      "learning_rate": 3.337643615911398e-06,
      "loss": 0.76103002,
      "num_input_tokens_seen": 51434160,
      "step": 2402,
      "time_per_iteration": 2.7818849086761475
    },
    {
      "auxiliary_loss_clip": 0.01135276,
      "auxiliary_loss_mlp": 0.01036456,
      "balance_loss_clip": 1.05300212,
      "balance_loss_mlp": 1.02635908,
      "epoch": 0.2889436662057356,
      "flos": 22272767856000.0,
      "grad_norm": 2.9934928500657687,
      "language_loss": 0.79184139,
      "learning_rate": 3.3370644092200026e-06,
      "loss": 0.81355876,
      "num_input_tokens_seen": 51451435,
      "step": 2403,
      "time_per_iteration": 3.8433034420013428
    },
    {
      "auxiliary_loss_clip": 0.01157986,
      "auxiliary_loss_mlp": 0.01036579,
      "balance_loss_clip": 1.05778241,
      "balance_loss_mlp": 1.02594543,
      "epoch": 0.2890639090963747,
      "flos": 21616931381760.0,
      "grad_norm": 2.9322857288537554,
      "language_loss": 0.79163063,
      "learning_rate": 3.3364849996969985e-06,
      "loss": 0.81357622,
      "num_input_tokens_seen": 51471455,
      "step": 2404,
      "time_per_iteration": 2.8602771759033203
    },
    {
      "auxiliary_loss_clip": 0.01136213,
      "auxiliary_loss_mlp": 0.01035758,
      "balance_loss_clip": 1.05431318,
      "balance_loss_mlp": 1.02651322,
      "epoch": 0.28918415198701375,
      "flos": 28585540333440.0,
      "grad_norm": 10.345993369862754,
      "language_loss": 0.85676825,
      "learning_rate": 3.335905387430283e-06,
      "loss": 0.878488,
      "num_input_tokens_seen": 51492890,
      "step": 2405,
      "time_per_iteration": 2.860274314880371
    },
    {
      "auxiliary_loss_clip": 0.01147886,
      "auxiliary_loss_mlp": 0.01042915,
      "balance_loss_clip": 1.05721188,
      "balance_loss_mlp": 1.03298545,
      "epoch": 0.28930439487765286,
      "flos": 21944688007680.0,
      "grad_norm": 2.609052329237414,
      "language_loss": 0.83120406,
      "learning_rate": 3.335325572507782e-06,
      "loss": 0.85311204,
      "num_input_tokens_seen": 51513390,
      "step": 2406,
      "time_per_iteration": 2.8501346111297607
    },
    {
      "auxiliary_loss_clip": 0.01183746,
      "auxiliary_loss_mlp": 0.01036936,
      "balance_loss_clip": 1.06381822,
      "balance_loss_mlp": 1.026088,
      "epoch": 0.28942463776829197,
      "flos": 19281911955840.0,
      "grad_norm": 1.9970081340853247,
      "language_loss": 0.74240541,
      "learning_rate": 3.3347455550174537e-06,
      "loss": 0.76461226,
      "num_input_tokens_seen": 51532730,
      "step": 2407,
      "time_per_iteration": 2.7140605449676514
    },
    {
      "auxiliary_loss_clip": 0.01187064,
      "auxiliary_loss_mlp": 0.00764464,
      "balance_loss_clip": 1.06408548,
      "balance_loss_mlp": 1.00065792,
      "epoch": 0.289544880658931,
      "flos": 14645700737280.0,
      "grad_norm": 2.1509880179390337,
      "language_loss": 0.68196154,
      "learning_rate": 3.3341653350472864e-06,
      "loss": 0.70147681,
      "num_input_tokens_seen": 51549560,
      "step": 2408,
      "time_per_iteration": 2.871652364730835
    },
    {
      "auxiliary_loss_clip": 0.01170531,
      "auxiliary_loss_mlp": 0.01039049,
      "balance_loss_clip": 1.06179225,
      "balance_loss_mlp": 1.02845764,
      "epoch": 0.28966512354957014,
      "flos": 28621881918720.0,
      "grad_norm": 2.660452895960593,
      "language_loss": 0.6940766,
      "learning_rate": 3.333584912685298e-06,
      "loss": 0.71617234,
      "num_input_tokens_seen": 51568180,
      "step": 2409,
      "time_per_iteration": 2.866236448287964
    },
    {
      "auxiliary_loss_clip": 0.01113423,
      "auxiliary_loss_mlp": 0.01004289,
      "balance_loss_clip": 1.02926517,
      "balance_loss_mlp": 1.002406,
      "epoch": 0.28978536644020925,
      "flos": 64711784511360.0,
      "grad_norm": 0.8842757508903664,
      "language_loss": 0.55519664,
      "learning_rate": 3.3330042880195385e-06,
      "loss": 0.57637376,
      "num_input_tokens_seen": 51622530,
      "step": 2410,
      "time_per_iteration": 3.2603394985198975
    },
    {
      "auxiliary_loss_clip": 0.01187195,
      "auxiliary_loss_mlp": 0.01036474,
      "balance_loss_clip": 1.05913675,
      "balance_loss_mlp": 1.02646041,
      "epoch": 0.2899056093308483,
      "flos": 18624638937600.0,
      "grad_norm": 2.089976528293924,
      "language_loss": 0.78376073,
      "learning_rate": 3.3324234611380888e-06,
      "loss": 0.80599737,
      "num_input_tokens_seen": 51641260,
      "step": 2411,
      "time_per_iteration": 2.844780683517456
    },
    {
      "auxiliary_loss_clip": 0.01157156,
      "auxiliary_loss_mlp": 0.01039215,
      "balance_loss_clip": 1.06098449,
      "balance_loss_mlp": 1.02864718,
      "epoch": 0.2900258522214874,
      "flos": 22893735202560.0,
      "grad_norm": 3.679313312425561,
      "language_loss": 0.82176805,
      "learning_rate": 3.3318424321290596e-06,
      "loss": 0.8437317,
      "num_input_tokens_seen": 51660975,
      "step": 2412,
      "time_per_iteration": 2.7634310722351074
    },
    {
      "auxiliary_loss_clip": 0.0104783,
      "auxiliary_loss_mlp": 0.01000697,
      "balance_loss_clip": 1.02703118,
      "balance_loss_mlp": 0.99886078,
      "epoch": 0.2901460951121265,
      "flos": 71106036013440.0,
      "grad_norm": 0.8418183826287228,
      "language_loss": 0.59874618,
      "learning_rate": 3.3312612010805917e-06,
      "loss": 0.61923146,
      "num_input_tokens_seen": 51720550,
      "step": 2413,
      "time_per_iteration": 3.439948558807373
    },
    {
      "auxiliary_loss_clip": 0.01192423,
      "auxiliary_loss_mlp": 0.01040787,
      "balance_loss_clip": 1.06226873,
      "balance_loss_mlp": 1.02952719,
      "epoch": 0.2902663380027656,
      "flos": 32160986081280.0,
      "grad_norm": 6.2144469972631144,
      "language_loss": 0.70335197,
      "learning_rate": 3.330679768080858e-06,
      "loss": 0.72568405,
      "num_input_tokens_seen": 51744435,
      "step": 2414,
      "time_per_iteration": 2.717602491378784
    },
    {
      "auxiliary_loss_clip": 0.01128567,
      "auxiliary_loss_mlp": 0.01042234,
      "balance_loss_clip": 1.05649197,
      "balance_loss_mlp": 1.03220272,
      "epoch": 0.2903865808934047,
      "flos": 29351658539520.0,
      "grad_norm": 3.265888864487854,
      "language_loss": 0.83679676,
      "learning_rate": 3.3300981332180627e-06,
      "loss": 0.85850471,
      "num_input_tokens_seen": 51763640,
      "step": 2415,
      "time_per_iteration": 2.8818211555480957
    },
    {
      "auxiliary_loss_clip": 0.01173252,
      "auxiliary_loss_mlp": 0.00763924,
      "balance_loss_clip": 1.06249642,
      "balance_loss_mlp": 1.00073361,
      "epoch": 0.29050682378404374,
      "flos": 17089026647040.0,
      "grad_norm": 2.00432934182267,
      "language_loss": 0.80422747,
      "learning_rate": 3.3295162965804373e-06,
      "loss": 0.82359922,
      "num_input_tokens_seen": 51782135,
      "step": 2416,
      "time_per_iteration": 2.762260675430298
    },
    {
      "auxiliary_loss_clip": 0.01185252,
      "auxiliary_loss_mlp": 0.01045823,
      "balance_loss_clip": 1.06286228,
      "balance_loss_mlp": 1.03612566,
      "epoch": 0.29062706667468285,
      "flos": 17858233422720.0,
      "grad_norm": 2.249664312714899,
      "language_loss": 0.7880491,
      "learning_rate": 3.328934258256247e-06,
      "loss": 0.81035984,
      "num_input_tokens_seen": 51800200,
      "step": 2417,
      "time_per_iteration": 2.6697733402252197
    },
    {
      "auxiliary_loss_clip": 0.01181876,
      "auxiliary_loss_mlp": 0.01038778,
      "balance_loss_clip": 1.05754256,
      "balance_loss_mlp": 1.02825832,
      "epoch": 0.29074730956532197,
      "flos": 24279815174400.0,
      "grad_norm": 2.3932351758461876,
      "language_loss": 0.67001712,
      "learning_rate": 3.3283520183337856e-06,
      "loss": 0.69222367,
      "num_input_tokens_seen": 51819905,
      "step": 2418,
      "time_per_iteration": 2.7234973907470703
    },
    {
      "auxiliary_loss_clip": 0.01152942,
      "auxiliary_loss_mlp": 0.01038957,
      "balance_loss_clip": 1.05380702,
      "balance_loss_mlp": 1.02920532,
      "epoch": 0.290867552455961,
      "flos": 22340961826560.0,
      "grad_norm": 2.1279145835583826,
      "language_loss": 0.69414872,
      "learning_rate": 3.3277695769013797e-06,
      "loss": 0.71606767,
      "num_input_tokens_seen": 51839350,
      "step": 2419,
      "time_per_iteration": 2.7549917697906494
    },
    {
      "auxiliary_loss_clip": 0.01148819,
      "auxiliary_loss_mlp": 0.00764489,
      "balance_loss_clip": 1.05356395,
      "balance_loss_mlp": 1.00068736,
      "epoch": 0.29098779534660013,
      "flos": 23186155824000.0,
      "grad_norm": 3.998972245271686,
      "language_loss": 0.77239317,
      "learning_rate": 3.327186934047385e-06,
      "loss": 0.7915262,
      "num_input_tokens_seen": 51858045,
      "step": 2420,
      "time_per_iteration": 2.816958427429199
    },
    {
      "auxiliary_loss_clip": 0.01200261,
      "auxiliary_loss_mlp": 0.01040016,
      "balance_loss_clip": 1.06591725,
      "balance_loss_mlp": 1.03062868,
      "epoch": 0.29110803823723924,
      "flos": 15304194817920.0,
      "grad_norm": 3.4017936869211884,
      "language_loss": 0.65766615,
      "learning_rate": 3.3266040898601877e-06,
      "loss": 0.68006897,
      "num_input_tokens_seen": 51875880,
      "step": 2421,
      "time_per_iteration": 2.690354347229004
    },
    {
      "auxiliary_loss_clip": 0.01162624,
      "auxiliary_loss_mlp": 0.01047715,
      "balance_loss_clip": 1.06341326,
      "balance_loss_mlp": 1.03754687,
      "epoch": 0.2912282811278783,
      "flos": 22595352923520.0,
      "grad_norm": 3.076961953396218,
      "language_loss": 0.77999443,
      "learning_rate": 3.3260210444282045e-06,
      "loss": 0.8020978,
      "num_input_tokens_seen": 51893835,
      "step": 2422,
      "time_per_iteration": 2.8119871616363525
    },
    {
      "auxiliary_loss_clip": 0.01149701,
      "auxiliary_loss_mlp": 0.01038101,
      "balance_loss_clip": 1.05410182,
      "balance_loss_mlp": 1.02811122,
      "epoch": 0.2913485240185174,
      "flos": 24497900599680.0,
      "grad_norm": 2.7081075618670694,
      "language_loss": 0.72859645,
      "learning_rate": 3.325437797839883e-06,
      "loss": 0.75047445,
      "num_input_tokens_seen": 51912205,
      "step": 2423,
      "time_per_iteration": 2.7564449310302734
    },
    {
      "auxiliary_loss_clip": 0.01164025,
      "auxiliary_loss_mlp": 0.01042038,
      "balance_loss_clip": 1.0587132,
      "balance_loss_mlp": 1.03126776,
      "epoch": 0.2914687669091565,
      "flos": 17931024334080.0,
      "grad_norm": 2.6736500590990233,
      "language_loss": 0.74469233,
      "learning_rate": 3.3248543501837015e-06,
      "loss": 0.76675302,
      "num_input_tokens_seen": 51929410,
      "step": 2424,
      "time_per_iteration": 2.794908046722412
    },
    {
      "auxiliary_loss_clip": 0.01203746,
      "auxiliary_loss_mlp": 0.01036743,
      "balance_loss_clip": 1.0682112,
      "balance_loss_mlp": 1.02725959,
      "epoch": 0.2915890097997956,
      "flos": 22529313768960.0,
      "grad_norm": 2.5392749077494745,
      "language_loss": 0.77629602,
      "learning_rate": 3.3242707015481684e-06,
      "loss": 0.79870087,
      "num_input_tokens_seen": 51949345,
      "step": 2425,
      "time_per_iteration": 3.6505563259124756
    },
    {
      "auxiliary_loss_clip": 0.01201808,
      "auxiliary_loss_mlp": 0.01037634,
      "balance_loss_clip": 1.06541598,
      "balance_loss_mlp": 1.02709603,
      "epoch": 0.2917092526904347,
      "flos": 13845216193920.0,
      "grad_norm": 1.9828486099499723,
      "language_loss": 0.80960447,
      "learning_rate": 3.323686852021823e-06,
      "loss": 0.83199894,
      "num_input_tokens_seen": 51966855,
      "step": 2426,
      "time_per_iteration": 2.8428215980529785
    },
    {
      "auxiliary_loss_clip": 0.011846,
      "auxiliary_loss_mlp": 0.01048655,
      "balance_loss_clip": 1.06406569,
      "balance_loss_mlp": 1.03784323,
      "epoch": 0.2918294955810738,
      "flos": 22674859678080.0,
      "grad_norm": 2.059710648319389,
      "language_loss": 0.79671556,
      "learning_rate": 3.323102801693235e-06,
      "loss": 0.81904805,
      "num_input_tokens_seen": 51985620,
      "step": 2427,
      "time_per_iteration": 2.790914297103882
    },
    {
      "auxiliary_loss_clip": 0.01206364,
      "auxiliary_loss_mlp": 0.01041676,
      "balance_loss_clip": 1.06723952,
      "balance_loss_mlp": 1.03156734,
      "epoch": 0.29194973847171285,
      "flos": 23438284364160.0,
      "grad_norm": 2.15260013853555,
      "language_loss": 0.80569744,
      "learning_rate": 3.322518550651003e-06,
      "loss": 0.82817781,
      "num_input_tokens_seen": 52004930,
      "step": 2428,
      "time_per_iteration": 3.609893321990967
    },
    {
      "auxiliary_loss_clip": 0.01218767,
      "auxiliary_loss_mlp": 0.01038745,
      "balance_loss_clip": 1.06733,
      "balance_loss_mlp": 1.02867162,
      "epoch": 0.29206998136235196,
      "flos": 21909064694400.0,
      "grad_norm": 2.9137661856939,
      "language_loss": 0.81593966,
      "learning_rate": 3.3219340989837586e-06,
      "loss": 0.8385148,
      "num_input_tokens_seen": 52024920,
      "step": 2429,
      "time_per_iteration": 3.5514559745788574
    },
    {
      "auxiliary_loss_clip": 0.0117146,
      "auxiliary_loss_mlp": 0.00763788,
      "balance_loss_clip": 1.05579925,
      "balance_loss_mlp": 1.00079906,
      "epoch": 0.292190224252991,
      "flos": 23215925220480.0,
      "grad_norm": 2.1074751087947154,
      "language_loss": 0.8044765,
      "learning_rate": 3.3213494467801625e-06,
      "loss": 0.82382894,
      "num_input_tokens_seen": 52044095,
      "step": 2430,
      "time_per_iteration": 2.737684488296509
    },
    {
      "auxiliary_loss_clip": 0.01178099,
      "auxiliary_loss_mlp": 0.01038453,
      "balance_loss_clip": 1.06064129,
      "balance_loss_mlp": 1.027933,
      "epoch": 0.2923104671436301,
      "flos": 20740818752640.0,
      "grad_norm": 3.256378468535536,
      "language_loss": 0.71479726,
      "learning_rate": 3.3207645941289063e-06,
      "loss": 0.7369628,
      "num_input_tokens_seen": 52062440,
      "step": 2431,
      "time_per_iteration": 2.6800689697265625
    },
    {
      "auxiliary_loss_clip": 0.01172731,
      "auxiliary_loss_mlp": 0.00764265,
      "balance_loss_clip": 1.05857718,
      "balance_loss_mlp": 1.00082278,
      "epoch": 0.29243071003426924,
      "flos": 35809114999680.0,
      "grad_norm": 7.263899794068548,
      "language_loss": 0.80355215,
      "learning_rate": 3.320179541118711e-06,
      "loss": 0.82292211,
      "num_input_tokens_seen": 52084940,
      "step": 2432,
      "time_per_iteration": 2.863912582397461
    },
    {
      "auxiliary_loss_clip": 0.01093763,
      "auxiliary_loss_mlp": 0.01005921,
      "balance_loss_clip": 1.02576208,
      "balance_loss_mlp": 1.00415707,
      "epoch": 0.2925509529249083,
      "flos": 58081598524800.0,
      "grad_norm": 1.0096759472921653,
      "language_loss": 0.60317814,
      "learning_rate": 3.3195942878383293e-06,
      "loss": 0.62417501,
      "num_input_tokens_seen": 52141040,
      "step": 2433,
      "time_per_iteration": 3.2472338676452637
    },
    {
      "auxiliary_loss_clip": 0.01163765,
      "auxiliary_loss_mlp": 0.01038229,
      "balance_loss_clip": 1.06255138,
      "balance_loss_mlp": 1.02764344,
      "epoch": 0.2926711958155474,
      "flos": 21397122103680.0,
      "grad_norm": 2.1421205985147957,
      "language_loss": 0.7812103,
      "learning_rate": 3.319008834376543e-06,
      "loss": 0.80323029,
      "num_input_tokens_seen": 52160730,
      "step": 2434,
      "time_per_iteration": 2.8362443447113037
    },
    {
      "auxiliary_loss_clip": 0.01186514,
      "auxiliary_loss_mlp": 0.0103241,
      "balance_loss_clip": 1.06168795,
      "balance_loss_mlp": 1.02192605,
      "epoch": 0.2927914387061865,
      "flos": 23185796688000.0,
      "grad_norm": 2.978711316676029,
      "language_loss": 0.88652992,
      "learning_rate": 3.3184231808221654e-06,
      "loss": 0.90871918,
      "num_input_tokens_seen": 52175055,
      "step": 2435,
      "time_per_iteration": 2.7822656631469727
    },
    {
      "auxiliary_loss_clip": 0.01181727,
      "auxiliary_loss_mlp": 0.00764378,
      "balance_loss_clip": 1.05873132,
      "balance_loss_mlp": 1.00079465,
      "epoch": 0.29291168159682557,
      "flos": 22455553190400.0,
      "grad_norm": 2.1563014517478236,
      "language_loss": 0.62408936,
      "learning_rate": 3.3178373272640394e-06,
      "loss": 0.6435504,
      "num_input_tokens_seen": 52194150,
      "step": 2436,
      "time_per_iteration": 2.77414608001709
    },
    {
      "auxiliary_loss_clip": 0.01190229,
      "auxiliary_loss_mlp": 0.01031989,
      "balance_loss_clip": 1.06497955,
      "balance_loss_mlp": 1.02247059,
      "epoch": 0.2930319244874647,
      "flos": 21170632896000.0,
      "grad_norm": 5.8557647969352535,
      "language_loss": 0.85574126,
      "learning_rate": 3.3172512737910387e-06,
      "loss": 0.87796348,
      "num_input_tokens_seen": 52211660,
      "step": 2437,
      "time_per_iteration": 2.7681515216827393
    },
    {
      "auxiliary_loss_clip": 0.01178335,
      "auxiliary_loss_mlp": 0.010439,
      "balance_loss_clip": 1.0613116,
      "balance_loss_mlp": 1.03338575,
      "epoch": 0.2931521673781038,
      "flos": 31357843931520.0,
      "grad_norm": 7.678828466819629,
      "language_loss": 0.88361889,
      "learning_rate": 3.3166650204920674e-06,
      "loss": 0.90584129,
      "num_input_tokens_seen": 52232830,
      "step": 2438,
      "time_per_iteration": 2.7844674587249756
    },
    {
      "auxiliary_loss_clip": 0.01207019,
      "auxiliary_loss_mlp": 0.01044075,
      "balance_loss_clip": 1.06805205,
      "balance_loss_mlp": 1.03288674,
      "epoch": 0.29327241026874284,
      "flos": 24200990778240.0,
      "grad_norm": 1.9414524464832104,
      "language_loss": 0.81831169,
      "learning_rate": 3.316078567456059e-06,
      "loss": 0.84082258,
      "num_input_tokens_seen": 52250670,
      "step": 2439,
      "time_per_iteration": 2.7390055656433105
    },
    {
      "auxiliary_loss_clip": 0.01189628,
      "auxiliary_loss_mlp": 0.01035877,
      "balance_loss_clip": 1.06031919,
      "balance_loss_mlp": 1.02576208,
      "epoch": 0.29339265315938196,
      "flos": 24242611662720.0,
      "grad_norm": 2.0959715585994867,
      "language_loss": 0.7615096,
      "learning_rate": 3.3154919147719786e-06,
      "loss": 0.78376466,
      "num_input_tokens_seen": 52271685,
      "step": 2440,
      "time_per_iteration": 2.670694351196289
    },
    {
      "auxiliary_loss_clip": 0.01174149,
      "auxiliary_loss_mlp": 0.01047598,
      "balance_loss_clip": 1.05880499,
      "balance_loss_mlp": 1.0372386,
      "epoch": 0.29351289605002107,
      "flos": 16946641134720.0,
      "grad_norm": 3.4909592053857055,
      "language_loss": 0.86864889,
      "learning_rate": 3.31490506252882e-06,
      "loss": 0.89086628,
      "num_input_tokens_seen": 52291065,
      "step": 2441,
      "time_per_iteration": 2.7125277519226074
    },
    {
      "auxiliary_loss_clip": 0.01200505,
      "auxiliary_loss_mlp": 0.01038474,
      "balance_loss_clip": 1.06685448,
      "balance_loss_mlp": 1.02901459,
      "epoch": 0.2936331389406601,
      "flos": 19829082810240.0,
      "grad_norm": 3.063946405334703,
      "language_loss": 0.84427851,
      "learning_rate": 3.31431801081561e-06,
      "loss": 0.86666834,
      "num_input_tokens_seen": 52310000,
      "step": 2442,
      "time_per_iteration": 2.67232608795166
    },
    {
      "auxiliary_loss_clip": 0.01069709,
      "auxiliary_loss_mlp": 0.01009183,
      "balance_loss_clip": 1.03639638,
      "balance_loss_mlp": 1.00734735,
      "epoch": 0.29375338183129923,
      "flos": 71416844398080.0,
      "grad_norm": 0.9036064755247719,
      "language_loss": 0.67875183,
      "learning_rate": 3.313730759721402e-06,
      "loss": 0.69954073,
      "num_input_tokens_seen": 52372930,
      "step": 2443,
      "time_per_iteration": 3.485297679901123
    },
    {
      "auxiliary_loss_clip": 0.01176173,
      "auxiliary_loss_mlp": 0.01042416,
      "balance_loss_clip": 1.06152105,
      "balance_loss_mlp": 1.03261745,
      "epoch": 0.29387362472193834,
      "flos": 22054502862720.0,
      "grad_norm": 2.2279845364942554,
      "language_loss": 0.86309481,
      "learning_rate": 3.313143309335282e-06,
      "loss": 0.88528067,
      "num_input_tokens_seen": 52391420,
      "step": 2444,
      "time_per_iteration": 3.1973564624786377
    },
    {
      "auxiliary_loss_clip": 0.01176229,
      "auxiliary_loss_mlp": 0.01036246,
      "balance_loss_clip": 1.0607326,
      "balance_loss_mlp": 1.02669144,
      "epoch": 0.2939938676125774,
      "flos": 22966418373120.0,
      "grad_norm": 2.138749454104239,
      "language_loss": 0.85060084,
      "learning_rate": 3.3125556597463665e-06,
      "loss": 0.87272561,
      "num_input_tokens_seen": 52410725,
      "step": 2445,
      "time_per_iteration": 2.8586528301239014
    },
    {
      "auxiliary_loss_clip": 0.01204924,
      "auxiliary_loss_mlp": 0.01038147,
      "balance_loss_clip": 1.06882977,
      "balance_loss_mlp": 1.02859259,
      "epoch": 0.2941141105032165,
      "flos": 31358705857920.0,
      "grad_norm": 2.3892761733333505,
      "language_loss": 0.66482359,
      "learning_rate": 3.311967811043801e-06,
      "loss": 0.68725431,
      "num_input_tokens_seen": 52432645,
      "step": 2446,
      "time_per_iteration": 2.7898857593536377
    },
    {
      "auxiliary_loss_clip": 0.01219222,
      "auxiliary_loss_mlp": 0.01042206,
      "balance_loss_clip": 1.0682745,
      "balance_loss_mlp": 1.03266978,
      "epoch": 0.29423435339385556,
      "flos": 23222138273280.0,
      "grad_norm": 2.2186441845213345,
      "language_loss": 0.82240283,
      "learning_rate": 3.3113797633167617e-06,
      "loss": 0.84501708,
      "num_input_tokens_seen": 52450940,
      "step": 2447,
      "time_per_iteration": 2.655651807785034
    },
    {
      "auxiliary_loss_clip": 0.01128512,
      "auxiliary_loss_mlp": 0.00764632,
      "balance_loss_clip": 1.05029202,
      "balance_loss_mlp": 1.00077462,
      "epoch": 0.2943545962844947,
      "flos": 26864054138880.0,
      "grad_norm": 8.277096330235867,
      "language_loss": 0.69530857,
      "learning_rate": 3.310791516654455e-06,
      "loss": 0.71424007,
      "num_input_tokens_seen": 52468000,
      "step": 2448,
      "time_per_iteration": 2.898733139038086
    },
    {
      "auxiliary_loss_clip": 0.01211074,
      "auxiliary_loss_mlp": 0.0103643,
      "balance_loss_clip": 1.06713533,
      "balance_loss_mlp": 1.02680981,
      "epoch": 0.2944748391751338,
      "flos": 20231677422720.0,
      "grad_norm": 2.406363014800737,
      "language_loss": 0.79463166,
      "learning_rate": 3.3102030711461177e-06,
      "loss": 0.81710672,
      "num_input_tokens_seen": 52487575,
      "step": 2449,
      "time_per_iteration": 2.707669258117676
    },
    {
      "auxiliary_loss_clip": 0.01171701,
      "auxiliary_loss_mlp": 0.00765285,
      "balance_loss_clip": 1.05645943,
      "balance_loss_mlp": 1.00073469,
      "epoch": 0.29459508206577284,
      "flos": 15960965045760.0,
      "grad_norm": 2.613564838551364,
      "language_loss": 0.68539643,
      "learning_rate": 3.3096144268810156e-06,
      "loss": 0.70476627,
      "num_input_tokens_seen": 52506335,
      "step": 2450,
      "time_per_iteration": 2.7204015254974365
    },
    {
      "auxiliary_loss_clip": 0.01161824,
      "auxiliary_loss_mlp": 0.01033869,
      "balance_loss_clip": 1.06008971,
      "balance_loss_mlp": 1.02374792,
      "epoch": 0.29471532495641195,
      "flos": 20412882558720.0,
      "grad_norm": 3.5344003464279106,
      "language_loss": 0.73125899,
      "learning_rate": 3.3090255839484462e-06,
      "loss": 0.75321591,
      "num_input_tokens_seen": 52524330,
      "step": 2451,
      "time_per_iteration": 3.737929105758667
    },
    {
      "auxiliary_loss_clip": 0.01169796,
      "auxiliary_loss_mlp": 0.01032661,
      "balance_loss_clip": 1.05684114,
      "balance_loss_mlp": 1.02279067,
      "epoch": 0.29483556784705106,
      "flos": 20376576887040.0,
      "grad_norm": 7.753252075553705,
      "language_loss": 0.85973269,
      "learning_rate": 3.3084365424377366e-06,
      "loss": 0.88175726,
      "num_input_tokens_seen": 52543095,
      "step": 2452,
      "time_per_iteration": 2.754371404647827
    },
    {
      "auxiliary_loss_clip": 0.0107091,
      "auxiliary_loss_mlp": 0.01000859,
      "balance_loss_clip": 1.02484787,
      "balance_loss_mlp": 0.9990595,
      "epoch": 0.2949558107376901,
      "flos": 68555660595840.0,
      "grad_norm": 0.728421420489077,
      "language_loss": 0.55947316,
      "learning_rate": 3.307847302438245e-06,
      "loss": 0.58019078,
      "num_input_tokens_seen": 52597075,
      "step": 2453,
      "time_per_iteration": 3.157928228378296
    },
    {
      "auxiliary_loss_clip": 0.01220022,
      "auxiliary_loss_mlp": 0.01037274,
      "balance_loss_clip": 1.06609416,
      "balance_loss_mlp": 1.02625895,
      "epoch": 0.2950760536283292,
      "flos": 16107085572480.0,
      "grad_norm": 11.694568461012906,
      "language_loss": 0.77820575,
      "learning_rate": 3.3072578640393562e-06,
      "loss": 0.80077869,
      "num_input_tokens_seen": 52614410,
      "step": 2454,
      "time_per_iteration": 3.6101043224334717
    },
    {
      "auxiliary_loss_clip": 0.01167648,
      "auxiliary_loss_mlp": 0.0104181,
      "balance_loss_clip": 1.05419075,
      "balance_loss_mlp": 1.03071785,
      "epoch": 0.29519629651896834,
      "flos": 20483626394880.0,
      "grad_norm": 3.432431409306598,
      "language_loss": 0.8002497,
      "learning_rate": 3.3066682273304886e-06,
      "loss": 0.8223443,
      "num_input_tokens_seen": 52632055,
      "step": 2455,
      "time_per_iteration": 3.7788424491882324
    },
    {
      "auxiliary_loss_clip": 0.01175404,
      "auxiliary_loss_mlp": 0.01040693,
      "balance_loss_clip": 1.06341374,
      "balance_loss_mlp": 1.03010106,
      "epoch": 0.2953165394096074,
      "flos": 18916484941440.0,
      "grad_norm": 1.998523108164261,
      "language_loss": 0.78648055,
      "learning_rate": 3.3060783924010904e-06,
      "loss": 0.80864155,
      "num_input_tokens_seen": 52649980,
      "step": 2456,
      "time_per_iteration": 2.712679862976074
    },
    {
      "auxiliary_loss_clip": 0.01149926,
      "auxiliary_loss_mlp": 0.01040512,
      "balance_loss_clip": 1.05590725,
      "balance_loss_mlp": 1.03047466,
      "epoch": 0.2954367823002465,
      "flos": 20624467622400.0,
      "grad_norm": 3.437152503827753,
      "language_loss": 0.85047913,
      "learning_rate": 3.3054883593406387e-06,
      "loss": 0.87238348,
      "num_input_tokens_seen": 52664730,
      "step": 2457,
      "time_per_iteration": 2.7721762657165527
    },
    {
      "auxiliary_loss_clip": 0.01159333,
      "auxiliary_loss_mlp": 0.01034197,
      "balance_loss_clip": 1.05434775,
      "balance_loss_mlp": 1.02361131,
      "epoch": 0.2955570251908856,
      "flos": 31175525473920.0,
      "grad_norm": 5.031555969338875,
      "language_loss": 0.65663362,
      "learning_rate": 3.3048981282386404e-06,
      "loss": 0.67856896,
      "num_input_tokens_seen": 52686040,
      "step": 2458,
      "time_per_iteration": 2.764948606491089
    },
    {
      "auxiliary_loss_clip": 0.011767,
      "auxiliary_loss_mlp": 0.00764501,
      "balance_loss_clip": 1.06333661,
      "balance_loss_mlp": 1.00082374,
      "epoch": 0.29567726808152467,
      "flos": 21650328051840.0,
      "grad_norm": 2.049203630338951,
      "language_loss": 0.82417762,
      "learning_rate": 3.304307699184634e-06,
      "loss": 0.84358966,
      "num_input_tokens_seen": 52704630,
      "step": 2459,
      "time_per_iteration": 2.7513413429260254
    },
    {
      "auxiliary_loss_clip": 0.01203507,
      "auxiliary_loss_mlp": 0.01031657,
      "balance_loss_clip": 1.06752706,
      "balance_loss_mlp": 1.02150035,
      "epoch": 0.2957975109721638,
      "flos": 24243868638720.0,
      "grad_norm": 4.60222828604539,
      "language_loss": 0.78910321,
      "learning_rate": 3.3037170722681866e-06,
      "loss": 0.81145477,
      "num_input_tokens_seen": 52725465,
      "step": 2460,
      "time_per_iteration": 2.665679693222046
    },
    {
      "auxiliary_loss_clip": 0.01199118,
      "auxiliary_loss_mlp": 0.01031755,
      "balance_loss_clip": 1.06477237,
      "balance_loss_mlp": 1.02155113,
      "epoch": 0.29591775386280283,
      "flos": 13479717352320.0,
      "grad_norm": 2.11198465678162,
      "language_loss": 0.68413806,
      "learning_rate": 3.3031262475788956e-06,
      "loss": 0.70644683,
      "num_input_tokens_seen": 52742405,
      "step": 2461,
      "time_per_iteration": 2.6741538047790527
    },
    {
      "auxiliary_loss_clip": 0.01150771,
      "auxiliary_loss_mlp": 0.01038919,
      "balance_loss_clip": 1.05359745,
      "balance_loss_mlp": 1.02838063,
      "epoch": 0.29603799675344195,
      "flos": 17749783284480.0,
      "grad_norm": 2.124199049253745,
      "language_loss": 0.72938162,
      "learning_rate": 3.3025352252063897e-06,
      "loss": 0.75127852,
      "num_input_tokens_seen": 52761100,
      "step": 2462,
      "time_per_iteration": 2.6747798919677734
    },
    {
      "auxiliary_loss_clip": 0.01122464,
      "auxiliary_loss_mlp": 0.01046809,
      "balance_loss_clip": 1.05285048,
      "balance_loss_mlp": 1.03681338,
      "epoch": 0.29615823964408106,
      "flos": 22783920347520.0,
      "grad_norm": 4.704921976873968,
      "language_loss": 0.75463313,
      "learning_rate": 3.3019440052403252e-06,
      "loss": 0.77632582,
      "num_input_tokens_seen": 52780965,
      "step": 2463,
      "time_per_iteration": 2.8259940147399902
    },
    {
      "auxiliary_loss_clip": 0.01171432,
      "auxiliary_loss_mlp": 0.01039009,
      "balance_loss_clip": 1.06152391,
      "balance_loss_mlp": 1.02944255,
      "epoch": 0.2962784825347201,
      "flos": 23514199758720.0,
      "grad_norm": 2.766246760351656,
      "language_loss": 0.71351361,
      "learning_rate": 3.30135258777039e-06,
      "loss": 0.73561805,
      "num_input_tokens_seen": 52800335,
      "step": 2464,
      "time_per_iteration": 2.7920334339141846
    },
    {
      "auxiliary_loss_clip": 0.01199129,
      "auxiliary_loss_mlp": 0.00764876,
      "balance_loss_clip": 1.06325746,
      "balance_loss_mlp": 1.00079787,
      "epoch": 0.2963987254253592,
      "flos": 16362769559040.0,
      "grad_norm": 2.6815428040543123,
      "language_loss": 0.7031846,
      "learning_rate": 3.3007609728863024e-06,
      "loss": 0.72282463,
      "num_input_tokens_seen": 52818425,
      "step": 2465,
      "time_per_iteration": 2.6108310222625732
    },
    {
      "auxiliary_loss_clip": 0.0116312,
      "auxiliary_loss_mlp": 0.0103865,
      "balance_loss_clip": 1.05930984,
      "balance_loss_mlp": 1.0289526,
      "epoch": 0.29651896831599833,
      "flos": 33472263980160.0,
      "grad_norm": 2.356929310146188,
      "language_loss": 0.72938961,
      "learning_rate": 3.300169160677809e-06,
      "loss": 0.75140733,
      "num_input_tokens_seen": 52842340,
      "step": 2466,
      "time_per_iteration": 2.8607590198516846
    },
    {
      "auxiliary_loss_clip": 0.01148444,
      "auxiliary_loss_mlp": 0.01038038,
      "balance_loss_clip": 1.05443764,
      "balance_loss_mlp": 1.02736902,
      "epoch": 0.2966392112066374,
      "flos": 23805363404160.0,
      "grad_norm": 2.5873295905011995,
      "language_loss": 0.77917331,
      "learning_rate": 3.2995771512346878e-06,
      "loss": 0.80103815,
      "num_input_tokens_seen": 52860690,
      "step": 2467,
      "time_per_iteration": 2.8231234550476074
    },
    {
      "auxiliary_loss_clip": 0.01106801,
      "auxiliary_loss_mlp": 0.0104499,
      "balance_loss_clip": 1.0514524,
      "balance_loss_mlp": 1.03522682,
      "epoch": 0.2967594540972765,
      "flos": 19938466702080.0,
      "grad_norm": 2.883461389374325,
      "language_loss": 0.73048317,
      "learning_rate": 3.298984944646746e-06,
      "loss": 0.75200111,
      "num_input_tokens_seen": 52879370,
      "step": 2468,
      "time_per_iteration": 2.8293440341949463
    },
    {
      "auxiliary_loss_clip": 0.01162483,
      "auxiliary_loss_mlp": 0.01035323,
      "balance_loss_clip": 1.06499314,
      "balance_loss_mlp": 1.02597642,
      "epoch": 0.2968796969879156,
      "flos": 23732823888000.0,
      "grad_norm": 7.870507061535371,
      "language_loss": 0.81734419,
      "learning_rate": 3.298392541003822e-06,
      "loss": 0.83932221,
      "num_input_tokens_seen": 52898775,
      "step": 2469,
      "time_per_iteration": 2.758204221725464
    },
    {
      "auxiliary_loss_clip": 0.01158758,
      "auxiliary_loss_mlp": 0.01035024,
      "balance_loss_clip": 1.05806315,
      "balance_loss_mlp": 1.02535057,
      "epoch": 0.29699993987855466,
      "flos": 22893699288960.0,
      "grad_norm": 1.7306580427814668,
      "language_loss": 0.8963204,
      "learning_rate": 3.2977999403957806e-06,
      "loss": 0.91825819,
      "num_input_tokens_seen": 52917535,
      "step": 2470,
      "time_per_iteration": 2.7485008239746094
    },
    {
      "auxiliary_loss_clip": 0.01209604,
      "auxiliary_loss_mlp": 0.01034566,
      "balance_loss_clip": 1.06709754,
      "balance_loss_mlp": 1.02432609,
      "epoch": 0.2971201827691938,
      "flos": 33832555349760.0,
      "grad_norm": 5.0670593991045525,
      "language_loss": 0.6713357,
      "learning_rate": 3.2972071429125207e-06,
      "loss": 0.69377738,
      "num_input_tokens_seen": 52938755,
      "step": 2471,
      "time_per_iteration": 2.764099597930908
    },
    {
      "auxiliary_loss_clip": 0.01172603,
      "auxiliary_loss_mlp": 0.01035992,
      "balance_loss_clip": 1.05987263,
      "balance_loss_mlp": 1.02602589,
      "epoch": 0.2972404256598329,
      "flos": 22054359208320.0,
      "grad_norm": 3.825278477424707,
      "language_loss": 0.88720036,
      "learning_rate": 3.2966141486439682e-06,
      "loss": 0.90928632,
      "num_input_tokens_seen": 52957945,
      "step": 2472,
      "time_per_iteration": 2.75862979888916
    },
    {
      "auxiliary_loss_clip": 0.01156775,
      "auxiliary_loss_mlp": 0.0103434,
      "balance_loss_clip": 1.05735397,
      "balance_loss_mlp": 1.02441549,
      "epoch": 0.29736066855047194,
      "flos": 31978595796480.0,
      "grad_norm": 2.418504707680682,
      "language_loss": 0.64086521,
      "learning_rate": 3.29602095768008e-06,
      "loss": 0.66277629,
      "num_input_tokens_seen": 52978460,
      "step": 2473,
      "time_per_iteration": 2.8573765754699707
    },
    {
      "auxiliary_loss_clip": 0.01175989,
      "auxiliary_loss_mlp": 0.01033245,
      "balance_loss_clip": 1.05883622,
      "balance_loss_mlp": 1.02374434,
      "epoch": 0.29748091144111105,
      "flos": 33510401245440.0,
      "grad_norm": 2.0385426901738493,
      "language_loss": 0.63939834,
      "learning_rate": 3.2954275701108437e-06,
      "loss": 0.66149068,
      "num_input_tokens_seen": 52999640,
      "step": 2474,
      "time_per_iteration": 2.8104641437530518
    },
    {
      "auxiliary_loss_clip": 0.01148273,
      "auxiliary_loss_mlp": 0.01037398,
      "balance_loss_clip": 1.05859852,
      "balance_loss_mlp": 1.02791536,
      "epoch": 0.29760115433175016,
      "flos": 41283373409280.0,
      "grad_norm": 3.4786808521604113,
      "language_loss": 0.68837589,
      "learning_rate": 3.294833986026275e-06,
      "loss": 0.71023262,
      "num_input_tokens_seen": 53022880,
      "step": 2475,
      "time_per_iteration": 2.920926570892334
    },
    {
      "auxiliary_loss_clip": 0.01145065,
      "auxiliary_loss_mlp": 0.00764018,
      "balance_loss_clip": 1.05771327,
      "balance_loss_mlp": 1.0007596,
      "epoch": 0.2977213972223892,
      "flos": 24493339572480.0,
      "grad_norm": 2.2107987037213324,
      "language_loss": 0.85285664,
      "learning_rate": 3.29424020551642e-06,
      "loss": 0.87194753,
      "num_input_tokens_seen": 53041515,
      "step": 2476,
      "time_per_iteration": 3.683598041534424
    },
    {
      "auxiliary_loss_clip": 0.01164276,
      "auxiliary_loss_mlp": 0.01034433,
      "balance_loss_clip": 1.06133795,
      "balance_loss_mlp": 1.02344215,
      "epoch": 0.2978416401130283,
      "flos": 21285116519040.0,
      "grad_norm": 7.527673217250313,
      "language_loss": 0.72007537,
      "learning_rate": 3.2936462286713546e-06,
      "loss": 0.74206245,
      "num_input_tokens_seen": 53059865,
      "step": 2477,
      "time_per_iteration": 3.766331672668457
    },
    {
      "auxiliary_loss_clip": 0.01189304,
      "auxiliary_loss_mlp": 0.01032933,
      "balance_loss_clip": 1.06271172,
      "balance_loss_mlp": 1.02206707,
      "epoch": 0.2979618830036674,
      "flos": 25772154554880.0,
      "grad_norm": 2.134265605248342,
      "language_loss": 0.77032697,
      "learning_rate": 3.2930520555811846e-06,
      "loss": 0.79254931,
      "num_input_tokens_seen": 53079490,
      "step": 2478,
      "time_per_iteration": 2.8114326000213623
    },
    {
      "auxiliary_loss_clip": 0.01148746,
      "auxiliary_loss_mlp": 0.01042117,
      "balance_loss_clip": 1.05663347,
      "balance_loss_mlp": 1.0321753,
      "epoch": 0.2980821258943065,
      "flos": 23476996247040.0,
      "grad_norm": 3.09199285477796,
      "language_loss": 0.80347377,
      "learning_rate": 3.292457686336046e-06,
      "loss": 0.82538241,
      "num_input_tokens_seen": 53098810,
      "step": 2479,
      "time_per_iteration": 2.7908875942230225
    },
    {
      "auxiliary_loss_clip": 0.01083042,
      "auxiliary_loss_mlp": 0.01002056,
      "balance_loss_clip": 1.02689815,
      "balance_loss_mlp": 1.00036335,
      "epoch": 0.2982023687849456,
      "flos": 69752314195200.0,
      "grad_norm": 0.8766772287962142,
      "language_loss": 0.61222756,
      "learning_rate": 3.291863121026105e-06,
      "loss": 0.63307858,
      "num_input_tokens_seen": 53162590,
      "step": 2480,
      "time_per_iteration": 4.3194780349731445
    },
    {
      "auxiliary_loss_clip": 0.01153593,
      "auxiliary_loss_mlp": 0.01039464,
      "balance_loss_clip": 1.06151676,
      "balance_loss_mlp": 1.02948642,
      "epoch": 0.29832261167558466,
      "flos": 29825930741760.0,
      "grad_norm": 2.2519939439948726,
      "language_loss": 0.76712978,
      "learning_rate": 3.2912683597415547e-06,
      "loss": 0.78906035,
      "num_input_tokens_seen": 53186675,
      "step": 2481,
      "time_per_iteration": 3.759535312652588
    },
    {
      "auxiliary_loss_clip": 0.01126147,
      "auxiliary_loss_mlp": 0.01045996,
      "balance_loss_clip": 1.05283546,
      "balance_loss_mlp": 1.03591061,
      "epoch": 0.29844285456622377,
      "flos": 33910158683520.0,
      "grad_norm": 2.0193696429404353,
      "language_loss": 0.78273594,
      "learning_rate": 3.2906734025726213e-06,
      "loss": 0.80445737,
      "num_input_tokens_seen": 53205940,
      "step": 2482,
      "time_per_iteration": 2.886530876159668
    },
    {
      "auxiliary_loss_clip": 0.01206823,
      "auxiliary_loss_mlp": 0.01045575,
      "balance_loss_clip": 1.06887472,
      "balance_loss_mlp": 1.03481615,
      "epoch": 0.2985630974568629,
      "flos": 23876933253120.0,
      "grad_norm": 4.724469646677941,
      "language_loss": 0.88106561,
      "learning_rate": 3.290078249609559e-06,
      "loss": 0.90358961,
      "num_input_tokens_seen": 53225360,
      "step": 2483,
      "time_per_iteration": 2.7108418941497803
    },
    {
      "auxiliary_loss_clip": 0.0121693,
      "auxiliary_loss_mlp": 0.01039594,
      "balance_loss_clip": 1.06898808,
      "balance_loss_mlp": 1.03027821,
      "epoch": 0.29868334034750194,
      "flos": 21799106184960.0,
      "grad_norm": 3.109516433175004,
      "language_loss": 0.88181746,
      "learning_rate": 3.2894829009426514e-06,
      "loss": 0.90438271,
      "num_input_tokens_seen": 53243195,
      "step": 2484,
      "time_per_iteration": 2.6640801429748535
    },
    {
      "auxiliary_loss_clip": 0.01192636,
      "auxiliary_loss_mlp": 0.01039749,
      "balance_loss_clip": 1.06289196,
      "balance_loss_mlp": 1.03048015,
      "epoch": 0.29880358323814105,
      "flos": 25666649331840.0,
      "grad_norm": 4.636399087022581,
      "language_loss": 0.77656341,
      "learning_rate": 3.288887356662213e-06,
      "loss": 0.79888725,
      "num_input_tokens_seen": 53264530,
      "step": 2485,
      "time_per_iteration": 2.697113513946533
    },
    {
      "auxiliary_loss_clip": 0.01076781,
      "auxiliary_loss_mlp": 0.01003482,
      "balance_loss_clip": 1.02258134,
      "balance_loss_mlp": 1.00167024,
      "epoch": 0.29892382612878016,
      "flos": 71005846003200.0,
      "grad_norm": 0.7703783248675532,
      "language_loss": 0.59608054,
      "learning_rate": 3.288291616858588e-06,
      "loss": 0.61688316,
      "num_input_tokens_seen": 53319920,
      "step": 2486,
      "time_per_iteration": 3.195819139480591
    },
    {
      "auxiliary_loss_clip": 0.01161297,
      "auxiliary_loss_mlp": 0.01043309,
      "balance_loss_clip": 1.05635333,
      "balance_loss_mlp": 1.03377807,
      "epoch": 0.2990440690194192,
      "flos": 25481134563840.0,
      "grad_norm": 3.7072941322655133,
      "language_loss": 0.76895785,
      "learning_rate": 3.287695681622149e-06,
      "loss": 0.79100394,
      "num_input_tokens_seen": 53339270,
      "step": 2487,
      "time_per_iteration": 2.726757287979126
    },
    {
      "auxiliary_loss_clip": 0.01147687,
      "auxiliary_loss_mlp": 0.01028864,
      "balance_loss_clip": 1.05431223,
      "balance_loss_mlp": 1.01889789,
      "epoch": 0.2991643119100583,
      "flos": 23732357011200.0,
      "grad_norm": 4.915675334419143,
      "language_loss": 0.80791301,
      "learning_rate": 3.2870995510432982e-06,
      "loss": 0.82967854,
      "num_input_tokens_seen": 53357750,
      "step": 2488,
      "time_per_iteration": 2.7753169536590576
    },
    {
      "auxiliary_loss_clip": 0.01140251,
      "auxiliary_loss_mlp": 0.01032364,
      "balance_loss_clip": 1.05275512,
      "balance_loss_mlp": 1.02371526,
      "epoch": 0.29928455480069743,
      "flos": 27417545786880.0,
      "grad_norm": 2.48271141317864,
      "language_loss": 0.77048767,
      "learning_rate": 3.2865032252124697e-06,
      "loss": 0.79221386,
      "num_input_tokens_seen": 53378265,
      "step": 2489,
      "time_per_iteration": 2.7696211338043213
    },
    {
      "auxiliary_loss_clip": 0.01201016,
      "auxiliary_loss_mlp": 0.00763319,
      "balance_loss_clip": 1.06755924,
      "balance_loss_mlp": 1.00066805,
      "epoch": 0.2994047976913365,
      "flos": 33692935184640.0,
      "grad_norm": 2.1663648029849734,
      "language_loss": 0.77670032,
      "learning_rate": 3.2859067042201243e-06,
      "loss": 0.79634368,
      "num_input_tokens_seen": 53400305,
      "step": 2490,
      "time_per_iteration": 2.8206515312194824
    },
    {
      "auxiliary_loss_clip": 0.01171502,
      "auxiliary_loss_mlp": 0.01031679,
      "balance_loss_clip": 1.05961871,
      "balance_loss_mlp": 1.02289319,
      "epoch": 0.2995250405819756,
      "flos": 16763963541120.0,
      "grad_norm": 3.369628748756632,
      "language_loss": 0.78212559,
      "learning_rate": 3.2853099881567544e-06,
      "loss": 0.80415738,
      "num_input_tokens_seen": 53418705,
      "step": 2491,
      "time_per_iteration": 2.7472875118255615
    },
    {
      "auxiliary_loss_clip": 0.01152321,
      "auxiliary_loss_mlp": 0.0103218,
      "balance_loss_clip": 1.05984712,
      "balance_loss_mlp": 1.02315629,
      "epoch": 0.29964528347261465,
      "flos": 22963976248320.0,
      "grad_norm": 2.6175099888913147,
      "language_loss": 0.79387343,
      "learning_rate": 3.284713077112881e-06,
      "loss": 0.81571841,
      "num_input_tokens_seen": 53438135,
      "step": 2492,
      "time_per_iteration": 2.8114173412323
    },
    {
      "auxiliary_loss_clip": 0.01222686,
      "auxiliary_loss_mlp": 0.01036984,
      "balance_loss_clip": 1.07125425,
      "balance_loss_mlp": 1.0266006,
      "epoch": 0.29976552636325376,
      "flos": 16938021870720.0,
      "grad_norm": 7.506426220851604,
      "language_loss": 0.86844039,
      "learning_rate": 3.284115971179056e-06,
      "loss": 0.89103711,
      "num_input_tokens_seen": 53452165,
      "step": 2493,
      "time_per_iteration": 2.682318687438965
    },
    {
      "auxiliary_loss_clip": 0.01186558,
      "auxiliary_loss_mlp": 0.01039703,
      "balance_loss_clip": 1.06519961,
      "balance_loss_mlp": 1.0302316,
      "epoch": 0.2998857692538929,
      "flos": 17056455989760.0,
      "grad_norm": 2.4166442935390577,
      "language_loss": 0.78306389,
      "learning_rate": 3.283518670445859e-06,
      "loss": 0.80532646,
      "num_input_tokens_seen": 53470075,
      "step": 2494,
      "time_per_iteration": 2.7920308113098145
    },
    {
      "auxiliary_loss_clip": 0.01089746,
      "auxiliary_loss_mlp": 0.01005803,
      "balance_loss_clip": 1.02476788,
      "balance_loss_mlp": 1.00400257,
      "epoch": 0.30000601214453193,
      "flos": 68831528025600.0,
      "grad_norm": 0.6966671298212611,
      "language_loss": 0.54298389,
      "learning_rate": 3.2829211750038995e-06,
      "loss": 0.56393939,
      "num_input_tokens_seen": 53538705,
      "step": 2495,
      "time_per_iteration": 3.371133327484131
    },
    {
      "auxiliary_loss_clip": 0.01176048,
      "auxiliary_loss_mlp": 0.01031612,
      "balance_loss_clip": 1.05864465,
      "balance_loss_mlp": 1.02218843,
      "epoch": 0.30012625503517104,
      "flos": 17603267708160.0,
      "grad_norm": 3.203051008664324,
      "language_loss": 0.89137673,
      "learning_rate": 3.2823234849438183e-06,
      "loss": 0.91345334,
      "num_input_tokens_seen": 53556740,
      "step": 2496,
      "time_per_iteration": 2.6669907569885254
    },
    {
      "auxiliary_loss_clip": 0.01171498,
      "auxiliary_loss_mlp": 0.01033201,
      "balance_loss_clip": 1.06063724,
      "balance_loss_mlp": 1.02403331,
      "epoch": 0.30024649792581015,
      "flos": 21252581775360.0,
      "grad_norm": 2.4013452428596986,
      "language_loss": 0.76006877,
      "learning_rate": 3.2817256003562836e-06,
      "loss": 0.78211582,
      "num_input_tokens_seen": 53577115,
      "step": 2497,
      "time_per_iteration": 2.7516000270843506
    },
    {
      "auxiliary_loss_clip": 0.0119194,
      "auxiliary_loss_mlp": 0.01037102,
      "balance_loss_clip": 1.06506658,
      "balance_loss_mlp": 1.02704644,
      "epoch": 0.3003667408164492,
      "flos": 23003262748800.0,
      "grad_norm": 2.4845756067594906,
      "language_loss": 0.65869725,
      "learning_rate": 3.281127521331995e-06,
      "loss": 0.68098772,
      "num_input_tokens_seen": 53598295,
      "step": 2498,
      "time_per_iteration": 2.762742042541504
    },
    {
      "auxiliary_loss_clip": 0.0107476,
      "auxiliary_loss_mlp": 0.01026931,
      "balance_loss_clip": 1.03735876,
      "balance_loss_mlp": 1.02526212,
      "epoch": 0.3004869837070883,
      "flos": 64232340750720.0,
      "grad_norm": 0.8910002756714612,
      "language_loss": 0.60686839,
      "learning_rate": 3.2805292479616798e-06,
      "loss": 0.62788522,
      "num_input_tokens_seen": 53657160,
      "step": 2499,
      "time_per_iteration": 3.288325309753418
    },
    {
      "auxiliary_loss_clip": 0.01180459,
      "auxiliary_loss_mlp": 0.0104014,
      "balance_loss_clip": 1.05998623,
      "balance_loss_mlp": 1.03172982,
      "epoch": 0.30060722659772743,
      "flos": 26248653400320.0,
      "grad_norm": 3.0779698163230633,
      "language_loss": 0.9174087,
      "learning_rate": 3.2799307803360955e-06,
      "loss": 0.93961465,
      "num_input_tokens_seen": 53673090,
      "step": 2500,
      "time_per_iteration": 2.768329381942749
    },
    {
      "auxiliary_loss_clip": 0.01200898,
      "auxiliary_loss_mlp": 0.01034844,
      "balance_loss_clip": 1.06271422,
      "balance_loss_mlp": 1.02592158,
      "epoch": 0.3007274694883665,
      "flos": 24970879912320.0,
      "grad_norm": 1.5264553106289724,
      "language_loss": 0.81390309,
      "learning_rate": 3.27933211854603e-06,
      "loss": 0.83626056,
      "num_input_tokens_seen": 53692145,
      "step": 2501,
      "time_per_iteration": 2.695216655731201
    },
    {
      "auxiliary_loss_clip": 0.01142461,
      "auxiliary_loss_mlp": 0.01039211,
      "balance_loss_clip": 1.05194187,
      "balance_loss_mlp": 1.02945971,
      "epoch": 0.3008477123790056,
      "flos": 17055845458560.0,
      "grad_norm": 2.3714403862391435,
      "language_loss": 0.86860025,
      "learning_rate": 3.278733262682299e-06,
      "loss": 0.89041698,
      "num_input_tokens_seen": 53710000,
      "step": 2502,
      "time_per_iteration": 3.712397813796997
    },
    {
      "auxiliary_loss_clip": 0.01160453,
      "auxiliary_loss_mlp": 0.01044262,
      "balance_loss_clip": 1.05377913,
      "balance_loss_mlp": 1.03525007,
      "epoch": 0.3009679552696447,
      "flos": 21506398254720.0,
      "grad_norm": 3.5088560430733007,
      "language_loss": 0.8225106,
      "learning_rate": 3.2781342128357484e-06,
      "loss": 0.84455776,
      "num_input_tokens_seen": 53729355,
      "step": 2503,
      "time_per_iteration": 3.686117649078369
    },
    {
      "auxiliary_loss_clip": 0.01186122,
      "auxiliary_loss_mlp": 0.01038508,
      "balance_loss_clip": 1.06517553,
      "balance_loss_mlp": 1.02943063,
      "epoch": 0.30108819816028376,
      "flos": 21134004001920.0,
      "grad_norm": 2.8017196348622817,
      "language_loss": 0.80389428,
      "learning_rate": 3.2775349690972547e-06,
      "loss": 0.82614064,
      "num_input_tokens_seen": 53743505,
      "step": 2504,
      "time_per_iteration": 2.6917521953582764
    },
    {
      "auxiliary_loss_clip": 0.01074423,
      "auxiliary_loss_mlp": 0.01008042,
      "balance_loss_clip": 1.02249956,
      "balance_loss_mlp": 1.00633705,
      "epoch": 0.30120844105092287,
      "flos": 71126434938240.0,
      "grad_norm": 0.7562702788602668,
      "language_loss": 0.51805222,
      "learning_rate": 3.276935531557722e-06,
      "loss": 0.53887677,
      "num_input_tokens_seen": 53808725,
      "step": 2505,
      "time_per_iteration": 3.37138295173645
    },
    {
      "auxiliary_loss_clip": 0.01187844,
      "auxiliary_loss_mlp": 0.0103597,
      "balance_loss_clip": 1.06451583,
      "balance_loss_mlp": 1.02673101,
      "epoch": 0.301328683941562,
      "flos": 20264571302400.0,
      "grad_norm": 4.262180627667278,
      "language_loss": 0.79879177,
      "learning_rate": 3.2763359003080837e-06,
      "loss": 0.82102996,
      "num_input_tokens_seen": 53825680,
      "step": 2506,
      "time_per_iteration": 3.6970391273498535
    },
    {
      "auxiliary_loss_clip": 0.0109017,
      "auxiliary_loss_mlp": 0.00754896,
      "balance_loss_clip": 1.0230521,
      "balance_loss_mlp": 1.00108933,
      "epoch": 0.30144892683220104,
      "flos": 70648212240000.0,
      "grad_norm": 0.8334649879595509,
      "language_loss": 0.62439841,
      "learning_rate": 3.2757360754393047e-06,
      "loss": 0.64284909,
      "num_input_tokens_seen": 53889750,
      "step": 2507,
      "time_per_iteration": 4.317520380020142
    },
    {
      "auxiliary_loss_clip": 0.01138785,
      "auxiliary_loss_mlp": 0.01041846,
      "balance_loss_clip": 1.05396175,
      "balance_loss_mlp": 1.03140962,
      "epoch": 0.30156916972284015,
      "flos": 22820549241600.0,
      "grad_norm": 5.4454531233328725,
      "language_loss": 0.64245945,
      "learning_rate": 3.2751360570423767e-06,
      "loss": 0.66426575,
      "num_input_tokens_seen": 53908135,
      "step": 2508,
      "time_per_iteration": 2.791762351989746
    },
    {
      "auxiliary_loss_clip": 0.01181859,
      "auxiliary_loss_mlp": 0.01034095,
      "balance_loss_clip": 1.05692506,
      "balance_loss_mlp": 1.02430797,
      "epoch": 0.3016894126134792,
      "flos": 29899188529920.0,
      "grad_norm": 2.187655372626552,
      "language_loss": 0.75827599,
      "learning_rate": 3.2745358452083236e-06,
      "loss": 0.78043556,
      "num_input_tokens_seen": 53931035,
      "step": 2509,
      "time_per_iteration": 2.8038408756256104
    },
    {
      "auxiliary_loss_clip": 0.0116323,
      "auxiliary_loss_mlp": 0.0103814,
      "balance_loss_clip": 1.06073451,
      "balance_loss_mlp": 1.02908015,
      "epoch": 0.3018096555041183,
      "flos": 21546331200000.0,
      "grad_norm": 2.104225640801367,
      "language_loss": 0.82415259,
      "learning_rate": 3.2739354400281955e-06,
      "loss": 0.84616631,
      "num_input_tokens_seen": 53952255,
      "step": 2510,
      "time_per_iteration": 2.7398948669433594
    },
    {
      "auxiliary_loss_clip": 0.01087279,
      "auxiliary_loss_mlp": 0.01004336,
      "balance_loss_clip": 1.02169847,
      "balance_loss_mlp": 1.00263107,
      "epoch": 0.3019298983947574,
      "flos": 59136294597120.0,
      "grad_norm": 0.869529949799666,
      "language_loss": 0.63613582,
      "learning_rate": 3.2733348415930744e-06,
      "loss": 0.65705204,
      "num_input_tokens_seen": 54014125,
      "step": 2511,
      "time_per_iteration": 3.259089708328247
    },
    {
      "auxiliary_loss_clip": 0.01150822,
      "auxiliary_loss_mlp": 0.0103429,
      "balance_loss_clip": 1.05701578,
      "balance_loss_mlp": 1.02504539,
      "epoch": 0.3020501412853965,
      "flos": 34423070941440.0,
      "grad_norm": 2.6635853962409515,
      "language_loss": 0.80884916,
      "learning_rate": 3.27273404999407e-06,
      "loss": 0.83070034,
      "num_input_tokens_seen": 54036345,
      "step": 2512,
      "time_per_iteration": 2.859686851501465
    },
    {
      "auxiliary_loss_clip": 0.01095284,
      "auxiliary_loss_mlp": 0.0100421,
      "balance_loss_clip": 1.02259433,
      "balance_loss_mlp": 1.00251698,
      "epoch": 0.3021703841760356,
      "flos": 71008288128000.0,
      "grad_norm": 0.808132251349923,
      "language_loss": 0.60387385,
      "learning_rate": 3.272133065322322e-06,
      "loss": 0.62486875,
      "num_input_tokens_seen": 54094615,
      "step": 2513,
      "time_per_iteration": 3.233365774154663
    },
    {
      "auxiliary_loss_clip": 0.01199214,
      "auxiliary_loss_mlp": 0.01036679,
      "balance_loss_clip": 1.06270683,
      "balance_loss_mlp": 1.02725565,
      "epoch": 0.3022906270666747,
      "flos": 21510528318720.0,
      "grad_norm": 1.6717842996531753,
      "language_loss": 0.79594076,
      "learning_rate": 3.271531887669e-06,
      "loss": 0.81829971,
      "num_input_tokens_seen": 54114675,
      "step": 2514,
      "time_per_iteration": 2.741175413131714
    },
    {
      "auxiliary_loss_clip": 0.01186095,
      "auxiliary_loss_mlp": 0.01029653,
      "balance_loss_clip": 1.05925083,
      "balance_loss_mlp": 1.01949596,
      "epoch": 0.30241086995731375,
      "flos": 31132001168640.0,
      "grad_norm": 5.644851702937754,
      "language_loss": 0.63540614,
      "learning_rate": 3.2709305171253015e-06,
      "loss": 0.65756363,
      "num_input_tokens_seen": 54134795,
      "step": 2515,
      "time_per_iteration": 2.7375423908233643
    },
    {
      "auxiliary_loss_clip": 0.01124968,
      "auxiliary_loss_mlp": 0.01036395,
      "balance_loss_clip": 1.05363727,
      "balance_loss_mlp": 1.0272398,
      "epoch": 0.30253111284795287,
      "flos": 23511542152320.0,
      "grad_norm": 3.1296275616358296,
      "language_loss": 0.78097844,
      "learning_rate": 3.2703289537824536e-06,
      "loss": 0.80259204,
      "num_input_tokens_seen": 54154595,
      "step": 2516,
      "time_per_iteration": 2.8446061611175537
    },
    {
      "auxiliary_loss_clip": 0.0116688,
      "auxiliary_loss_mlp": 0.01039687,
      "balance_loss_clip": 1.05282092,
      "balance_loss_mlp": 1.03050208,
      "epoch": 0.302651355738592,
      "flos": 18725367651840.0,
      "grad_norm": 5.327933526178594,
      "language_loss": 0.78903711,
      "learning_rate": 3.269727197731714e-06,
      "loss": 0.81110281,
      "num_input_tokens_seen": 54167360,
      "step": 2517,
      "time_per_iteration": 2.684258460998535
    },
    {
      "auxiliary_loss_clip": 0.01196077,
      "auxiliary_loss_mlp": 0.01036347,
      "balance_loss_clip": 1.06230271,
      "balance_loss_mlp": 1.02680409,
      "epoch": 0.30277159862923103,
      "flos": 22418888382720.0,
      "grad_norm": 1.7134725301165596,
      "language_loss": 0.77807879,
      "learning_rate": 3.269125249064367e-06,
      "loss": 0.800403,
      "num_input_tokens_seen": 54187055,
      "step": 2518,
      "time_per_iteration": 2.656874656677246
    },
    {
      "auxiliary_loss_clip": 0.01158305,
      "auxiliary_loss_mlp": 0.01036116,
      "balance_loss_clip": 1.05453229,
      "balance_loss_mlp": 1.0266571,
      "epoch": 0.30289184151987014,
      "flos": 22273126992000.0,
      "grad_norm": 3.389402340914403,
      "language_loss": 0.83322048,
      "learning_rate": 3.2685231078717297e-06,
      "loss": 0.85516465,
      "num_input_tokens_seen": 54207245,
      "step": 2519,
      "time_per_iteration": 2.741889238357544
    },
    {
      "auxiliary_loss_clip": 0.01135624,
      "auxiliary_loss_mlp": 0.01039804,
      "balance_loss_clip": 1.05242133,
      "balance_loss_mlp": 1.02995729,
      "epoch": 0.30301208441050925,
      "flos": 25225594231680.0,
      "grad_norm": 3.7280182081765854,
      "language_loss": 0.7530756,
      "learning_rate": 3.267920774245145e-06,
      "loss": 0.77482992,
      "num_input_tokens_seen": 54226650,
      "step": 2520,
      "time_per_iteration": 2.8480422496795654
    },
    {
      "auxiliary_loss_clip": 0.01176514,
      "auxiliary_loss_mlp": 0.01038495,
      "balance_loss_clip": 1.06054592,
      "balance_loss_mlp": 1.02891612,
      "epoch": 0.3031323273011483,
      "flos": 23039245198080.0,
      "grad_norm": 2.6212551274590528,
      "language_loss": 0.84741068,
      "learning_rate": 3.2673182482759876e-06,
      "loss": 0.86956078,
      "num_input_tokens_seen": 54245765,
      "step": 2521,
      "time_per_iteration": 2.745957136154175
    },
    {
      "auxiliary_loss_clip": 0.01134845,
      "auxiliary_loss_mlp": 0.01030812,
      "balance_loss_clip": 1.05810666,
      "balance_loss_mlp": 1.02054834,
      "epoch": 0.3032525701917874,
      "flos": 18876695650560.0,
      "grad_norm": 3.5605480613961293,
      "language_loss": 0.66582334,
      "learning_rate": 3.266715530055659e-06,
      "loss": 0.68747985,
      "num_input_tokens_seen": 54263915,
      "step": 2522,
      "time_per_iteration": 2.8991503715515137
    },
    {
      "auxiliary_loss_clip": 0.0121337,
      "auxiliary_loss_mlp": 0.01035155,
      "balance_loss_clip": 1.06445384,
      "balance_loss_mlp": 1.0251472,
      "epoch": 0.30337281308242653,
      "flos": 17782641250560.0,
      "grad_norm": 1.7118962929054864,
      "language_loss": 0.80557108,
      "learning_rate": 3.2661126196755927e-06,
      "loss": 0.82805634,
      "num_input_tokens_seen": 54283025,
      "step": 2523,
      "time_per_iteration": 2.657116174697876
    },
    {
      "auxiliary_loss_clip": 0.01093191,
      "auxiliary_loss_mlp": 0.01013636,
      "balance_loss_clip": 1.02094352,
      "balance_loss_mlp": 1.01200283,
      "epoch": 0.3034930559730656,
      "flos": 57824298426240.0,
      "grad_norm": 0.7861565301429562,
      "language_loss": 0.55935925,
      "learning_rate": 3.265509517227248e-06,
      "loss": 0.58042753,
      "num_input_tokens_seen": 54339840,
      "step": 2524,
      "time_per_iteration": 3.2481040954589844
    },
    {
      "auxiliary_loss_clip": 0.01152157,
      "auxiliary_loss_mlp": 0.01038506,
      "balance_loss_clip": 1.05519795,
      "balance_loss_mlp": 1.02999473,
      "epoch": 0.3036132988637047,
      "flos": 14755587419520.0,
      "grad_norm": 2.191003413819213,
      "language_loss": 0.80462646,
      "learning_rate": 3.264906222802115e-06,
      "loss": 0.82653314,
      "num_input_tokens_seen": 54357690,
      "step": 2525,
      "time_per_iteration": 2.7761154174804688
    },
    {
      "auxiliary_loss_clip": 0.01201629,
      "auxiliary_loss_mlp": 0.01029687,
      "balance_loss_clip": 1.06212914,
      "balance_loss_mlp": 1.02003706,
      "epoch": 0.30373354175434375,
      "flos": 21033203460480.0,
      "grad_norm": 10.629853351060069,
      "language_loss": 0.77871674,
      "learning_rate": 3.264302736491715e-06,
      "loss": 0.80102992,
      "num_input_tokens_seen": 54377810,
      "step": 2526,
      "time_per_iteration": 2.6224160194396973
    },
    {
      "auxiliary_loss_clip": 0.01180066,
      "auxiliary_loss_mlp": 0.01036602,
      "balance_loss_clip": 1.06335616,
      "balance_loss_mlp": 1.02708316,
      "epoch": 0.30385378464498286,
      "flos": 21143233797120.0,
      "grad_norm": 3.0808349349439776,
      "language_loss": 0.87646699,
      "learning_rate": 3.263699058387594e-06,
      "loss": 0.89863366,
      "num_input_tokens_seen": 54395245,
      "step": 2527,
      "time_per_iteration": 2.7140204906463623
    },
    {
      "auxiliary_loss_clip": 0.01128443,
      "auxiliary_loss_mlp": 0.0103719,
      "balance_loss_clip": 1.05529916,
      "balance_loss_mlp": 1.02704477,
      "epoch": 0.30397402753562197,
      "flos": 20629244131200.0,
      "grad_norm": 12.41983417878483,
      "language_loss": 0.89996386,
      "learning_rate": 3.2630951885813315e-06,
      "loss": 0.92162013,
      "num_input_tokens_seen": 54412640,
      "step": 2528,
      "time_per_iteration": 3.751495122909546
    },
    {
      "auxiliary_loss_clip": 0.01164288,
      "auxiliary_loss_mlp": 0.01033724,
      "balance_loss_clip": 1.05744028,
      "balance_loss_mlp": 1.0237999,
      "epoch": 0.304094270426261,
      "flos": 15085678429440.0,
      "grad_norm": 2.711709434601873,
      "language_loss": 0.78559035,
      "learning_rate": 3.262491127164533e-06,
      "loss": 0.80757046,
      "num_input_tokens_seen": 54431455,
      "step": 2529,
      "time_per_iteration": 3.6863720417022705
    },
    {
      "auxiliary_loss_clip": 0.0112891,
      "auxiliary_loss_mlp": 0.01033261,
      "balance_loss_clip": 1.04950905,
      "balance_loss_mlp": 1.02284193,
      "epoch": 0.30421451331690014,
      "flos": 13845216193920.0,
      "grad_norm": 2.873462243968574,
      "language_loss": 0.80542696,
      "learning_rate": 3.2618868742288337e-06,
      "loss": 0.82704866,
      "num_input_tokens_seen": 54448380,
      "step": 2530,
      "time_per_iteration": 2.849853754043579
    },
    {
      "auxiliary_loss_clip": 0.01160436,
      "auxiliary_loss_mlp": 0.01044051,
      "balance_loss_clip": 1.05502701,
      "balance_loss_mlp": 1.03499675,
      "epoch": 0.30433475620753925,
      "flos": 17384212615680.0,
      "grad_norm": 2.019303472476211,
      "language_loss": 0.72386658,
      "learning_rate": 3.261282429865899e-06,
      "loss": 0.74591148,
      "num_input_tokens_seen": 54466385,
      "step": 2531,
      "time_per_iteration": 3.6361050605773926
    },
    {
      "auxiliary_loss_clip": 0.0118381,
      "auxiliary_loss_mlp": 0.01038213,
      "balance_loss_clip": 1.06047583,
      "balance_loss_mlp": 1.0287056,
      "epoch": 0.3044549990981783,
      "flos": 18916951818240.0,
      "grad_norm": 1.9531275409228483,
      "language_loss": 0.72640419,
      "learning_rate": 3.2606777941674225e-06,
      "loss": 0.74862438,
      "num_input_tokens_seen": 54485040,
      "step": 2532,
      "time_per_iteration": 2.650459051132202
    },
    {
      "auxiliary_loss_clip": 0.0114197,
      "auxiliary_loss_mlp": 0.01033449,
      "balance_loss_clip": 1.05554461,
      "balance_loss_mlp": 1.02431798,
      "epoch": 0.3045752419888174,
      "flos": 21068431724160.0,
      "grad_norm": 2.702513834918221,
      "language_loss": 0.84873056,
      "learning_rate": 3.2600729672251276e-06,
      "loss": 0.87048471,
      "num_input_tokens_seen": 54502755,
      "step": 2533,
      "time_per_iteration": 3.706951141357422
    },
    {
      "auxiliary_loss_clip": 0.01188964,
      "auxiliary_loss_mlp": 0.01030701,
      "balance_loss_clip": 1.06500649,
      "balance_loss_mlp": 1.02069926,
      "epoch": 0.3046954848794565,
      "flos": 29096405516160.0,
      "grad_norm": 2.3526712480024163,
      "language_loss": 0.65486252,
      "learning_rate": 3.259467949130765e-06,
      "loss": 0.67705917,
      "num_input_tokens_seen": 54524165,
      "step": 2534,
      "time_per_iteration": 2.777273654937744
    },
    {
      "auxiliary_loss_clip": 0.01185172,
      "auxiliary_loss_mlp": 0.01039636,
      "balance_loss_clip": 1.05821562,
      "balance_loss_mlp": 1.02990818,
      "epoch": 0.3048157277700956,
      "flos": 20295346279680.0,
      "grad_norm": 2.7616065338750495,
      "language_loss": 0.82371593,
      "learning_rate": 3.2588627399761164e-06,
      "loss": 0.84596401,
      "num_input_tokens_seen": 54540160,
      "step": 2535,
      "time_per_iteration": 2.678421974182129
    },
    {
      "auxiliary_loss_clip": 0.01214243,
      "auxiliary_loss_mlp": 0.01035195,
      "balance_loss_clip": 1.06656551,
      "balance_loss_mlp": 1.02592063,
      "epoch": 0.3049359706607347,
      "flos": 22739929165440.0,
      "grad_norm": 3.2192215963806063,
      "language_loss": 0.71325159,
      "learning_rate": 3.2582573398529903e-06,
      "loss": 0.73574591,
      "num_input_tokens_seen": 54557515,
      "step": 2536,
      "time_per_iteration": 2.615190029144287
    },
    {
      "auxiliary_loss_clip": 0.01174266,
      "auxiliary_loss_mlp": 0.01044096,
      "balance_loss_clip": 1.06025934,
      "balance_loss_mlp": 1.03448784,
      "epoch": 0.3050562135513738,
      "flos": 18434634969600.0,
      "grad_norm": 2.9026459527933537,
      "language_loss": 0.73863709,
      "learning_rate": 3.2576517488532265e-06,
      "loss": 0.76082075,
      "num_input_tokens_seen": 54573865,
      "step": 2537,
      "time_per_iteration": 2.633841037750244
    },
    {
      "auxiliary_loss_clip": 0.01196527,
      "auxiliary_loss_mlp": 0.01037241,
      "balance_loss_clip": 1.06052816,
      "balance_loss_mlp": 1.02784133,
      "epoch": 0.30517645644201286,
      "flos": 20370327920640.0,
      "grad_norm": 1.867657133685427,
      "language_loss": 0.87090898,
      "learning_rate": 3.257045967068692e-06,
      "loss": 0.89324665,
      "num_input_tokens_seen": 54593120,
      "step": 2538,
      "time_per_iteration": 2.6590452194213867
    },
    {
      "auxiliary_loss_clip": 0.0117245,
      "auxiliary_loss_mlp": 0.0103686,
      "balance_loss_clip": 1.06196868,
      "balance_loss_mlp": 1.02632153,
      "epoch": 0.30529669933265197,
      "flos": 21945118970880.0,
      "grad_norm": 2.3766088856597904,
      "language_loss": 0.82475984,
      "learning_rate": 3.2564399945912848e-06,
      "loss": 0.8468529,
      "num_input_tokens_seen": 54612910,
      "step": 2539,
      "time_per_iteration": 2.698746919631958
    },
    {
      "auxiliary_loss_clip": 0.01213184,
      "auxiliary_loss_mlp": 0.01039525,
      "balance_loss_clip": 1.06822097,
      "balance_loss_mlp": 1.03103757,
      "epoch": 0.305416942223291,
      "flos": 21835411856640.0,
      "grad_norm": 2.860360455833658,
      "language_loss": 0.82290047,
      "learning_rate": 3.2558338315129287e-06,
      "loss": 0.84542751,
      "num_input_tokens_seen": 54631055,
      "step": 2540,
      "time_per_iteration": 2.6892824172973633
    },
    {
      "auxiliary_loss_clip": 0.01200203,
      "auxiliary_loss_mlp": 0.01034347,
      "balance_loss_clip": 1.06361401,
      "balance_loss_mlp": 1.02455378,
      "epoch": 0.30553718511393013,
      "flos": 33911810709120.0,
      "grad_norm": 2.0725669159459974,
      "language_loss": 0.75698966,
      "learning_rate": 3.2552274779255785e-06,
      "loss": 0.77933514,
      "num_input_tokens_seen": 54651985,
      "step": 2541,
      "time_per_iteration": 2.741487741470337
    },
    {
      "auxiliary_loss_clip": 0.01210544,
      "auxiliary_loss_mlp": 0.00763059,
      "balance_loss_clip": 1.06252813,
      "balance_loss_mlp": 1.00061297,
      "epoch": 0.30565742800456924,
      "flos": 22268530051200.0,
      "grad_norm": 2.4423803290347426,
      "language_loss": 0.77181387,
      "learning_rate": 3.2546209339212184e-06,
      "loss": 0.79154986,
      "num_input_tokens_seen": 54671005,
      "step": 2542,
      "time_per_iteration": 2.707143783569336
    },
    {
      "auxiliary_loss_clip": 0.01168733,
      "auxiliary_loss_mlp": 0.01034861,
      "balance_loss_clip": 1.05760729,
      "balance_loss_mlp": 1.02553892,
      "epoch": 0.3057776708952083,
      "flos": 22565044823040.0,
      "grad_norm": 2.268135936763462,
      "language_loss": 0.77634811,
      "learning_rate": 3.25401419959186e-06,
      "loss": 0.79838407,
      "num_input_tokens_seen": 54691615,
      "step": 2543,
      "time_per_iteration": 2.745425224304199
    },
    {
      "auxiliary_loss_clip": 0.0118614,
      "auxiliary_loss_mlp": 0.00763188,
      "balance_loss_clip": 1.06382692,
      "balance_loss_mlp": 1.00066054,
      "epoch": 0.3058979137858474,
      "flos": 21799213925760.0,
      "grad_norm": 2.6945888294282963,
      "language_loss": 0.76478726,
      "learning_rate": 3.253407275029545e-06,
      "loss": 0.78428054,
      "num_input_tokens_seen": 54710520,
      "step": 2544,
      "time_per_iteration": 2.684481382369995
    },
    {
      "auxiliary_loss_clip": 0.01141779,
      "auxiliary_loss_mlp": 0.01033986,
      "balance_loss_clip": 1.05729365,
      "balance_loss_mlp": 1.02388322,
      "epoch": 0.3060181566764865,
      "flos": 26979435601920.0,
      "grad_norm": 2.5795933327675407,
      "language_loss": 0.8031081,
      "learning_rate": 3.2528001603263425e-06,
      "loss": 0.82486582,
      "num_input_tokens_seen": 54732590,
      "step": 2545,
      "time_per_iteration": 2.827211856842041
    },
    {
      "auxiliary_loss_clip": 0.01146271,
      "auxiliary_loss_mlp": 0.01039763,
      "balance_loss_clip": 1.0510726,
      "balance_loss_mlp": 1.02960587,
      "epoch": 0.3061383995671256,
      "flos": 19865101173120.0,
      "grad_norm": 2.036440754317123,
      "language_loss": 0.81575346,
      "learning_rate": 3.2521928555743514e-06,
      "loss": 0.83761382,
      "num_input_tokens_seen": 54749935,
      "step": 2546,
      "time_per_iteration": 2.7872376441955566
    },
    {
      "auxiliary_loss_clip": 0.01201005,
      "auxiliary_loss_mlp": 0.01033971,
      "balance_loss_clip": 1.06528378,
      "balance_loss_mlp": 1.02445221,
      "epoch": 0.3062586424577647,
      "flos": 22127509255680.0,
      "grad_norm": 1.9313170417053234,
      "language_loss": 0.67267114,
      "learning_rate": 3.2515853608657e-06,
      "loss": 0.69502091,
      "num_input_tokens_seen": 54767935,
      "step": 2547,
      "time_per_iteration": 2.7498276233673096
    },
    {
      "auxiliary_loss_clip": 0.01166917,
      "auxiliary_loss_mlp": 0.01041858,
      "balance_loss_clip": 1.06180954,
      "balance_loss_mlp": 1.03236294,
      "epoch": 0.3063788853484038,
      "flos": 20845497962880.0,
      "grad_norm": 2.763146920665566,
      "language_loss": 0.75621015,
      "learning_rate": 3.250977676292545e-06,
      "loss": 0.7782979,
      "num_input_tokens_seen": 54786175,
      "step": 2548,
      "time_per_iteration": 2.7502799034118652
    },
    {
      "auxiliary_loss_clip": 0.01155829,
      "auxiliary_loss_mlp": 0.01037178,
      "balance_loss_clip": 1.05637658,
      "balance_loss_mlp": 1.02808249,
      "epoch": 0.30649912823904285,
      "flos": 16209717707520.0,
      "grad_norm": 2.419472917454689,
      "language_loss": 0.79321772,
      "learning_rate": 3.2503698019470712e-06,
      "loss": 0.81514782,
      "num_input_tokens_seen": 54801945,
      "step": 2549,
      "time_per_iteration": 2.7255804538726807
    },
    {
      "auxiliary_loss_clip": 0.01188868,
      "auxiliary_loss_mlp": 0.01036766,
      "balance_loss_clip": 1.05788183,
      "balance_loss_mlp": 1.02699041,
      "epoch": 0.30661937112968196,
      "flos": 18617815353600.0,
      "grad_norm": 2.2402020470594928,
      "language_loss": 0.78257585,
      "learning_rate": 3.249761737921492e-06,
      "loss": 0.80483222,
      "num_input_tokens_seen": 54818475,
      "step": 2550,
      "time_per_iteration": 2.711674928665161
    },
    {
      "auxiliary_loss_clip": 0.01148167,
      "auxiliary_loss_mlp": 0.01044039,
      "balance_loss_clip": 1.05460691,
      "balance_loss_mlp": 1.03440118,
      "epoch": 0.30673961402032107,
      "flos": 31390809638400.0,
      "grad_norm": 2.4616342957904513,
      "language_loss": 0.74436659,
      "learning_rate": 3.249153484308051e-06,
      "loss": 0.76628864,
      "num_input_tokens_seen": 54837090,
      "step": 2551,
      "time_per_iteration": 2.939643144607544
    },
    {
      "auxiliary_loss_clip": 0.01191343,
      "auxiliary_loss_mlp": 0.01039033,
      "balance_loss_clip": 1.05608845,
      "balance_loss_mlp": 1.02928138,
      "epoch": 0.3068598569109601,
      "flos": 20229809915520.0,
      "grad_norm": 5.486961096128423,
      "language_loss": 0.7816838,
      "learning_rate": 3.2485450411990194e-06,
      "loss": 0.8039875,
      "num_input_tokens_seen": 54856445,
      "step": 2552,
      "time_per_iteration": 2.648223400115967
    },
    {
      "auxiliary_loss_clip": 0.01211226,
      "auxiliary_loss_mlp": 0.0103729,
      "balance_loss_clip": 1.0601933,
      "balance_loss_mlp": 1.02765799,
      "epoch": 0.30698009980159924,
      "flos": 29601991399680.0,
      "grad_norm": 10.47819971143871,
      "language_loss": 0.82575381,
      "learning_rate": 3.2479364086866983e-06,
      "loss": 0.84823889,
      "num_input_tokens_seen": 54876700,
      "step": 2553,
      "time_per_iteration": 2.744112014770508
    },
    {
      "auxiliary_loss_clip": 0.01214902,
      "auxiliary_loss_mlp": 0.01033542,
      "balance_loss_clip": 1.06788135,
      "balance_loss_mlp": 1.0246129,
      "epoch": 0.30710034269223835,
      "flos": 23842423261440.0,
      "grad_norm": 1.859046902650137,
      "language_loss": 0.81555808,
      "learning_rate": 3.247327586863416e-06,
      "loss": 0.8380425,
      "num_input_tokens_seen": 54897580,
      "step": 2554,
      "time_per_iteration": 3.549311876296997
    },
    {
      "auxiliary_loss_clip": 0.01191725,
      "auxiliary_loss_mlp": 0.01030331,
      "balance_loss_clip": 1.0597837,
      "balance_loss_mlp": 1.02116346,
      "epoch": 0.3072205855828774,
      "flos": 25884986152320.0,
      "grad_norm": 2.7508030820200204,
      "language_loss": 0.77088737,
      "learning_rate": 3.2467185758215304e-06,
      "loss": 0.79310787,
      "num_input_tokens_seen": 54917320,
      "step": 2555,
      "time_per_iteration": 2.6806623935699463
    },
    {
      "auxiliary_loss_clip": 0.0118066,
      "auxiliary_loss_mlp": 0.01041743,
      "balance_loss_clip": 1.06369054,
      "balance_loss_mlp": 1.03206909,
      "epoch": 0.3073408284735165,
      "flos": 22236390357120.0,
      "grad_norm": 3.9339874203100678,
      "language_loss": 0.85481954,
      "learning_rate": 3.246109375653428e-06,
      "loss": 0.8770436,
      "num_input_tokens_seen": 54934085,
      "step": 2556,
      "time_per_iteration": 3.6598281860351562
    },
    {
      "auxiliary_loss_clip": 0.01215995,
      "auxiliary_loss_mlp": 0.01042897,
      "balance_loss_clip": 1.0661509,
      "balance_loss_mlp": 1.03381968,
      "epoch": 0.30746107136415557,
      "flos": 19500284689920.0,
      "grad_norm": 3.9347641722473994,
      "language_loss": 0.78242362,
      "learning_rate": 3.2454999864515243e-06,
      "loss": 0.80501258,
      "num_input_tokens_seen": 54953460,
      "step": 2557,
      "time_per_iteration": 3.605720043182373
    },
    {
      "auxiliary_loss_clip": 0.0116235,
      "auxiliary_loss_mlp": 0.01034744,
      "balance_loss_clip": 1.05890155,
      "balance_loss_mlp": 1.02539182,
      "epoch": 0.3075813142547947,
      "flos": 21724806902400.0,
      "grad_norm": 1.7864027219918615,
      "language_loss": 0.69716585,
      "learning_rate": 3.244890408308263e-06,
      "loss": 0.71913683,
      "num_input_tokens_seen": 54974165,
      "step": 2558,
      "time_per_iteration": 2.725595474243164
    },
    {
      "auxiliary_loss_clip": 0.01152143,
      "auxiliary_loss_mlp": 0.01034725,
      "balance_loss_clip": 1.05506682,
      "balance_loss_mlp": 1.02553391,
      "epoch": 0.3077015571454338,
      "flos": 24097963593600.0,
      "grad_norm": 2.3782898632815908,
      "language_loss": 0.61704028,
      "learning_rate": 3.2442806413161165e-06,
      "loss": 0.63890892,
      "num_input_tokens_seen": 54993810,
      "step": 2559,
      "time_per_iteration": 3.749783992767334
    },
    {
      "auxiliary_loss_clip": 0.0117543,
      "auxiliary_loss_mlp": 0.01034535,
      "balance_loss_clip": 1.06016064,
      "balance_loss_mlp": 1.02474749,
      "epoch": 0.30782180003607285,
      "flos": 18405476104320.0,
      "grad_norm": 2.9533410710988113,
      "language_loss": 0.76139909,
      "learning_rate": 3.243670685567586e-06,
      "loss": 0.78349876,
      "num_input_tokens_seen": 55011210,
      "step": 2560,
      "time_per_iteration": 2.7289087772369385
    },
    {
      "auxiliary_loss_clip": 0.01149784,
      "auxiliary_loss_mlp": 0.01036129,
      "balance_loss_clip": 1.05597043,
      "balance_loss_mlp": 1.02737904,
      "epoch": 0.30794204292671196,
      "flos": 23878549365120.0,
      "grad_norm": 2.2488246319706655,
      "language_loss": 0.81129074,
      "learning_rate": 3.2430605411552012e-06,
      "loss": 0.83314979,
      "num_input_tokens_seen": 55031325,
      "step": 2561,
      "time_per_iteration": 2.8275678157806396
    },
    {
      "auxiliary_loss_clip": 0.01101402,
      "auxiliary_loss_mlp": 0.01001748,
      "balance_loss_clip": 1.02057159,
      "balance_loss_mlp": 1.00009048,
      "epoch": 0.30806228581735107,
      "flos": 67927800816000.0,
      "grad_norm": 0.8903621834404118,
      "language_loss": 0.70567471,
      "learning_rate": 3.2424502081715205e-06,
      "loss": 0.72670621,
      "num_input_tokens_seen": 55094440,
      "step": 2562,
      "time_per_iteration": 3.331042766571045
    },
    {
      "auxiliary_loss_clip": 0.01167478,
      "auxiliary_loss_mlp": 0.01031503,
      "balance_loss_clip": 1.05799246,
      "balance_loss_mlp": 1.02175164,
      "epoch": 0.3081825287079901,
      "flos": 23843213360640.0,
      "grad_norm": 2.082274879725013,
      "language_loss": 0.78242958,
      "learning_rate": 3.241839686709132e-06,
      "loss": 0.8044194,
      "num_input_tokens_seen": 55115375,
      "step": 2563,
      "time_per_iteration": 2.8704586029052734
    },
    {
      "auxiliary_loss_clip": 0.01156999,
      "auxiliary_loss_mlp": 0.01041272,
      "balance_loss_clip": 1.05491054,
      "balance_loss_mlp": 1.03102565,
      "epoch": 0.30830277159862923,
      "flos": 16209969102720.0,
      "grad_norm": 2.7939570914102636,
      "language_loss": 0.82492846,
      "learning_rate": 3.2412289768606495e-06,
      "loss": 0.84691119,
      "num_input_tokens_seen": 55131945,
      "step": 2564,
      "time_per_iteration": 2.7837257385253906
    },
    {
      "auxiliary_loss_clip": 0.01186989,
      "auxiliary_loss_mlp": 0.01044605,
      "balance_loss_clip": 1.06298637,
      "balance_loss_mlp": 1.0346632,
      "epoch": 0.30842301448926834,
      "flos": 29349503723520.0,
      "grad_norm": 2.4274482413199205,
      "language_loss": 0.82460177,
      "learning_rate": 3.240618078718718e-06,
      "loss": 0.84691769,
      "num_input_tokens_seen": 55153405,
      "step": 2565,
      "time_per_iteration": 2.7701637744903564
    },
    {
      "auxiliary_loss_clip": 0.01173743,
      "auxiliary_loss_mlp": 0.0104169,
      "balance_loss_clip": 1.06002748,
      "balance_loss_mlp": 1.03143787,
      "epoch": 0.3085432573799074,
      "flos": 21945190798080.0,
      "grad_norm": 2.9177650624429945,
      "language_loss": 0.74760056,
      "learning_rate": 3.240006992376011e-06,
      "loss": 0.76975489,
      "num_input_tokens_seen": 55173030,
      "step": 2566,
      "time_per_iteration": 2.817187786102295
    },
    {
      "auxiliary_loss_clip": 0.01170747,
      "auxiliary_loss_mlp": 0.01035348,
      "balance_loss_clip": 1.05933392,
      "balance_loss_mlp": 1.02610373,
      "epoch": 0.3086635002705465,
      "flos": 22054718344320.0,
      "grad_norm": 2.141987761180646,
      "language_loss": 0.75966525,
      "learning_rate": 3.2393957179252284e-06,
      "loss": 0.78172618,
      "num_input_tokens_seen": 55189565,
      "step": 2567,
      "time_per_iteration": 2.7034196853637695
    },
    {
      "auxiliary_loss_clip": 0.01200994,
      "auxiliary_loss_mlp": 0.01034386,
      "balance_loss_clip": 1.06599939,
      "balance_loss_mlp": 1.02444375,
      "epoch": 0.3087837431611856,
      "flos": 32665925520000.0,
      "grad_norm": 1.9136563785831058,
      "language_loss": 0.80582172,
      "learning_rate": 3.2387842554591016e-06,
      "loss": 0.82817549,
      "num_input_tokens_seen": 55210380,
      "step": 2568,
      "time_per_iteration": 2.8552074432373047
    },
    {
      "auxiliary_loss_clip": 0.01167349,
      "auxiliary_loss_mlp": 0.01036725,
      "balance_loss_clip": 1.05757594,
      "balance_loss_mlp": 1.02667618,
      "epoch": 0.3089039860518247,
      "flos": 17599245384960.0,
      "grad_norm": 4.879115768100588,
      "language_loss": 0.88307124,
      "learning_rate": 3.238172605070388e-06,
      "loss": 0.90511203,
      "num_input_tokens_seen": 55225795,
      "step": 2569,
      "time_per_iteration": 2.704549551010132
    },
    {
      "auxiliary_loss_clip": 0.01186174,
      "auxiliary_loss_mlp": 0.01036921,
      "balance_loss_clip": 1.0619452,
      "balance_loss_mlp": 1.02733612,
      "epoch": 0.3090242289424638,
      "flos": 14383839611520.0,
      "grad_norm": 15.603039474848567,
      "language_loss": 0.78529215,
      "learning_rate": 3.2375607668518745e-06,
      "loss": 0.80752313,
      "num_input_tokens_seen": 55238830,
      "step": 2570,
      "time_per_iteration": 2.672715187072754
    },
    {
      "auxiliary_loss_clip": 0.01154205,
      "auxiliary_loss_mlp": 0.01034561,
      "balance_loss_clip": 1.05425107,
      "balance_loss_mlp": 1.02562046,
      "epoch": 0.30914447183310284,
      "flos": 16068625084800.0,
      "grad_norm": 3.3555570492559577,
      "language_loss": 0.9018355,
      "learning_rate": 3.236948740896377e-06,
      "loss": 0.92372316,
      "num_input_tokens_seen": 55253630,
      "step": 2571,
      "time_per_iteration": 2.7680256366729736
    },
    {
      "auxiliary_loss_clip": 0.01194317,
      "auxiliary_loss_mlp": 0.0103891,
      "balance_loss_clip": 1.06405044,
      "balance_loss_mlp": 1.02875912,
      "epoch": 0.30926471472374195,
      "flos": 32230221546240.0,
      "grad_norm": 2.6754306857216634,
      "language_loss": 0.84637666,
      "learning_rate": 3.2363365272967384e-06,
      "loss": 0.86870891,
      "num_input_tokens_seen": 55276200,
      "step": 2572,
      "time_per_iteration": 2.778756856918335
    },
    {
      "auxiliary_loss_clip": 0.01175375,
      "auxiliary_loss_mlp": 0.01039769,
      "balance_loss_clip": 1.06192029,
      "balance_loss_mlp": 1.02976704,
      "epoch": 0.30938495761438106,
      "flos": 20370722970240.0,
      "grad_norm": 2.828993525507045,
      "language_loss": 0.81975925,
      "learning_rate": 3.235724126145832e-06,
      "loss": 0.8419106,
      "num_input_tokens_seen": 55292235,
      "step": 2573,
      "time_per_iteration": 2.7241508960723877
    },
    {
      "auxiliary_loss_clip": 0.01164802,
      "auxiliary_loss_mlp": 0.01041599,
      "balance_loss_clip": 1.05702376,
      "balance_loss_mlp": 1.03193152,
      "epoch": 0.3095052005050201,
      "flos": 24061155131520.0,
      "grad_norm": 1.669808757071099,
      "language_loss": 0.77949661,
      "learning_rate": 3.235111537536558e-06,
      "loss": 0.80156064,
      "num_input_tokens_seen": 55313050,
      "step": 2574,
      "time_per_iteration": 2.7712137699127197
    },
    {
      "auxiliary_loss_clip": 0.0117183,
      "auxiliary_loss_mlp": 0.0103222,
      "balance_loss_clip": 1.05914545,
      "balance_loss_mlp": 1.02257562,
      "epoch": 0.30962544339565923,
      "flos": 23401547729280.0,
      "grad_norm": 6.202587949129426,
      "language_loss": 0.83082175,
      "learning_rate": 3.2344987615618456e-06,
      "loss": 0.85286224,
      "num_input_tokens_seen": 55332885,
      "step": 2575,
      "time_per_iteration": 2.8213083744049072
    },
    {
      "auxiliary_loss_clip": 0.01213311,
      "auxiliary_loss_mlp": 0.01039469,
      "balance_loss_clip": 1.0636425,
      "balance_loss_mlp": 1.03023624,
      "epoch": 0.30974568628629834,
      "flos": 33799984692480.0,
      "grad_norm": 4.522640067872931,
      "language_loss": 0.78786218,
      "learning_rate": 3.2338857983146533e-06,
      "loss": 0.81038988,
      "num_input_tokens_seen": 55354385,
      "step": 2576,
      "time_per_iteration": 2.701995849609375
    },
    {
      "auxiliary_loss_clip": 0.01154161,
      "auxiliary_loss_mlp": 0.01033899,
      "balance_loss_clip": 1.05287325,
      "balance_loss_mlp": 1.02389693,
      "epoch": 0.3098659291769374,
      "flos": 20229594433920.0,
      "grad_norm": 4.394459213979865,
      "language_loss": 0.76927346,
      "learning_rate": 3.233272647887966e-06,
      "loss": 0.79115403,
      "num_input_tokens_seen": 55373275,
      "step": 2577,
      "time_per_iteration": 2.777229070663452
    },
    {
      "auxiliary_loss_clip": 0.01217473,
      "auxiliary_loss_mlp": 0.01042498,
      "balance_loss_clip": 1.06670427,
      "balance_loss_mlp": 1.03183508,
      "epoch": 0.3099861720675765,
      "flos": 24748556682240.0,
      "grad_norm": 2.1594939933085424,
      "language_loss": 0.90338528,
      "learning_rate": 3.2326593103747985e-06,
      "loss": 0.92598498,
      "num_input_tokens_seen": 55392290,
      "step": 2578,
      "time_per_iteration": 2.6831271648406982
    },
    {
      "auxiliary_loss_clip": 0.01147183,
      "auxiliary_loss_mlp": 0.01035612,
      "balance_loss_clip": 1.05299592,
      "balance_loss_mlp": 1.02569366,
      "epoch": 0.3101064149582156,
      "flos": 11765485704960.0,
      "grad_norm": 2.254633704343681,
      "language_loss": 0.85189235,
      "learning_rate": 3.2320457858681936e-06,
      "loss": 0.87372029,
      "num_input_tokens_seen": 55410680,
      "step": 2579,
      "time_per_iteration": 2.767103433609009
    },
    {
      "auxiliary_loss_clip": 0.01183004,
      "auxiliary_loss_mlp": 0.01033746,
      "balance_loss_clip": 1.06271791,
      "balance_loss_mlp": 1.02448893,
      "epoch": 0.31022665784885467,
      "flos": 23033247626880.0,
      "grad_norm": 3.1348683593922897,
      "language_loss": 0.85424775,
      "learning_rate": 3.2314320744612228e-06,
      "loss": 0.87641531,
      "num_input_tokens_seen": 55425980,
      "step": 2580,
      "time_per_iteration": 3.5973727703094482
    },
    {
      "auxiliary_loss_clip": 0.01211335,
      "auxiliary_loss_mlp": 0.01032586,
      "balance_loss_clip": 1.06398714,
      "balance_loss_mlp": 1.02273893,
      "epoch": 0.3103469007394938,
      "flos": 16289188548480.0,
      "grad_norm": 2.36977521993293,
      "language_loss": 0.77086079,
      "learning_rate": 3.2308181762469854e-06,
      "loss": 0.79330003,
      "num_input_tokens_seen": 55443925,
      "step": 2581,
      "time_per_iteration": 3.5583887100219727
    },
    {
      "auxiliary_loss_clip": 0.01199922,
      "auxiliary_loss_mlp": 0.01032219,
      "balance_loss_clip": 1.06393611,
      "balance_loss_mlp": 1.02228832,
      "epoch": 0.3104671436301329,
      "flos": 30515271626880.0,
      "grad_norm": 2.7909843357642656,
      "language_loss": 0.78910482,
      "learning_rate": 3.230204091318609e-06,
      "loss": 0.81142622,
      "num_input_tokens_seen": 55464465,
      "step": 2582,
      "time_per_iteration": 2.7540414333343506
    },
    {
      "auxiliary_loss_clip": 0.01199686,
      "auxiliary_loss_mlp": 0.01034633,
      "balance_loss_clip": 1.06296277,
      "balance_loss_mlp": 1.02501845,
      "epoch": 0.31058738652077195,
      "flos": 20047240062720.0,
      "grad_norm": 2.624271021740333,
      "language_loss": 0.85203558,
      "learning_rate": 3.2295898197692503e-06,
      "loss": 0.8743788,
      "num_input_tokens_seen": 55483425,
      "step": 2583,
      "time_per_iteration": 2.6997225284576416
    },
    {
      "auxiliary_loss_clip": 0.01156245,
      "auxiliary_loss_mlp": 0.01036825,
      "balance_loss_clip": 1.05512595,
      "balance_loss_mlp": 1.02779531,
      "epoch": 0.31070762941141106,
      "flos": 28074639237120.0,
      "grad_norm": 1.933042322570062,
      "language_loss": 0.79280353,
      "learning_rate": 3.228975361692094e-06,
      "loss": 0.81473422,
      "num_input_tokens_seen": 55504445,
      "step": 2584,
      "time_per_iteration": 3.799053430557251
    },
    {
      "auxiliary_loss_clip": 0.01198809,
      "auxiliary_loss_mlp": 0.01036137,
      "balance_loss_clip": 1.06237531,
      "balance_loss_mlp": 1.02585554,
      "epoch": 0.31082787230205017,
      "flos": 20521907314560.0,
      "grad_norm": 3.2304891567613407,
      "language_loss": 0.80406511,
      "learning_rate": 3.228360717180352e-06,
      "loss": 0.82641453,
      "num_input_tokens_seen": 55521970,
      "step": 2585,
      "time_per_iteration": 3.541933059692383
    },
    {
      "auxiliary_loss_clip": 0.01077487,
      "auxiliary_loss_mlp": 0.01007397,
      "balance_loss_clip": 1.02086282,
      "balance_loss_mlp": 1.00568056,
      "epoch": 0.3109481151926892,
      "flos": 62445928723200.0,
      "grad_norm": 0.8420307218497816,
      "language_loss": 0.59339815,
      "learning_rate": 3.227745886327266e-06,
      "loss": 0.61424702,
      "num_input_tokens_seen": 55580665,
      "step": 2586,
      "time_per_iteration": 3.206221342086792
    },
    {
      "auxiliary_loss_clip": 0.01080055,
      "auxiliary_loss_mlp": 0.00753876,
      "balance_loss_clip": 1.01967072,
      "balance_loss_mlp": 1.0004586,
      "epoch": 0.31106835808332833,
      "flos": 44746744723200.0,
      "grad_norm": 0.8019583987524078,
      "language_loss": 0.55796176,
      "learning_rate": 3.227130869226105e-06,
      "loss": 0.57630104,
      "num_input_tokens_seen": 55637825,
      "step": 2587,
      "time_per_iteration": 3.249063491821289
    },
    {
      "auxiliary_loss_clip": 0.01154213,
      "auxiliary_loss_mlp": 0.01030271,
      "balance_loss_clip": 1.05596304,
      "balance_loss_mlp": 1.02061534,
      "epoch": 0.3111886009739674,
      "flos": 23403056100480.0,
      "grad_norm": 7.505114212272787,
      "language_loss": 0.83195305,
      "learning_rate": 3.226515665970167e-06,
      "loss": 0.85379791,
      "num_input_tokens_seen": 55655365,
      "step": 2588,
      "time_per_iteration": 2.7917559146881104
    },
    {
      "auxiliary_loss_clip": 0.01198008,
      "auxiliary_loss_mlp": 0.01037051,
      "balance_loss_clip": 1.06268692,
      "balance_loss_mlp": 1.02634048,
      "epoch": 0.3113088438646065,
      "flos": 17530728192000.0,
      "grad_norm": 2.476801887019921,
      "language_loss": 0.87273645,
      "learning_rate": 3.225900276652777e-06,
      "loss": 0.89508706,
      "num_input_tokens_seen": 55672140,
      "step": 2589,
      "time_per_iteration": 2.6922245025634766
    },
    {
      "auxiliary_loss_clip": 0.01168703,
      "auxiliary_loss_mlp": 0.01035325,
      "balance_loss_clip": 1.05461669,
      "balance_loss_mlp": 1.02648532,
      "epoch": 0.3114290867552456,
      "flos": 28365802882560.0,
      "grad_norm": 1.830897643633984,
      "language_loss": 0.76070923,
      "learning_rate": 3.2252847013672906e-06,
      "loss": 0.78274947,
      "num_input_tokens_seen": 55694800,
      "step": 2590,
      "time_per_iteration": 2.7968664169311523
    },
    {
      "auxiliary_loss_clip": 0.01162851,
      "auxiliary_loss_mlp": 0.01032818,
      "balance_loss_clip": 1.0564245,
      "balance_loss_mlp": 1.02363932,
      "epoch": 0.31154932964588467,
      "flos": 27379157126400.0,
      "grad_norm": 3.9152130470925997,
      "language_loss": 0.76754838,
      "learning_rate": 3.224668940207089e-06,
      "loss": 0.789505,
      "num_input_tokens_seen": 55713785,
      "step": 2591,
      "time_per_iteration": 2.8526670932769775
    },
    {
      "auxiliary_loss_clip": 0.01201473,
      "auxiliary_loss_mlp": 0.01041986,
      "balance_loss_clip": 1.06278872,
      "balance_loss_mlp": 1.03227592,
      "epoch": 0.3116695725365238,
      "flos": 26541864120960.0,
      "grad_norm": 2.1858623441964498,
      "language_loss": 0.87289453,
      "learning_rate": 3.2240529932655828e-06,
      "loss": 0.89532912,
      "num_input_tokens_seen": 55733050,
      "step": 2592,
      "time_per_iteration": 2.7566661834716797
    },
    {
      "auxiliary_loss_clip": 0.01176036,
      "auxiliary_loss_mlp": 0.01037193,
      "balance_loss_clip": 1.05994534,
      "balance_loss_mlp": 1.02827621,
      "epoch": 0.3117898154271629,
      "flos": 21177600134400.0,
      "grad_norm": 4.539424595733985,
      "language_loss": 0.8870573,
      "learning_rate": 3.223436860636211e-06,
      "loss": 0.90918958,
      "num_input_tokens_seen": 55748685,
      "step": 2593,
      "time_per_iteration": 2.7671031951904297
    },
    {
      "auxiliary_loss_clip": 0.01176425,
      "auxiliary_loss_mlp": 0.01041528,
      "balance_loss_clip": 1.06062746,
      "balance_loss_mlp": 1.03211617,
      "epoch": 0.31191005831780194,
      "flos": 27272430840960.0,
      "grad_norm": 2.4896894060276766,
      "language_loss": 0.74249768,
      "learning_rate": 3.2228205424124403e-06,
      "loss": 0.76467717,
      "num_input_tokens_seen": 55771840,
      "step": 2594,
      "time_per_iteration": 2.7616331577301025
    },
    {
      "auxiliary_loss_clip": 0.01144642,
      "auxiliary_loss_mlp": 0.01036642,
      "balance_loss_clip": 1.05217528,
      "balance_loss_mlp": 1.02732587,
      "epoch": 0.31203030120844105,
      "flos": 12963501043200.0,
      "grad_norm": 2.398028488760043,
      "language_loss": 0.74405849,
      "learning_rate": 3.222204038687765e-06,
      "loss": 0.76587135,
      "num_input_tokens_seen": 55784975,
      "step": 2595,
      "time_per_iteration": 2.781034469604492
    },
    {
      "auxiliary_loss_clip": 0.01184313,
      "auxiliary_loss_mlp": 0.01038212,
      "balance_loss_clip": 1.05747974,
      "balance_loss_mlp": 1.02856827,
      "epoch": 0.31215054409908016,
      "flos": 27562014288000.0,
      "grad_norm": 1.7074049352274803,
      "language_loss": 0.88287431,
      "learning_rate": 3.221587349555709e-06,
      "loss": 0.90509951,
      "num_input_tokens_seen": 55805235,
      "step": 2596,
      "time_per_iteration": 2.7472732067108154
    },
    {
      "auxiliary_loss_clip": 0.01182859,
      "auxiliary_loss_mlp": 0.0076425,
      "balance_loss_clip": 1.05775762,
      "balance_loss_mlp": 1.00069177,
      "epoch": 0.3122707869897192,
      "flos": 21506326427520.0,
      "grad_norm": 2.3788084069574498,
      "language_loss": 0.69414252,
      "learning_rate": 3.2209704751098236e-06,
      "loss": 0.71361363,
      "num_input_tokens_seen": 55824265,
      "step": 2597,
      "time_per_iteration": 2.743433713912964
    },
    {
      "auxiliary_loss_clip": 0.01158193,
      "auxiliary_loss_mlp": 0.01032875,
      "balance_loss_clip": 1.05625916,
      "balance_loss_mlp": 1.02341032,
      "epoch": 0.31239102988035833,
      "flos": 15187017674880.0,
      "grad_norm": 2.7968943900166887,
      "language_loss": 0.82616138,
      "learning_rate": 3.2203534154436875e-06,
      "loss": 0.84807211,
      "num_input_tokens_seen": 55838620,
      "step": 2598,
      "time_per_iteration": 2.7194175720214844
    },
    {
      "auxiliary_loss_clip": 0.0116859,
      "auxiliary_loss_mlp": 0.0076375,
      "balance_loss_clip": 1.05730033,
      "balance_loss_mlp": 1.00073361,
      "epoch": 0.31251127277099744,
      "flos": 22053712763520.0,
      "grad_norm": 2.2804802990441737,
      "language_loss": 0.76121104,
      "learning_rate": 3.2197361706509084e-06,
      "loss": 0.78053439,
      "num_input_tokens_seen": 55859375,
      "step": 2599,
      "time_per_iteration": 2.792937755584717
    },
    {
      "auxiliary_loss_clip": 0.01162575,
      "auxiliary_loss_mlp": 0.01037918,
      "balance_loss_clip": 1.0542146,
      "balance_loss_mlp": 1.02758288,
      "epoch": 0.3126315156616365,
      "flos": 15193984913280.0,
      "grad_norm": 3.5418272909819835,
      "language_loss": 0.84183145,
      "learning_rate": 3.2191187408251228e-06,
      "loss": 0.86383641,
      "num_input_tokens_seen": 55876535,
      "step": 2600,
      "time_per_iteration": 2.7336745262145996
    },
    {
      "auxiliary_loss_clip": 0.01124125,
      "auxiliary_loss_mlp": 0.01041325,
      "balance_loss_clip": 1.05094719,
      "balance_loss_mlp": 1.03080511,
      "epoch": 0.3127517585522756,
      "flos": 18145338831360.0,
      "grad_norm": 9.782173705452768,
      "language_loss": 0.79204214,
      "learning_rate": 3.218501126059993e-06,
      "loss": 0.81369662,
      "num_input_tokens_seen": 55891930,
      "step": 2601,
      "time_per_iteration": 2.8098466396331787
    },
    {
      "auxiliary_loss_clip": 0.01144289,
      "auxiliary_loss_mlp": 0.01039267,
      "balance_loss_clip": 1.04796529,
      "balance_loss_mlp": 1.02901506,
      "epoch": 0.31287200144291466,
      "flos": 21908633731200.0,
      "grad_norm": 2.448122721609388,
      "language_loss": 0.81769323,
      "learning_rate": 3.2178833264492116e-06,
      "loss": 0.83952874,
      "num_input_tokens_seen": 55910635,
      "step": 2602,
      "time_per_iteration": 2.7664170265197754
    },
    {
      "auxiliary_loss_clip": 0.01172316,
      "auxiliary_loss_mlp": 0.01039635,
      "balance_loss_clip": 1.06087542,
      "balance_loss_mlp": 1.02974117,
      "epoch": 0.31299224433355377,
      "flos": 29896997800320.0,
      "grad_norm": 1.8324541387958808,
      "language_loss": 0.76126075,
      "learning_rate": 3.217265342086498e-06,
      "loss": 0.78338027,
      "num_input_tokens_seen": 55931125,
      "step": 2603,
      "time_per_iteration": 2.7568492889404297
    },
    {
      "auxiliary_loss_clip": 0.01172798,
      "auxiliary_loss_mlp": 0.01039783,
      "balance_loss_clip": 1.05806494,
      "balance_loss_mlp": 1.02960205,
      "epoch": 0.3131124872241929,
      "flos": 11655886331520.0,
      "grad_norm": 3.688759510430917,
      "language_loss": 0.73080438,
      "learning_rate": 3.216647173065599e-06,
      "loss": 0.75293016,
      "num_input_tokens_seen": 55946590,
      "step": 2604,
      "time_per_iteration": 2.704495668411255
    },
    {
      "auxiliary_loss_clip": 0.0119363,
      "auxiliary_loss_mlp": 0.01038313,
      "balance_loss_clip": 1.05941582,
      "balance_loss_mlp": 1.02887762,
      "epoch": 0.31323273011483194,
      "flos": 49848785470080.0,
      "grad_norm": 1.9211044572509932,
      "language_loss": 0.74051201,
      "learning_rate": 3.216028819480292e-06,
      "loss": 0.76283139,
      "num_input_tokens_seen": 55967930,
      "step": 2605,
      "time_per_iteration": 2.8938934803009033
    },
    {
      "auxiliary_loss_clip": 0.01137015,
      "auxiliary_loss_mlp": 0.01033033,
      "balance_loss_clip": 1.04860234,
      "balance_loss_mlp": 1.02280486,
      "epoch": 0.31335297300547105,
      "flos": 22601278667520.0,
      "grad_norm": 2.2842922022598535,
      "language_loss": 0.75732088,
      "learning_rate": 3.2154102814243793e-06,
      "loss": 0.77902138,
      "num_input_tokens_seen": 55987070,
      "step": 2606,
      "time_per_iteration": 3.748685121536255
    },
    {
      "auxiliary_loss_clip": 0.01157879,
      "auxiliary_loss_mlp": 0.01037499,
      "balance_loss_clip": 1.05419171,
      "balance_loss_mlp": 1.02710366,
      "epoch": 0.31347321589611016,
      "flos": 34710858708480.0,
      "grad_norm": 2.0448530128118985,
      "language_loss": 0.66945904,
      "learning_rate": 3.2147915589916937e-06,
      "loss": 0.69141287,
      "num_input_tokens_seen": 56008630,
      "step": 2607,
      "time_per_iteration": 3.788355588912964
    },
    {
      "auxiliary_loss_clip": 0.01153084,
      "auxiliary_loss_mlp": 0.01044117,
      "balance_loss_clip": 1.05202925,
      "balance_loss_mlp": 1.03388321,
      "epoch": 0.3135934587867492,
      "flos": 19755789108480.0,
      "grad_norm": 2.2769373651126505,
      "language_loss": 0.82942653,
      "learning_rate": 3.2141726522760938e-06,
      "loss": 0.85139853,
      "num_input_tokens_seen": 56026690,
      "step": 2608,
      "time_per_iteration": 2.7211833000183105
    },
    {
      "auxiliary_loss_clip": 0.01074247,
      "auxiliary_loss_mlp": 0.0100644,
      "balance_loss_clip": 1.0203799,
      "balance_loss_mlp": 1.00460434,
      "epoch": 0.3137137016773883,
      "flos": 65815535583360.0,
      "grad_norm": 0.7003168851128521,
      "language_loss": 0.52573323,
      "learning_rate": 3.213553561371469e-06,
      "loss": 0.54654008,
      "num_input_tokens_seen": 56090425,
      "step": 2609,
      "time_per_iteration": 3.338794231414795
    },
    {
      "auxiliary_loss_clip": 0.01163901,
      "auxiliary_loss_mlp": 0.01038556,
      "balance_loss_clip": 1.0588305,
      "balance_loss_mlp": 1.02947235,
      "epoch": 0.31383394456802743,
      "flos": 16252739222400.0,
      "grad_norm": 2.370594182505211,
      "language_loss": 0.96053898,
      "learning_rate": 3.212934286371733e-06,
      "loss": 0.98256356,
      "num_input_tokens_seen": 56107135,
      "step": 2610,
      "time_per_iteration": 3.6715309619903564
    },
    {
      "auxiliary_loss_clip": 0.01151049,
      "auxiliary_loss_mlp": 0.00764582,
      "balance_loss_clip": 1.0550127,
      "balance_loss_mlp": 1.00057304,
      "epoch": 0.3139541874586665,
      "flos": 38795517613440.0,
      "grad_norm": 3.777146716227442,
      "language_loss": 0.82989788,
      "learning_rate": 3.2123148273708304e-06,
      "loss": 0.84905422,
      "num_input_tokens_seen": 56127325,
      "step": 2611,
      "time_per_iteration": 3.8315072059631348
    },
    {
      "auxiliary_loss_clip": 0.01145618,
      "auxiliary_loss_mlp": 0.01037614,
      "balance_loss_clip": 1.05659032,
      "balance_loss_mlp": 1.02776706,
      "epoch": 0.3140744303493056,
      "flos": 25046328430080.0,
      "grad_norm": 2.281957032985867,
      "language_loss": 0.76940238,
      "learning_rate": 3.211695184462733e-06,
      "loss": 0.79123473,
      "num_input_tokens_seen": 56148500,
      "step": 2612,
      "time_per_iteration": 2.7769670486450195
    },
    {
      "auxiliary_loss_clip": 0.01071058,
      "auxiliary_loss_mlp": 0.0101568,
      "balance_loss_clip": 1.0283438,
      "balance_loss_mlp": 1.01384401,
      "epoch": 0.3141946732399447,
      "flos": 72504254782080.0,
      "grad_norm": 0.8853483922550226,
      "language_loss": 0.60419536,
      "learning_rate": 3.2110753577414383e-06,
      "loss": 0.62506276,
      "num_input_tokens_seen": 56210080,
      "step": 2613,
      "time_per_iteration": 3.267606496810913
    },
    {
      "auxiliary_loss_clip": 0.01176119,
      "auxiliary_loss_mlp": 0.01034888,
      "balance_loss_clip": 1.05789685,
      "balance_loss_mlp": 1.02509499,
      "epoch": 0.31431491613058377,
      "flos": 19239788280960.0,
      "grad_norm": 3.0180356478722463,
      "language_loss": 0.78982008,
      "learning_rate": 3.2104553473009757e-06,
      "loss": 0.81193006,
      "num_input_tokens_seen": 56228200,
      "step": 2614,
      "time_per_iteration": 2.6531131267547607
    },
    {
      "auxiliary_loss_clip": 0.01151004,
      "auxiliary_loss_mlp": 0.01029764,
      "balance_loss_clip": 1.04992366,
      "balance_loss_mlp": 1.02041245,
      "epoch": 0.3144351590212229,
      "flos": 36210596290560.0,
      "grad_norm": 1.9760102373050348,
      "language_loss": 0.67932266,
      "learning_rate": 3.209835153235399e-06,
      "loss": 0.70113033,
      "num_input_tokens_seen": 56249755,
      "step": 2615,
      "time_per_iteration": 2.8517043590545654
    },
    {
      "auxiliary_loss_clip": 0.01172279,
      "auxiliary_loss_mlp": 0.01047616,
      "balance_loss_clip": 1.0586282,
      "balance_loss_mlp": 1.03886044,
      "epoch": 0.314555401911862,
      "flos": 18551740285440.0,
      "grad_norm": 1.994857455328856,
      "language_loss": 0.67700303,
      "learning_rate": 3.2092147756387916e-06,
      "loss": 0.699202,
      "num_input_tokens_seen": 56270080,
      "step": 2616,
      "time_per_iteration": 2.7188186645507812
    },
    {
      "auxiliary_loss_clip": 0.01180088,
      "auxiliary_loss_mlp": 0.00763542,
      "balance_loss_clip": 1.06074214,
      "balance_loss_mlp": 1.00061131,
      "epoch": 0.31467564480250104,
      "flos": 16362877299840.0,
      "grad_norm": 3.2110549161859097,
      "language_loss": 0.83624434,
      "learning_rate": 3.208594214605264e-06,
      "loss": 0.85568064,
      "num_input_tokens_seen": 56288625,
      "step": 2617,
      "time_per_iteration": 2.7247276306152344
    },
    {
      "auxiliary_loss_clip": 0.01151251,
      "auxiliary_loss_mlp": 0.01039192,
      "balance_loss_clip": 1.05118191,
      "balance_loss_mlp": 1.03056097,
      "epoch": 0.31479588769314015,
      "flos": 21652375127040.0,
      "grad_norm": 2.614012737267209,
      "language_loss": 0.77197206,
      "learning_rate": 3.2079734702289553e-06,
      "loss": 0.79387653,
      "num_input_tokens_seen": 56307520,
      "step": 2618,
      "time_per_iteration": 2.773045301437378
    },
    {
      "auxiliary_loss_clip": 0.01070776,
      "auxiliary_loss_mlp": 0.00753566,
      "balance_loss_clip": 1.01736307,
      "balance_loss_mlp": 0.99998778,
      "epoch": 0.3149161305837792,
      "flos": 66051072040320.0,
      "grad_norm": 0.803448494324474,
      "language_loss": 0.60297263,
      "learning_rate": 3.207352542604031e-06,
      "loss": 0.621216,
      "num_input_tokens_seen": 56369855,
      "step": 2619,
      "time_per_iteration": 3.4032111167907715
    },
    {
      "auxiliary_loss_clip": 0.01191537,
      "auxiliary_loss_mlp": 0.01032461,
      "balance_loss_clip": 1.05786812,
      "balance_loss_mlp": 1.023669,
      "epoch": 0.3150363734744183,
      "flos": 28987201192320.0,
      "grad_norm": 3.3644711683398567,
      "language_loss": 0.78277552,
      "learning_rate": 3.2067314318246864e-06,
      "loss": 0.80501556,
      "num_input_tokens_seen": 56390570,
      "step": 2620,
      "time_per_iteration": 2.821519613265991
    },
    {
      "auxiliary_loss_clip": 0.01165683,
      "auxiliary_loss_mlp": 0.00763423,
      "balance_loss_clip": 1.06081915,
      "balance_loss_mlp": 1.00064421,
      "epoch": 0.31515661636505743,
      "flos": 27636600879360.0,
      "grad_norm": 2.199592729414684,
      "language_loss": 0.77858102,
      "learning_rate": 3.206110137985143e-06,
      "loss": 0.79787207,
      "num_input_tokens_seen": 56410775,
      "step": 2621,
      "time_per_iteration": 2.7419931888580322
    },
    {
      "auxiliary_loss_clip": 0.01149328,
      "auxiliary_loss_mlp": 0.00763619,
      "balance_loss_clip": 1.05480146,
      "balance_loss_mlp": 1.00064158,
      "epoch": 0.3152768592556965,
      "flos": 24605632465920.0,
      "grad_norm": 3.5172672783994474,
      "language_loss": 0.92315835,
      "learning_rate": 3.2054886611796505e-06,
      "loss": 0.9422878,
      "num_input_tokens_seen": 56429770,
      "step": 2622,
      "time_per_iteration": 2.8382887840270996
    },
    {
      "auxiliary_loss_clip": 0.01079897,
      "auxiliary_loss_mlp": 0.01005536,
      "balance_loss_clip": 1.01796579,
      "balance_loss_mlp": 1.00378346,
      "epoch": 0.3153971021463356,
      "flos": 68476908026880.0,
      "grad_norm": 0.896317929278778,
      "language_loss": 0.63490266,
      "learning_rate": 3.204867001502487e-06,
      "loss": 0.65575701,
      "num_input_tokens_seen": 56488425,
      "step": 2623,
      "time_per_iteration": 3.274658441543579
    },
    {
      "auxiliary_loss_clip": 0.01156518,
      "auxiliary_loss_mlp": 0.01035903,
      "balance_loss_clip": 1.05970383,
      "balance_loss_mlp": 1.02625287,
      "epoch": 0.3155173450369747,
      "flos": 25593714766080.0,
      "grad_norm": 2.005616989313868,
      "language_loss": 0.80691433,
      "learning_rate": 3.2042451590479567e-06,
      "loss": 0.82883859,
      "num_input_tokens_seen": 56508940,
      "step": 2624,
      "time_per_iteration": 2.7945926189422607
    },
    {
      "auxiliary_loss_clip": 0.01193525,
      "auxiliary_loss_mlp": 0.01033419,
      "balance_loss_clip": 1.05911005,
      "balance_loss_mlp": 1.02466559,
      "epoch": 0.31563758792761376,
      "flos": 24309333175680.0,
      "grad_norm": 3.872600477953325,
      "language_loss": 0.86871219,
      "learning_rate": 3.203623133910394e-06,
      "loss": 0.89098155,
      "num_input_tokens_seen": 56527245,
      "step": 2625,
      "time_per_iteration": 2.700404405593872
    },
    {
      "auxiliary_loss_clip": 0.01156498,
      "auxiliary_loss_mlp": 0.0103546,
      "balance_loss_clip": 1.05367696,
      "balance_loss_mlp": 1.02641821,
      "epoch": 0.31575783081825287,
      "flos": 31903865550720.0,
      "grad_norm": 2.4531786144769723,
      "language_loss": 0.77094543,
      "learning_rate": 3.203000926184158e-06,
      "loss": 0.79286504,
      "num_input_tokens_seen": 56546170,
      "step": 2626,
      "time_per_iteration": 2.8043041229248047
    },
    {
      "auxiliary_loss_clip": 0.01176895,
      "auxiliary_loss_mlp": 0.01039048,
      "balance_loss_clip": 1.05757272,
      "balance_loss_mlp": 1.03032804,
      "epoch": 0.315878073708892,
      "flos": 30810960385920.0,
      "grad_norm": 2.5965965263104005,
      "language_loss": 0.77918887,
      "learning_rate": 3.202378535963639e-06,
      "loss": 0.80134833,
      "num_input_tokens_seen": 56567085,
      "step": 2627,
      "time_per_iteration": 2.784625768661499
    },
    {
      "auxiliary_loss_clip": 0.01141636,
      "auxiliary_loss_mlp": 0.01040377,
      "balance_loss_clip": 1.0513854,
      "balance_loss_mlp": 1.03031623,
      "epoch": 0.31599831659953104,
      "flos": 22200264253440.0,
      "grad_norm": 2.0437599172274017,
      "language_loss": 0.84074706,
      "learning_rate": 3.2017559633432516e-06,
      "loss": 0.86256719,
      "num_input_tokens_seen": 56586715,
      "step": 2628,
      "time_per_iteration": 2.812119722366333
    },
    {
      "auxiliary_loss_clip": 0.0116624,
      "auxiliary_loss_mlp": 0.01039175,
      "balance_loss_clip": 1.05746973,
      "balance_loss_mlp": 1.02925718,
      "epoch": 0.31611855949017015,
      "flos": 25593463370880.0,
      "grad_norm": 3.267688213778092,
      "language_loss": 0.66208744,
      "learning_rate": 3.2011332084174398e-06,
      "loss": 0.68414164,
      "num_input_tokens_seen": 56607585,
      "step": 2629,
      "time_per_iteration": 2.7514824867248535
    },
    {
      "auxiliary_loss_clip": 0.01182355,
      "auxiliary_loss_mlp": 0.01038267,
      "balance_loss_clip": 1.05680156,
      "balance_loss_mlp": 1.02864671,
      "epoch": 0.31623880238080926,
      "flos": 20594087694720.0,
      "grad_norm": 18.65965155540562,
      "language_loss": 0.89564711,
      "learning_rate": 3.2005102712806756e-06,
      "loss": 0.9178533,
      "num_input_tokens_seen": 56626415,
      "step": 2630,
      "time_per_iteration": 2.596219301223755
    },
    {
      "auxiliary_loss_clip": 0.01160892,
      "auxiliary_loss_mlp": 0.01040004,
      "balance_loss_clip": 1.05454779,
      "balance_loss_mlp": 1.02993059,
      "epoch": 0.3163590452714483,
      "flos": 12784917600000.0,
      "grad_norm": 3.5515771705449852,
      "language_loss": 0.72929418,
      "learning_rate": 3.1998871520274575e-06,
      "loss": 0.7513032,
      "num_input_tokens_seen": 56641750,
      "step": 2631,
      "time_per_iteration": 2.6285407543182373
    },
    {
      "auxiliary_loss_clip": 0.0115731,
      "auxiliary_loss_mlp": 0.01037797,
      "balance_loss_clip": 1.05317211,
      "balance_loss_mlp": 1.02840304,
      "epoch": 0.3164792881620874,
      "flos": 23041292273280.0,
      "grad_norm": 2.5978656241893554,
      "language_loss": 0.85242283,
      "learning_rate": 3.199263850752312e-06,
      "loss": 0.87437391,
      "num_input_tokens_seen": 56662585,
      "step": 2632,
      "time_per_iteration": 3.4425950050354004
    },
    {
      "auxiliary_loss_clip": 0.01186844,
      "auxiliary_loss_mlp": 0.01038498,
      "balance_loss_clip": 1.05776215,
      "balance_loss_mlp": 1.02925372,
      "epoch": 0.31659953105272653,
      "flos": 18296271780480.0,
      "grad_norm": 16.15628419994406,
      "language_loss": 0.85634875,
      "learning_rate": 3.198640367549795e-06,
      "loss": 0.87860215,
      "num_input_tokens_seen": 56681480,
      "step": 2633,
      "time_per_iteration": 3.4939215183258057
    },
    {
      "auxiliary_loss_clip": 0.01163384,
      "auxiliary_loss_mlp": 0.01036618,
      "balance_loss_clip": 1.05612111,
      "balance_loss_mlp": 1.02689648,
      "epoch": 0.3167197739433656,
      "flos": 25703421880320.0,
      "grad_norm": 4.8807531075502695,
      "language_loss": 0.85834891,
      "learning_rate": 3.198016702514487e-06,
      "loss": 0.88034892,
      "num_input_tokens_seen": 56701760,
      "step": 2634,
      "time_per_iteration": 2.748396158218384
    },
    {
      "auxiliary_loss_clip": 0.01190562,
      "auxiliary_loss_mlp": 0.0076327,
      "balance_loss_clip": 1.06065798,
      "balance_loss_mlp": 1.00057626,
      "epoch": 0.3168400168340047,
      "flos": 23546016230400.0,
      "grad_norm": 4.1091626883664985,
      "language_loss": 0.84756565,
      "learning_rate": 3.1973928557409972e-06,
      "loss": 0.86710405,
      "num_input_tokens_seen": 56719800,
      "step": 2635,
      "time_per_iteration": 3.6447174549102783
    },
    {
      "auxiliary_loss_clip": 0.01189946,
      "auxiliary_loss_mlp": 0.01034024,
      "balance_loss_clip": 1.05987954,
      "balance_loss_mlp": 1.02516127,
      "epoch": 0.31696025972464376,
      "flos": 28366449327360.0,
      "grad_norm": 2.4111837969886754,
      "language_loss": 0.7155807,
      "learning_rate": 3.1967688273239636e-06,
      "loss": 0.73782039,
      "num_input_tokens_seen": 56739605,
      "step": 2636,
      "time_per_iteration": 2.67578125
    },
    {
      "auxiliary_loss_clip": 0.01177497,
      "auxiliary_loss_mlp": 0.01035157,
      "balance_loss_clip": 1.05527031,
      "balance_loss_mlp": 1.02553093,
      "epoch": 0.31708050261528287,
      "flos": 16399111144320.0,
      "grad_norm": 8.980627619609406,
      "language_loss": 0.82463801,
      "learning_rate": 3.1961446173580503e-06,
      "loss": 0.84676456,
      "num_input_tokens_seen": 56756545,
      "step": 2637,
      "time_per_iteration": 3.617272138595581
    },
    {
      "auxiliary_loss_clip": 0.01186498,
      "auxiliary_loss_mlp": 0.01042464,
      "balance_loss_clip": 1.06144738,
      "balance_loss_mlp": 1.03347552,
      "epoch": 0.317200745505922,
      "flos": 26212347728640.0,
      "grad_norm": 2.8435360554759677,
      "language_loss": 0.77531207,
      "learning_rate": 3.1955202259379502e-06,
      "loss": 0.7976017,
      "num_input_tokens_seen": 56778275,
      "step": 2638,
      "time_per_iteration": 2.742436647415161
    },
    {
      "auxiliary_loss_clip": 0.01190806,
      "auxiliary_loss_mlp": 0.01035909,
      "balance_loss_clip": 1.05665255,
      "balance_loss_mlp": 1.02653325,
      "epoch": 0.31732098839656103,
      "flos": 31350876693120.0,
      "grad_norm": 2.662597425791309,
      "language_loss": 0.83038312,
      "learning_rate": 3.194895653158381e-06,
      "loss": 0.85265017,
      "num_input_tokens_seen": 56797215,
      "step": 2639,
      "time_per_iteration": 2.691288948059082
    },
    {
      "auxiliary_loss_clip": 0.01049973,
      "auxiliary_loss_mlp": 0.01017023,
      "balance_loss_clip": 1.01558197,
      "balance_loss_mlp": 1.01527023,
      "epoch": 0.31744123128720014,
      "flos": 58989024835200.0,
      "grad_norm": 0.7804303129469493,
      "language_loss": 0.55516779,
      "learning_rate": 3.194270899114093e-06,
      "loss": 0.57583773,
      "num_input_tokens_seen": 56863010,
      "step": 2640,
      "time_per_iteration": 3.420010805130005
    },
    {
      "auxiliary_loss_clip": 0.01183725,
      "auxiliary_loss_mlp": 0.0104392,
      "balance_loss_clip": 1.0612582,
      "balance_loss_mlp": 1.03432345,
      "epoch": 0.31756147417783925,
      "flos": 17417573372160.0,
      "grad_norm": 2.3222890000388294,
      "language_loss": 0.82375455,
      "learning_rate": 3.193645963899858e-06,
      "loss": 0.84603101,
      "num_input_tokens_seen": 56880625,
      "step": 2641,
      "time_per_iteration": 2.731848955154419
    },
    {
      "auxiliary_loss_clip": 0.01132455,
      "auxiliary_loss_mlp": 0.01041497,
      "balance_loss_clip": 1.05484402,
      "balance_loss_mlp": 1.03273535,
      "epoch": 0.3176817170684783,
      "flos": 25481673267840.0,
      "grad_norm": 3.128828013564592,
      "language_loss": 0.84330237,
      "learning_rate": 3.193020847610479e-06,
      "loss": 0.86504185,
      "num_input_tokens_seen": 56900945,
      "step": 2642,
      "time_per_iteration": 2.8598151206970215
    },
    {
      "auxiliary_loss_clip": 0.01189476,
      "auxiliary_loss_mlp": 0.01032689,
      "balance_loss_clip": 1.06133652,
      "balance_loss_mlp": 1.02309871,
      "epoch": 0.3178019599591174,
      "flos": 24972603765120.0,
      "grad_norm": 2.5841185609448134,
      "language_loss": 0.71937466,
      "learning_rate": 3.192395550340787e-06,
      "loss": 0.74159634,
      "num_input_tokens_seen": 56918895,
      "step": 2643,
      "time_per_iteration": 2.743185520172119
    },
    {
      "auxiliary_loss_clip": 0.01193101,
      "auxiliary_loss_mlp": 0.01037888,
      "balance_loss_clip": 1.06085527,
      "balance_loss_mlp": 1.029037,
      "epoch": 0.31792220284975653,
      "flos": 12422220019200.0,
      "grad_norm": 3.682511474746362,
      "language_loss": 0.769894,
      "learning_rate": 3.191770072185638e-06,
      "loss": 0.79220378,
      "num_input_tokens_seen": 56935890,
      "step": 2644,
      "time_per_iteration": 2.616135835647583
    },
    {
      "auxiliary_loss_clip": 0.01171357,
      "auxiliary_loss_mlp": 0.01032233,
      "balance_loss_clip": 1.05560875,
      "balance_loss_mlp": 1.02305961,
      "epoch": 0.3180424457403956,
      "flos": 15485759089920.0,
      "grad_norm": 3.0069143757469967,
      "language_loss": 0.73385799,
      "learning_rate": 3.191144413239916e-06,
      "loss": 0.75589383,
      "num_input_tokens_seen": 56952460,
      "step": 2645,
      "time_per_iteration": 2.603717565536499
    },
    {
      "auxiliary_loss_clip": 0.01197895,
      "auxiliary_loss_mlp": 0.01028821,
      "balance_loss_clip": 1.06133056,
      "balance_loss_mlp": 1.01940382,
      "epoch": 0.3181626886310347,
      "flos": 26174964648960.0,
      "grad_norm": 3.938467696495203,
      "language_loss": 0.88505256,
      "learning_rate": 3.190518573598534e-06,
      "loss": 0.90731978,
      "num_input_tokens_seen": 56969065,
      "step": 2646,
      "time_per_iteration": 2.7088818550109863
    },
    {
      "auxiliary_loss_clip": 0.01146042,
      "auxiliary_loss_mlp": 0.01033586,
      "balance_loss_clip": 1.05813634,
      "balance_loss_mlp": 1.02384675,
      "epoch": 0.3182829315216738,
      "flos": 25483109811840.0,
      "grad_norm": 1.7660638595654854,
      "language_loss": 0.77767682,
      "learning_rate": 3.1898925533564308e-06,
      "loss": 0.79947305,
      "num_input_tokens_seen": 56990535,
      "step": 2647,
      "time_per_iteration": 2.7577967643737793
    },
    {
      "auxiliary_loss_clip": 0.01193123,
      "auxiliary_loss_mlp": 0.01041706,
      "balance_loss_clip": 1.05955935,
      "balance_loss_mlp": 1.03290296,
      "epoch": 0.31840317441231286,
      "flos": 18113701927680.0,
      "grad_norm": 2.7336058633241653,
      "language_loss": 0.64196026,
      "learning_rate": 3.1892663526085733e-06,
      "loss": 0.66430855,
      "num_input_tokens_seen": 57008910,
      "step": 2648,
      "time_per_iteration": 2.740830421447754
    },
    {
      "auxiliary_loss_clip": 0.01043465,
      "auxiliary_loss_mlp": 0.01001297,
      "balance_loss_clip": 1.01603556,
      "balance_loss_mlp": 0.99962777,
      "epoch": 0.31852341730295197,
      "flos": 64741948957440.0,
      "grad_norm": 0.7459663537290835,
      "language_loss": 0.56857944,
      "learning_rate": 3.188639971449956e-06,
      "loss": 0.58902705,
      "num_input_tokens_seen": 57074960,
      "step": 2649,
      "time_per_iteration": 3.314793825149536
    },
    {
      "auxiliary_loss_clip": 0.01185048,
      "auxiliary_loss_mlp": 0.01036349,
      "balance_loss_clip": 1.05729413,
      "balance_loss_mlp": 1.02635288,
      "epoch": 0.318643660193591,
      "flos": 20668135582080.0,
      "grad_norm": 2.5152226771190036,
      "language_loss": 0.72590226,
      "learning_rate": 3.1880134099756e-06,
      "loss": 0.7481162,
      "num_input_tokens_seen": 57094595,
      "step": 2650,
      "time_per_iteration": 2.691962480545044
    },
    {
      "auxiliary_loss_clip": 0.01150307,
      "auxiliary_loss_mlp": 0.0103053,
      "balance_loss_clip": 1.05052745,
      "balance_loss_mlp": 1.02129173,
      "epoch": 0.31876390308423014,
      "flos": 26943345411840.0,
      "grad_norm": 2.0901928636602807,
      "language_loss": 0.69522774,
      "learning_rate": 3.1873866682805535e-06,
      "loss": 0.71703613,
      "num_input_tokens_seen": 57115290,
      "step": 2651,
      "time_per_iteration": 2.794619083404541
    },
    {
      "auxiliary_loss_clip": 0.0118232,
      "auxiliary_loss_mlp": 0.01036064,
      "balance_loss_clip": 1.05970621,
      "balance_loss_mlp": 1.0270462,
      "epoch": 0.31888414597486925,
      "flos": 18041916597120.0,
      "grad_norm": 1.9353825449114452,
      "language_loss": 0.89155334,
      "learning_rate": 3.186759746459894e-06,
      "loss": 0.91373724,
      "num_input_tokens_seen": 57134400,
      "step": 2652,
      "time_per_iteration": 2.6666016578674316
    },
    {
      "auxiliary_loss_clip": 0.01168097,
      "auxiliary_loss_mlp": 0.01034143,
      "balance_loss_clip": 1.05080128,
      "balance_loss_mlp": 1.02457047,
      "epoch": 0.3190043888655083,
      "flos": 25149319701120.0,
      "grad_norm": 3.484951833909127,
      "language_loss": 0.79700041,
      "learning_rate": 3.1861326446087246e-06,
      "loss": 0.81902283,
      "num_input_tokens_seen": 57153140,
      "step": 2653,
      "time_per_iteration": 2.7477288246154785
    },
    {
      "auxiliary_loss_clip": 0.01150578,
      "auxiliary_loss_mlp": 0.01036917,
      "balance_loss_clip": 1.05300665,
      "balance_loss_mlp": 1.02811933,
      "epoch": 0.3191246317561474,
      "flos": 22053892331520.0,
      "grad_norm": 2.809895477418415,
      "language_loss": 0.72125936,
      "learning_rate": 3.1855053628221763e-06,
      "loss": 0.74313426,
      "num_input_tokens_seen": 57172395,
      "step": 2654,
      "time_per_iteration": 2.7137980461120605
    },
    {
      "auxiliary_loss_clip": 0.01193037,
      "auxiliary_loss_mlp": 0.01038705,
      "balance_loss_clip": 1.05961025,
      "balance_loss_mlp": 1.02985358,
      "epoch": 0.3192448746467865,
      "flos": 14901815687040.0,
      "grad_norm": 4.391648501173869,
      "language_loss": 0.90114635,
      "learning_rate": 3.184877901195407e-06,
      "loss": 0.92346376,
      "num_input_tokens_seen": 57189090,
      "step": 2655,
      "time_per_iteration": 2.6982369422912598
    },
    {
      "auxiliary_loss_clip": 0.01052961,
      "auxiliary_loss_mlp": 0.01004119,
      "balance_loss_clip": 1.01317596,
      "balance_loss_mlp": 1.00252175,
      "epoch": 0.3193651175374256,
      "flos": 67234832657280.0,
      "grad_norm": 0.7923775334047771,
      "language_loss": 0.62789261,
      "learning_rate": 3.184250259823602e-06,
      "loss": 0.64846337,
      "num_input_tokens_seen": 57251620,
      "step": 2656,
      "time_per_iteration": 3.3471755981445312
    },
    {
      "auxiliary_loss_clip": 0.01171911,
      "auxiliary_loss_mlp": 0.0103796,
      "balance_loss_clip": 1.05809736,
      "balance_loss_mlp": 1.02771425,
      "epoch": 0.3194853604280647,
      "flos": 12233077977600.0,
      "grad_norm": 2.4731553024364086,
      "language_loss": 0.81811762,
      "learning_rate": 3.183622438801974e-06,
      "loss": 0.84021628,
      "num_input_tokens_seen": 57266910,
      "step": 2657,
      "time_per_iteration": 2.7038838863372803
    },
    {
      "auxiliary_loss_clip": 0.01192982,
      "auxiliary_loss_mlp": 0.01034391,
      "balance_loss_clip": 1.06178379,
      "balance_loss_mlp": 1.02604079,
      "epoch": 0.3196056033187038,
      "flos": 14939917038720.0,
      "grad_norm": 8.32457197822414,
      "language_loss": 0.75465047,
      "learning_rate": 3.1829944382257637e-06,
      "loss": 0.77692419,
      "num_input_tokens_seen": 57285040,
      "step": 2658,
      "time_per_iteration": 4.64120078086853
    },
    {
      "auxiliary_loss_clip": 0.01182181,
      "auxiliary_loss_mlp": 0.01037992,
      "balance_loss_clip": 1.05749643,
      "balance_loss_mlp": 1.02903891,
      "epoch": 0.31972584620934286,
      "flos": 23768878164480.0,
      "grad_norm": 17.819152720771243,
      "language_loss": 0.81568909,
      "learning_rate": 3.1823662581902373e-06,
      "loss": 0.83789074,
      "num_input_tokens_seen": 57302725,
      "step": 2659,
      "time_per_iteration": 2.6853978633880615
    },
    {
      "auxiliary_loss_clip": 0.01160943,
      "auxiliary_loss_mlp": 0.01038242,
      "balance_loss_clip": 1.05324173,
      "balance_loss_mlp": 1.02917624,
      "epoch": 0.31984608909998197,
      "flos": 21251540280960.0,
      "grad_norm": 4.561782543521667,
      "language_loss": 0.75544804,
      "learning_rate": 3.1817378987906896e-06,
      "loss": 0.77743989,
      "num_input_tokens_seen": 57322230,
      "step": 2660,
      "time_per_iteration": 2.813662528991699
    },
    {
      "auxiliary_loss_clip": 0.01182381,
      "auxiliary_loss_mlp": 0.01034806,
      "balance_loss_clip": 1.05925786,
      "balance_loss_mlp": 1.02631187,
      "epoch": 0.3199663319906211,
      "flos": 18296235866880.0,
      "grad_norm": 2.3011069852865305,
      "language_loss": 0.79924297,
      "learning_rate": 3.181109360122442e-06,
      "loss": 0.82141489,
      "num_input_tokens_seen": 57339820,
      "step": 2661,
      "time_per_iteration": 3.6070313453674316
    },
    {
      "auxiliary_loss_clip": 0.01139389,
      "auxiliary_loss_mlp": 0.01036341,
      "balance_loss_clip": 1.04849362,
      "balance_loss_mlp": 1.02704263,
      "epoch": 0.32008657488126013,
      "flos": 18733627779840.0,
      "grad_norm": 3.00604193805975,
      "language_loss": 0.78338778,
      "learning_rate": 3.1804806422808445e-06,
      "loss": 0.80514514,
      "num_input_tokens_seen": 57356955,
      "step": 2662,
      "time_per_iteration": 3.721055746078491
    },
    {
      "auxiliary_loss_clip": 0.01176112,
      "auxiliary_loss_mlp": 0.01036765,
      "balance_loss_clip": 1.05743456,
      "balance_loss_mlp": 1.02712727,
      "epoch": 0.32020681777189924,
      "flos": 20595344670720.0,
      "grad_norm": 2.5368395396954657,
      "language_loss": 0.73214054,
      "learning_rate": 3.1798517453612714e-06,
      "loss": 0.7542693,
      "num_input_tokens_seen": 57376760,
      "step": 2663,
      "time_per_iteration": 2.711953639984131
    },
    {
      "auxiliary_loss_clip": 0.01178092,
      "auxiliary_loss_mlp": 0.01031149,
      "balance_loss_clip": 1.06225491,
      "balance_loss_mlp": 1.02246499,
      "epoch": 0.32032706066253835,
      "flos": 35261692750080.0,
      "grad_norm": 3.507748383280235,
      "language_loss": 0.75309813,
      "learning_rate": 3.1792226694591265e-06,
      "loss": 0.77519059,
      "num_input_tokens_seen": 57398145,
      "step": 2664,
      "time_per_iteration": 2.80966854095459
    },
    {
      "auxiliary_loss_clip": 0.01168758,
      "auxiliary_loss_mlp": 0.01034305,
      "balance_loss_clip": 1.05393994,
      "balance_loss_mlp": 1.02524567,
      "epoch": 0.3204473035531774,
      "flos": 15304230731520.0,
      "grad_norm": 2.6948926932603374,
      "language_loss": 0.80593616,
      "learning_rate": 3.178593414669841e-06,
      "loss": 0.82796681,
      "num_input_tokens_seen": 57416730,
      "step": 2665,
      "time_per_iteration": 2.7100679874420166
    },
    {
      "auxiliary_loss_clip": 0.01167952,
      "auxiliary_loss_mlp": 0.01038034,
      "balance_loss_clip": 1.05344272,
      "balance_loss_mlp": 1.02759171,
      "epoch": 0.3205675464438165,
      "flos": 24462564595200.0,
      "grad_norm": 5.5948352294424675,
      "language_loss": 0.71075714,
      "learning_rate": 3.1779639810888707e-06,
      "loss": 0.73281693,
      "num_input_tokens_seen": 57436325,
      "step": 2666,
      "time_per_iteration": 2.747110366821289
    },
    {
      "auxiliary_loss_clip": 0.01162188,
      "auxiliary_loss_mlp": 0.00763536,
      "balance_loss_clip": 1.05254912,
      "balance_loss_mlp": 1.00077486,
      "epoch": 0.3206877893344556,
      "flos": 22456235548800.0,
      "grad_norm": 1.8051489013310205,
      "language_loss": 0.75832355,
      "learning_rate": 3.1773343688117013e-06,
      "loss": 0.77758086,
      "num_input_tokens_seen": 57457235,
      "step": 2667,
      "time_per_iteration": 2.691973924636841
    },
    {
      "auxiliary_loss_clip": 0.01151332,
      "auxiliary_loss_mlp": 0.01042334,
      "balance_loss_clip": 1.05370092,
      "balance_loss_mlp": 1.03331017,
      "epoch": 0.3208080322250947,
      "flos": 20412236113920.0,
      "grad_norm": 2.1160150405862166,
      "language_loss": 0.83756918,
      "learning_rate": 3.1767045779338445e-06,
      "loss": 0.85950583,
      "num_input_tokens_seen": 57474895,
      "step": 2668,
      "time_per_iteration": 2.773784875869751
    },
    {
      "auxiliary_loss_clip": 0.01145131,
      "auxiliary_loss_mlp": 0.01030032,
      "balance_loss_clip": 1.04783833,
      "balance_loss_mlp": 1.02147281,
      "epoch": 0.3209282751157338,
      "flos": 21762118154880.0,
      "grad_norm": 3.461095250438232,
      "language_loss": 0.91545951,
      "learning_rate": 3.176074608550839e-06,
      "loss": 0.93721116,
      "num_input_tokens_seen": 57490715,
      "step": 2669,
      "time_per_iteration": 2.71675443649292
    },
    {
      "auxiliary_loss_clip": 0.01166527,
      "auxiliary_loss_mlp": 0.01035884,
      "balance_loss_clip": 1.05142117,
      "balance_loss_mlp": 1.02718186,
      "epoch": 0.32104851800637285,
      "flos": 22055041566720.0,
      "grad_norm": 2.5124344119756077,
      "language_loss": 0.82232636,
      "learning_rate": 3.17544446075825e-06,
      "loss": 0.84435046,
      "num_input_tokens_seen": 57509880,
      "step": 2670,
      "time_per_iteration": 2.769746780395508
    },
    {
      "auxiliary_loss_clip": 0.01159013,
      "auxiliary_loss_mlp": 0.01034943,
      "balance_loss_clip": 1.05225348,
      "balance_loss_mlp": 1.02561522,
      "epoch": 0.32116876089701196,
      "flos": 37012301896320.0,
      "grad_norm": 1.6961762016312052,
      "language_loss": 0.70899123,
      "learning_rate": 3.174814134651671e-06,
      "loss": 0.73093075,
      "num_input_tokens_seen": 57532430,
      "step": 2671,
      "time_per_iteration": 2.8657617568969727
    },
    {
      "auxiliary_loss_clip": 0.01163195,
      "auxiliary_loss_mlp": 0.0103401,
      "balance_loss_clip": 1.05432248,
      "balance_loss_mlp": 1.0251174,
      "epoch": 0.3212890037876511,
      "flos": 21979233912960.0,
      "grad_norm": 1.9147794469060275,
      "language_loss": 0.80437249,
      "learning_rate": 3.1741836303267215e-06,
      "loss": 0.82634455,
      "num_input_tokens_seen": 57551965,
      "step": 2672,
      "time_per_iteration": 2.739240884780884
    },
    {
      "auxiliary_loss_clip": 0.01183281,
      "auxiliary_loss_mlp": 0.01040268,
      "balance_loss_clip": 1.05740428,
      "balance_loss_mlp": 1.03116643,
      "epoch": 0.32140924667829013,
      "flos": 10342345875840.0,
      "grad_norm": 3.9189109579429973,
      "language_loss": 0.75379813,
      "learning_rate": 3.1735529478790496e-06,
      "loss": 0.77603358,
      "num_input_tokens_seen": 57569955,
      "step": 2673,
      "time_per_iteration": 2.7256007194519043
    },
    {
      "auxiliary_loss_clip": 0.01112133,
      "auxiliary_loss_mlp": 0.01043759,
      "balance_loss_clip": 1.04959846,
      "balance_loss_mlp": 1.03416276,
      "epoch": 0.32152948956892924,
      "flos": 50798910072960.0,
      "grad_norm": 3.22073392741438,
      "language_loss": 0.80037975,
      "learning_rate": 3.172922087404328e-06,
      "loss": 0.82193863,
      "num_input_tokens_seen": 57592215,
      "step": 2674,
      "time_per_iteration": 3.161362409591675
    },
    {
      "auxiliary_loss_clip": 0.01067246,
      "auxiliary_loss_mlp": 0.01009566,
      "balance_loss_clip": 1.02239347,
      "balance_loss_mlp": 1.00807023,
      "epoch": 0.32164973245956835,
      "flos": 63863250549120.0,
      "grad_norm": 0.7775517521875535,
      "language_loss": 0.55207807,
      "learning_rate": 3.1722910489982586e-06,
      "loss": 0.57284617,
      "num_input_tokens_seen": 57652575,
      "step": 2675,
      "time_per_iteration": 3.4014317989349365
    },
    {
      "auxiliary_loss_clip": 0.01190509,
      "auxiliary_loss_mlp": 0.01035737,
      "balance_loss_clip": 1.06000328,
      "balance_loss_mlp": 1.02577162,
      "epoch": 0.3217699753502074,
      "flos": 23513948363520.0,
      "grad_norm": 3.1643125115745327,
      "language_loss": 0.7991972,
      "learning_rate": 3.1716598327565694e-06,
      "loss": 0.82145965,
      "num_input_tokens_seen": 57672215,
      "step": 2676,
      "time_per_iteration": 2.7159581184387207
    },
    {
      "auxiliary_loss_clip": 0.01156248,
      "auxiliary_loss_mlp": 0.01030395,
      "balance_loss_clip": 1.05227268,
      "balance_loss_mlp": 1.02139509,
      "epoch": 0.3218902182408465,
      "flos": 19062533640960.0,
      "grad_norm": 5.594811424035353,
      "language_loss": 0.84165329,
      "learning_rate": 3.171028438775015e-06,
      "loss": 0.86351979,
      "num_input_tokens_seen": 57691410,
      "step": 2677,
      "time_per_iteration": 2.73075795173645
    },
    {
      "auxiliary_loss_clip": 0.01206331,
      "auxiliary_loss_mlp": 0.010346,
      "balance_loss_clip": 1.06162369,
      "balance_loss_mlp": 1.02613044,
      "epoch": 0.3220104611314856,
      "flos": 20375571306240.0,
      "grad_norm": 5.19982513647309,
      "language_loss": 0.83774388,
      "learning_rate": 3.170396867149377e-06,
      "loss": 0.86015314,
      "num_input_tokens_seen": 57709415,
      "step": 2678,
      "time_per_iteration": 2.6406805515289307
    },
    {
      "auxiliary_loss_clip": 0.01144613,
      "auxiliary_loss_mlp": 0.01037478,
      "balance_loss_clip": 1.05212927,
      "balance_loss_mlp": 1.02883577,
      "epoch": 0.3221307040221247,
      "flos": 20117014231680.0,
      "grad_norm": 3.5026374625836536,
      "language_loss": 0.87037468,
      "learning_rate": 3.1697651179754653e-06,
      "loss": 0.89219558,
      "num_input_tokens_seen": 57728075,
      "step": 2679,
      "time_per_iteration": 2.775526762008667
    },
    {
      "auxiliary_loss_clip": 0.01192864,
      "auxiliary_loss_mlp": 0.01036563,
      "balance_loss_clip": 1.06298256,
      "balance_loss_mlp": 1.02780676,
      "epoch": 0.3222509469127638,
      "flos": 23987789602560.0,
      "grad_norm": 3.339474367869919,
      "language_loss": 0.73743832,
      "learning_rate": 3.1691331913491153e-06,
      "loss": 0.75973254,
      "num_input_tokens_seen": 57750645,
      "step": 2680,
      "time_per_iteration": 2.7033026218414307
    },
    {
      "auxiliary_loss_clip": 0.01181348,
      "auxiliary_loss_mlp": 0.01036339,
      "balance_loss_clip": 1.05669737,
      "balance_loss_mlp": 1.02760673,
      "epoch": 0.32237118980340285,
      "flos": 17675735397120.0,
      "grad_norm": 8.310007991928702,
      "language_loss": 0.85059905,
      "learning_rate": 3.1685010873661898e-06,
      "loss": 0.87277591,
      "num_input_tokens_seen": 57769820,
      "step": 2681,
      "time_per_iteration": 2.630660057067871
    },
    {
      "auxiliary_loss_clip": 0.01171876,
      "auxiliary_loss_mlp": 0.01040514,
      "balance_loss_clip": 1.05537915,
      "balance_loss_mlp": 1.03092384,
      "epoch": 0.32249143269404196,
      "flos": 23147982645120.0,
      "grad_norm": 3.020150256873246,
      "language_loss": 0.80324185,
      "learning_rate": 3.167868806122578e-06,
      "loss": 0.82536572,
      "num_input_tokens_seen": 57788870,
      "step": 2682,
      "time_per_iteration": 2.7826855182647705
    },
    {
      "auxiliary_loss_clip": 0.01162512,
      "auxiliary_loss_mlp": 0.01035383,
      "balance_loss_clip": 1.05455256,
      "balance_loss_mlp": 1.02570915,
      "epoch": 0.32261167558468107,
      "flos": 24422308427520.0,
      "grad_norm": 2.257107807242508,
      "language_loss": 0.6677739,
      "learning_rate": 3.1672363477141968e-06,
      "loss": 0.68975294,
      "num_input_tokens_seen": 57808165,
      "step": 2683,
      "time_per_iteration": 2.706432342529297
    },
    {
      "auxiliary_loss_clip": 0.0115108,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.05655003,
      "balance_loss_mlp": 1.0224719,
      "epoch": 0.3227319184753201,
      "flos": 30367175852160.0,
      "grad_norm": 2.6682995182263616,
      "language_loss": 0.85296565,
      "learning_rate": 3.1666037122369903e-06,
      "loss": 0.87479138,
      "num_input_tokens_seen": 57828825,
      "step": 2684,
      "time_per_iteration": 4.746642827987671
    },
    {
      "auxiliary_loss_clip": 0.01208076,
      "auxiliary_loss_mlp": 0.01033016,
      "balance_loss_clip": 1.06162512,
      "balance_loss_mlp": 1.0241648,
      "epoch": 0.32285216136595923,
      "flos": 16946174257920.0,
      "grad_norm": 2.5442940727756707,
      "language_loss": 0.86519951,
      "learning_rate": 3.165970899786928e-06,
      "loss": 0.88761044,
      "num_input_tokens_seen": 57846740,
      "step": 2685,
      "time_per_iteration": 2.628183364868164
    },
    {
      "auxiliary_loss_clip": 0.01194707,
      "auxiliary_loss_mlp": 0.01032171,
      "balance_loss_clip": 1.05914092,
      "balance_loss_mlp": 1.02278948,
      "epoch": 0.32297240425659834,
      "flos": 21981532383360.0,
      "grad_norm": 2.384500006320033,
      "language_loss": 0.75475466,
      "learning_rate": 3.1653379104600067e-06,
      "loss": 0.77702349,
      "num_input_tokens_seen": 57866885,
      "step": 2686,
      "time_per_iteration": 2.714128255844116
    },
    {
      "auxiliary_loss_clip": 0.01134863,
      "auxiliary_loss_mlp": 0.01034298,
      "balance_loss_clip": 1.05558705,
      "balance_loss_mlp": 1.02471995,
      "epoch": 0.3230926471472374,
      "flos": 22748045639040.0,
      "grad_norm": 1.612879258082278,
      "language_loss": 0.69593221,
      "learning_rate": 3.164704744352251e-06,
      "loss": 0.71762377,
      "num_input_tokens_seen": 57887690,
      "step": 2687,
      "time_per_iteration": 3.8925602436065674
    },
    {
      "auxiliary_loss_clip": 0.01202467,
      "auxiliary_loss_mlp": 0.01032624,
      "balance_loss_clip": 1.0585537,
      "balance_loss_mlp": 1.02370095,
      "epoch": 0.3232128900378765,
      "flos": 16942977947520.0,
      "grad_norm": 1.894318316379945,
      "language_loss": 0.81298101,
      "learning_rate": 3.164071401559713e-06,
      "loss": 0.83533192,
      "num_input_tokens_seen": 57905090,
      "step": 2688,
      "time_per_iteration": 3.607353925704956
    },
    {
      "auxiliary_loss_clip": 0.01164274,
      "auxiliary_loss_mlp": 0.01037325,
      "balance_loss_clip": 1.05800915,
      "balance_loss_mlp": 1.02761519,
      "epoch": 0.3233331329285156,
      "flos": 24023736138240.0,
      "grad_norm": 2.1914221043205813,
      "language_loss": 0.71004164,
      "learning_rate": 3.1634378821784674e-06,
      "loss": 0.73205757,
      "num_input_tokens_seen": 57925305,
      "step": 2689,
      "time_per_iteration": 2.7662081718444824
    },
    {
      "auxiliary_loss_clip": 0.01183868,
      "auxiliary_loss_mlp": 0.01036926,
      "balance_loss_clip": 1.05678487,
      "balance_loss_mlp": 1.02760434,
      "epoch": 0.3234533758191547,
      "flos": 18113845582080.0,
      "grad_norm": 3.1178647391487244,
      "language_loss": 0.74458283,
      "learning_rate": 3.1628041863046208e-06,
      "loss": 0.76679081,
      "num_input_tokens_seen": 57942720,
      "step": 2690,
      "time_per_iteration": 2.666707754135132
    },
    {
      "auxiliary_loss_clip": 0.01180733,
      "auxiliary_loss_mlp": 0.01037405,
      "balance_loss_clip": 1.0567708,
      "balance_loss_mlp": 1.02832711,
      "epoch": 0.3235736187097938,
      "flos": 16946138344320.0,
      "grad_norm": 2.4784194287570394,
      "language_loss": 0.9179908,
      "learning_rate": 3.162170314034304e-06,
      "loss": 0.9401722,
      "num_input_tokens_seen": 57960135,
      "step": 2691,
      "time_per_iteration": 2.6959245204925537
    },
    {
      "auxiliary_loss_clip": 0.01188866,
      "auxiliary_loss_mlp": 0.0076415,
      "balance_loss_clip": 1.05591607,
      "balance_loss_mlp": 1.00078154,
      "epoch": 0.3236938616004329,
      "flos": 22127150119680.0,
      "grad_norm": 1.9873745844548376,
      "language_loss": 0.80769157,
      "learning_rate": 3.1615362654636738e-06,
      "loss": 0.82722175,
      "num_input_tokens_seen": 57980875,
      "step": 2692,
      "time_per_iteration": 2.7421631813049316
    },
    {
      "auxiliary_loss_clip": 0.01175335,
      "auxiliary_loss_mlp": 0.0103488,
      "balance_loss_clip": 1.05815136,
      "balance_loss_mlp": 1.0263294,
      "epoch": 0.32381410449107195,
      "flos": 17164618819200.0,
      "grad_norm": 1.9134598903237412,
      "language_loss": 0.87251425,
      "learning_rate": 3.1609020406889163e-06,
      "loss": 0.89461648,
      "num_input_tokens_seen": 57998310,
      "step": 2693,
      "time_per_iteration": 2.660247564315796
    },
    {
      "auxiliary_loss_clip": 0.01146935,
      "auxiliary_loss_mlp": 0.0104145,
      "balance_loss_clip": 1.05474567,
      "balance_loss_mlp": 1.03221118,
      "epoch": 0.32393434738171106,
      "flos": 16578125550720.0,
      "grad_norm": 1.8688942024058233,
      "language_loss": 0.84833622,
      "learning_rate": 3.1602676398062416e-06,
      "loss": 0.87022007,
      "num_input_tokens_seen": 58017220,
      "step": 2694,
      "time_per_iteration": 2.7441763877868652
    },
    {
      "auxiliary_loss_clip": 0.01177783,
      "auxiliary_loss_mlp": 0.0103112,
      "balance_loss_clip": 1.05659795,
      "balance_loss_mlp": 1.02217364,
      "epoch": 0.3240545902723502,
      "flos": 25483612602240.0,
      "grad_norm": 3.068932229236035,
      "language_loss": 0.61491984,
      "learning_rate": 3.1596330629118886e-06,
      "loss": 0.63700885,
      "num_input_tokens_seen": 58037190,
      "step": 2695,
      "time_per_iteration": 2.7249362468719482
    },
    {
      "auxiliary_loss_clip": 0.01146093,
      "auxiliary_loss_mlp": 0.01038534,
      "balance_loss_clip": 1.05402303,
      "balance_loss_mlp": 1.02890754,
      "epoch": 0.32417483316298923,
      "flos": 35845851634560.0,
      "grad_norm": 4.655693148107992,
      "language_loss": 0.73339534,
      "learning_rate": 3.1589983101021223e-06,
      "loss": 0.75524157,
      "num_input_tokens_seen": 58055820,
      "step": 2696,
      "time_per_iteration": 2.8385725021362305
    },
    {
      "auxiliary_loss_clip": 0.01188336,
      "auxiliary_loss_mlp": 0.01029391,
      "balance_loss_clip": 1.05818141,
      "balance_loss_mlp": 1.02074862,
      "epoch": 0.32429507605362834,
      "flos": 30080501406720.0,
      "grad_norm": 2.5075526154423398,
      "language_loss": 0.84681231,
      "learning_rate": 3.1583633814732337e-06,
      "loss": 0.86898959,
      "num_input_tokens_seen": 58075340,
      "step": 2697,
      "time_per_iteration": 2.7256720066070557
    },
    {
      "auxiliary_loss_clip": 0.01128667,
      "auxiliary_loss_mlp": 0.01037224,
      "balance_loss_clip": 1.04983366,
      "balance_loss_mlp": 1.02786648,
      "epoch": 0.3244153189442674,
      "flos": 18223265387520.0,
      "grad_norm": 2.703448007789777,
      "language_loss": 0.72365725,
      "learning_rate": 3.157728277121541e-06,
      "loss": 0.74531615,
      "num_input_tokens_seen": 58093515,
      "step": 2698,
      "time_per_iteration": 2.7723355293273926
    },
    {
      "auxiliary_loss_clip": 0.01184836,
      "auxiliary_loss_mlp": 0.00763725,
      "balance_loss_clip": 1.05580258,
      "balance_loss_mlp": 1.0008297,
      "epoch": 0.3245355618349065,
      "flos": 17710317216000.0,
      "grad_norm": 3.4553669437962236,
      "language_loss": 0.78516269,
      "learning_rate": 3.1570929971433897e-06,
      "loss": 0.80464822,
      "num_input_tokens_seen": 58109300,
      "step": 2699,
      "time_per_iteration": 2.7205123901367188
    },
    {
      "auxiliary_loss_clip": 0.01165114,
      "auxiliary_loss_mlp": 0.01044468,
      "balance_loss_clip": 1.05519521,
      "balance_loss_mlp": 1.0355866,
      "epoch": 0.3246558047255456,
      "flos": 23440798316160.0,
      "grad_norm": 2.0988224232834156,
      "language_loss": 0.84463364,
      "learning_rate": 3.1564575416351504e-06,
      "loss": 0.8667295,
      "num_input_tokens_seen": 58128000,
      "step": 2700,
      "time_per_iteration": 2.721862316131592
    },
    {
      "auxiliary_loss_clip": 0.01182668,
      "auxiliary_loss_mlp": 0.01034831,
      "balance_loss_clip": 1.05948114,
      "balance_loss_mlp": 1.02543104,
      "epoch": 0.32477604761618467,
      "flos": 21760861178880.0,
      "grad_norm": 2.7656840233175504,
      "language_loss": 0.73882741,
      "learning_rate": 3.155821910693221e-06,
      "loss": 0.76100242,
      "num_input_tokens_seen": 58147415,
      "step": 2701,
      "time_per_iteration": 2.7870936393737793
    },
    {
      "auxiliary_loss_clip": 0.01142851,
      "auxiliary_loss_mlp": 0.0103403,
      "balance_loss_clip": 1.05104661,
      "balance_loss_mlp": 1.02454066,
      "epoch": 0.3248962905068238,
      "flos": 19828328624640.0,
      "grad_norm": 2.6959065964382125,
      "language_loss": 0.86376452,
      "learning_rate": 3.1551861044140275e-06,
      "loss": 0.88553333,
      "num_input_tokens_seen": 58167050,
      "step": 2702,
      "time_per_iteration": 2.79902720451355
    },
    {
      "auxiliary_loss_clip": 0.01148791,
      "auxiliary_loss_mlp": 0.01039647,
      "balance_loss_clip": 1.06144392,
      "balance_loss_mlp": 1.03062892,
      "epoch": 0.3250165333974629,
      "flos": 23948215793280.0,
      "grad_norm": 2.2834583465289895,
      "language_loss": 0.78133798,
      "learning_rate": 3.15455012289402e-06,
      "loss": 0.80322236,
      "num_input_tokens_seen": 58186695,
      "step": 2703,
      "time_per_iteration": 2.9903879165649414
    },
    {
      "auxiliary_loss_clip": 0.01189448,
      "auxiliary_loss_mlp": 0.01041337,
      "balance_loss_clip": 1.0595243,
      "balance_loss_mlp": 1.0316093,
      "epoch": 0.32513677628810195,
      "flos": 23989333887360.0,
      "grad_norm": 2.116403985880881,
      "language_loss": 0.844401,
      "learning_rate": 3.153913966229677e-06,
      "loss": 0.86670887,
      "num_input_tokens_seen": 58205815,
      "step": 2704,
      "time_per_iteration": 2.7718729972839355
    },
    {
      "auxiliary_loss_clip": 0.01077679,
      "auxiliary_loss_mlp": 0.0100495,
      "balance_loss_clip": 1.02005482,
      "balance_loss_mlp": 1.00336444,
      "epoch": 0.32525701917874106,
      "flos": 70655790009600.0,
      "grad_norm": 0.6428726119637393,
      "language_loss": 0.50196218,
      "learning_rate": 3.1532776345175027e-06,
      "loss": 0.52278852,
      "num_input_tokens_seen": 58270960,
      "step": 2705,
      "time_per_iteration": 3.2365260124206543
    },
    {
      "auxiliary_loss_clip": 0.01168435,
      "auxiliary_loss_mlp": 0.01033529,
      "balance_loss_clip": 1.05402875,
      "balance_loss_mlp": 1.02450502,
      "epoch": 0.32537726206938017,
      "flos": 19682639061120.0,
      "grad_norm": 2.500860295250161,
      "language_loss": 0.78757286,
      "learning_rate": 3.1526411278540285e-06,
      "loss": 0.80959249,
      "num_input_tokens_seen": 58289390,
      "step": 2706,
      "time_per_iteration": 2.690441846847534
    },
    {
      "auxiliary_loss_clip": 0.01168386,
      "auxiliary_loss_mlp": 0.01034267,
      "balance_loss_clip": 1.05899203,
      "balance_loss_mlp": 1.02478421,
      "epoch": 0.3254975049600192,
      "flos": 28760999293440.0,
      "grad_norm": 2.6894859068679415,
      "language_loss": 0.81479704,
      "learning_rate": 3.1520044463358116e-06,
      "loss": 0.83682358,
      "num_input_tokens_seen": 58306120,
      "step": 2707,
      "time_per_iteration": 2.7554855346679688
    },
    {
      "auxiliary_loss_clip": 0.01170559,
      "auxiliary_loss_mlp": 0.01034467,
      "balance_loss_clip": 1.06060696,
      "balance_loss_mlp": 1.02507877,
      "epoch": 0.32561774785065833,
      "flos": 18877378008960.0,
      "grad_norm": 2.2384733341511476,
      "language_loss": 0.80264968,
      "learning_rate": 3.151367590059436e-06,
      "loss": 0.82469994,
      "num_input_tokens_seen": 58324545,
      "step": 2708,
      "time_per_iteration": 2.78891658782959
    },
    {
      "auxiliary_loss_clip": 0.01154681,
      "auxiliary_loss_mlp": 0.00763678,
      "balance_loss_clip": 1.05853939,
      "balance_loss_mlp": 1.00076985,
      "epoch": 0.32573799074129745,
      "flos": 23112107936640.0,
      "grad_norm": 6.62384337972041,
      "language_loss": 0.8702966,
      "learning_rate": 3.1507305591215117e-06,
      "loss": 0.88948023,
      "num_input_tokens_seen": 58342455,
      "step": 2709,
      "time_per_iteration": 3.765852928161621
    },
    {
      "auxiliary_loss_clip": 0.01087838,
      "auxiliary_loss_mlp": 0.01002128,
      "balance_loss_clip": 1.01981461,
      "balance_loss_mlp": 1.00056684,
      "epoch": 0.3258582336319365,
      "flos": 71237650423680.0,
      "grad_norm": 0.6919946403058146,
      "language_loss": 0.55609703,
      "learning_rate": 3.150093353618677e-06,
      "loss": 0.57699668,
      "num_input_tokens_seen": 58407185,
      "step": 2710,
      "time_per_iteration": 4.284544944763184
    },
    {
      "auxiliary_loss_clip": 0.01140554,
      "auxiliary_loss_mlp": 0.01034696,
      "balance_loss_clip": 1.04938972,
      "balance_loss_mlp": 1.02505803,
      "epoch": 0.3259784765225756,
      "flos": 22456020067200.0,
      "grad_norm": 2.6715199473038735,
      "language_loss": 0.88355094,
      "learning_rate": 3.149455973647596e-06,
      "loss": 0.90530342,
      "num_input_tokens_seen": 58425245,
      "step": 2711,
      "time_per_iteration": 2.7456254959106445
    },
    {
      "auxiliary_loss_clip": 0.01193703,
      "auxiliary_loss_mlp": 0.00764235,
      "balance_loss_clip": 1.05805194,
      "balance_loss_mlp": 1.00095427,
      "epoch": 0.32609871941321467,
      "flos": 20484811543680.0,
      "grad_norm": 2.5939080350436905,
      "language_loss": 0.7759161,
      "learning_rate": 3.1488184193049563e-06,
      "loss": 0.79549551,
      "num_input_tokens_seen": 58444780,
      "step": 2712,
      "time_per_iteration": 2.703444719314575
    },
    {
      "auxiliary_loss_clip": 0.01159393,
      "auxiliary_loss_mlp": 0.01035702,
      "balance_loss_clip": 1.0542438,
      "balance_loss_mlp": 1.02671993,
      "epoch": 0.3262189623038538,
      "flos": 22416805393920.0,
      "grad_norm": 1.897619842891075,
      "language_loss": 0.72281277,
      "learning_rate": 3.1481806906874767e-06,
      "loss": 0.74476373,
      "num_input_tokens_seen": 58466090,
      "step": 2713,
      "time_per_iteration": 3.6597037315368652
    },
    {
      "auxiliary_loss_clip": 0.01139418,
      "auxiliary_loss_mlp": 0.01038544,
      "balance_loss_clip": 1.05549693,
      "balance_loss_mlp": 1.03012824,
      "epoch": 0.3263392051944929,
      "flos": 20923496346240.0,
      "grad_norm": 2.1414949937811705,
      "language_loss": 0.87851143,
      "learning_rate": 3.147542787891899e-06,
      "loss": 0.90029097,
      "num_input_tokens_seen": 58485435,
      "step": 2714,
      "time_per_iteration": 3.7239551544189453
    },
    {
      "auxiliary_loss_clip": 0.01131329,
      "auxiliary_loss_mlp": 0.00763644,
      "balance_loss_clip": 1.04853117,
      "balance_loss_mlp": 1.00083709,
      "epoch": 0.32645944808513194,
      "flos": 24025172682240.0,
      "grad_norm": 2.1084619490021783,
      "language_loss": 0.75876248,
      "learning_rate": 3.1469047110149926e-06,
      "loss": 0.77771223,
      "num_input_tokens_seen": 58504175,
      "step": 2715,
      "time_per_iteration": 2.7813727855682373
    },
    {
      "auxiliary_loss_clip": 0.01178466,
      "auxiliary_loss_mlp": 0.01039286,
      "balance_loss_clip": 1.06059122,
      "balance_loss_mlp": 1.03036904,
      "epoch": 0.32657969097577105,
      "flos": 21032413361280.0,
      "grad_norm": 3.5410406721262104,
      "language_loss": 0.85436273,
      "learning_rate": 3.146266460153554e-06,
      "loss": 0.87654018,
      "num_input_tokens_seen": 58523885,
      "step": 2716,
      "time_per_iteration": 2.6731386184692383
    },
    {
      "auxiliary_loss_clip": 0.01162745,
      "auxiliary_loss_mlp": 0.01048052,
      "balance_loss_clip": 1.05738366,
      "balance_loss_mlp": 1.0385927,
      "epoch": 0.32669993386641016,
      "flos": 22710267509760.0,
      "grad_norm": 1.807418321813281,
      "language_loss": 0.80445981,
      "learning_rate": 3.145628035404404e-06,
      "loss": 0.82656777,
      "num_input_tokens_seen": 58543085,
      "step": 2717,
      "time_per_iteration": 2.7795543670654297
    },
    {
      "auxiliary_loss_clip": 0.01077447,
      "auxiliary_loss_mlp": 0.01002533,
      "balance_loss_clip": 1.01961315,
      "balance_loss_mlp": 1.00089979,
      "epoch": 0.3268201767570492,
      "flos": 72105718406400.0,
      "grad_norm": 0.8890480524456754,
      "language_loss": 0.57474756,
      "learning_rate": 3.1449894368643922e-06,
      "loss": 0.59554732,
      "num_input_tokens_seen": 58605400,
      "step": 2718,
      "time_per_iteration": 3.2893331050872803
    },
    {
      "auxiliary_loss_clip": 0.01194206,
      "auxiliary_loss_mlp": 0.01039552,
      "balance_loss_clip": 1.06654489,
      "balance_loss_mlp": 1.0311296,
      "epoch": 0.32694041964768833,
      "flos": 24535175938560.0,
      "grad_norm": 1.6962729943484314,
      "language_loss": 0.72094083,
      "learning_rate": 3.1443506646303934e-06,
      "loss": 0.74327838,
      "num_input_tokens_seen": 58626700,
      "step": 2719,
      "time_per_iteration": 2.7943708896636963
    },
    {
      "auxiliary_loss_clip": 0.01173715,
      "auxiliary_loss_mlp": 0.01031384,
      "balance_loss_clip": 1.05503154,
      "balance_loss_mlp": 1.0222404,
      "epoch": 0.32706066253832744,
      "flos": 33183003755520.0,
      "grad_norm": 3.544522157327054,
      "language_loss": 0.67167246,
      "learning_rate": 3.1437117187993086e-06,
      "loss": 0.69372344,
      "num_input_tokens_seen": 58649020,
      "step": 2720,
      "time_per_iteration": 2.777027130126953
    },
    {
      "auxiliary_loss_clip": 0.01109679,
      "auxiliary_loss_mlp": 0.01041108,
      "balance_loss_clip": 1.05091488,
      "balance_loss_mlp": 1.03181601,
      "epoch": 0.3271809054289665,
      "flos": 24061622008320.0,
      "grad_norm": 2.001957028617369,
      "language_loss": 0.79861891,
      "learning_rate": 3.143072599468065e-06,
      "loss": 0.82012677,
      "num_input_tokens_seen": 58668845,
      "step": 2721,
      "time_per_iteration": 2.8641507625579834
    },
    {
      "auxiliary_loss_clip": 0.01179057,
      "auxiliary_loss_mlp": 0.01036677,
      "balance_loss_clip": 1.05895567,
      "balance_loss_mlp": 1.02741408,
      "epoch": 0.3273011483196056,
      "flos": 38253769712640.0,
      "grad_norm": 1.6161996196090986,
      "language_loss": 0.75423789,
      "learning_rate": 3.1424333067336174e-06,
      "loss": 0.7763952,
      "num_input_tokens_seen": 58691610,
      "step": 2722,
      "time_per_iteration": 2.850463390350342
    },
    {
      "auxiliary_loss_clip": 0.01153625,
      "auxiliary_loss_mlp": 0.0103549,
      "balance_loss_clip": 1.04908133,
      "balance_loss_mlp": 1.0260371,
      "epoch": 0.3274213912102447,
      "flos": 29054389582080.0,
      "grad_norm": 1.882371132332854,
      "language_loss": 0.78246212,
      "learning_rate": 3.141793840692945e-06,
      "loss": 0.80435324,
      "num_input_tokens_seen": 58712360,
      "step": 2723,
      "time_per_iteration": 2.7542548179626465
    },
    {
      "auxiliary_loss_clip": 0.01197213,
      "auxiliary_loss_mlp": 0.01035663,
      "balance_loss_clip": 1.0591532,
      "balance_loss_mlp": 1.02606654,
      "epoch": 0.32754163410088377,
      "flos": 29133249891840.0,
      "grad_norm": 3.402800184785409,
      "language_loss": 0.6147083,
      "learning_rate": 3.1411542014430553e-06,
      "loss": 0.6370371,
      "num_input_tokens_seen": 58733440,
      "step": 2724,
      "time_per_iteration": 2.7455132007598877
    },
    {
      "auxiliary_loss_clip": 0.01177736,
      "auxiliary_loss_mlp": 0.01031529,
      "balance_loss_clip": 1.05641222,
      "balance_loss_mlp": 1.02323222,
      "epoch": 0.3276618769915229,
      "flos": 20631075724800.0,
      "grad_norm": 2.421744323422328,
      "language_loss": 0.81921971,
      "learning_rate": 3.1405143890809804e-06,
      "loss": 0.84131229,
      "num_input_tokens_seen": 58752735,
      "step": 2725,
      "time_per_iteration": 2.7213568687438965
    },
    {
      "auxiliary_loss_clip": 0.01191143,
      "auxiliary_loss_mlp": 0.0103936,
      "balance_loss_clip": 1.0623914,
      "balance_loss_mlp": 1.03050327,
      "epoch": 0.327782119882162,
      "flos": 18657425076480.0,
      "grad_norm": 2.9212637982132605,
      "language_loss": 0.7033838,
      "learning_rate": 3.1398744037037796e-06,
      "loss": 0.72568887,
      "num_input_tokens_seen": 58772070,
      "step": 2726,
      "time_per_iteration": 2.6775317192077637
    },
    {
      "auxiliary_loss_clip": 0.01199267,
      "auxiliary_loss_mlp": 0.01036739,
      "balance_loss_clip": 1.06356621,
      "balance_loss_mlp": 1.02749395,
      "epoch": 0.32790236277280105,
      "flos": 21795802133760.0,
      "grad_norm": 2.034709347925944,
      "language_loss": 0.84214222,
      "learning_rate": 3.139234245408538e-06,
      "loss": 0.86450219,
      "num_input_tokens_seen": 58790950,
      "step": 2727,
      "time_per_iteration": 2.604055881500244
    },
    {
      "auxiliary_loss_clip": 0.01148764,
      "auxiliary_loss_mlp": 0.01034291,
      "balance_loss_clip": 1.05276382,
      "balance_loss_mlp": 1.02493918,
      "epoch": 0.32802260566344016,
      "flos": 23331414424320.0,
      "grad_norm": 1.6817035704675423,
      "language_loss": 0.76379609,
      "learning_rate": 3.1385939142923666e-06,
      "loss": 0.78562665,
      "num_input_tokens_seen": 58813340,
      "step": 2728,
      "time_per_iteration": 2.8203892707824707
    },
    {
      "auxiliary_loss_clip": 0.01165282,
      "auxiliary_loss_mlp": 0.01030815,
      "balance_loss_clip": 1.06005681,
      "balance_loss_mlp": 1.02120721,
      "epoch": 0.3281428485540792,
      "flos": 24206988349440.0,
      "grad_norm": 2.379695168217942,
      "language_loss": 0.78070498,
      "learning_rate": 3.137953410452405e-06,
      "loss": 0.80266601,
      "num_input_tokens_seen": 58833610,
      "step": 2729,
      "time_per_iteration": 2.75319504737854
    },
    {
      "auxiliary_loss_clip": 0.01169858,
      "auxiliary_loss_mlp": 0.00763574,
      "balance_loss_clip": 1.05507231,
      "balance_loss_mlp": 1.00075114,
      "epoch": 0.3282630914447183,
      "flos": 34128962380800.0,
      "grad_norm": 1.9659620236264106,
      "language_loss": 0.74908388,
      "learning_rate": 3.1373127339858146e-06,
      "loss": 0.76841819,
      "num_input_tokens_seen": 58856210,
      "step": 2730,
      "time_per_iteration": 2.796799898147583
    },
    {
      "auxiliary_loss_clip": 0.01172949,
      "auxiliary_loss_mlp": 0.01034947,
      "balance_loss_clip": 1.05625474,
      "balance_loss_mlp": 1.0261668,
      "epoch": 0.32838333433535744,
      "flos": 27600726170880.0,
      "grad_norm": 2.0390653088973543,
      "language_loss": 0.74634647,
      "learning_rate": 3.136671884989787e-06,
      "loss": 0.76842546,
      "num_input_tokens_seen": 58876120,
      "step": 2731,
      "time_per_iteration": 2.7443349361419678
    },
    {
      "auxiliary_loss_clip": 0.0120209,
      "auxiliary_loss_mlp": 0.01034564,
      "balance_loss_clip": 1.06219244,
      "balance_loss_mlp": 1.025617,
      "epoch": 0.3285035772259965,
      "flos": 12349500935040.0,
      "grad_norm": 2.421209760792761,
      "language_loss": 0.87381256,
      "learning_rate": 3.1360308635615383e-06,
      "loss": 0.89617908,
      "num_input_tokens_seen": 58894660,
      "step": 2732,
      "time_per_iteration": 2.625685214996338
    },
    {
      "auxiliary_loss_clip": 0.01197034,
      "auxiliary_loss_mlp": 0.01038969,
      "balance_loss_clip": 1.06122375,
      "balance_loss_mlp": 1.0289669,
      "epoch": 0.3286238201166356,
      "flos": 24316084932480.0,
      "grad_norm": 2.156323007576704,
      "language_loss": 0.78905153,
      "learning_rate": 3.135389669798311e-06,
      "loss": 0.8114115,
      "num_input_tokens_seen": 58912720,
      "step": 2733,
      "time_per_iteration": 2.728301525115967
    },
    {
      "auxiliary_loss_clip": 0.01147495,
      "auxiliary_loss_mlp": 0.01044444,
      "balance_loss_clip": 1.05212426,
      "balance_loss_mlp": 1.03524137,
      "epoch": 0.3287440630072747,
      "flos": 21392812471680.0,
      "grad_norm": 2.5839091539020758,
      "language_loss": 0.8013283,
      "learning_rate": 3.134748303797373e-06,
      "loss": 0.82324767,
      "num_input_tokens_seen": 58930090,
      "step": 2734,
      "time_per_iteration": 2.741848945617676
    },
    {
      "auxiliary_loss_clip": 0.01169991,
      "auxiliary_loss_mlp": 0.0076471,
      "balance_loss_clip": 1.05431628,
      "balance_loss_mlp": 1.00072742,
      "epoch": 0.32886430589791377,
      "flos": 23732536579200.0,
      "grad_norm": 2.0713155830380408,
      "language_loss": 0.81630242,
      "learning_rate": 3.1341067656560203e-06,
      "loss": 0.83564937,
      "num_input_tokens_seen": 58947935,
      "step": 2735,
      "time_per_iteration": 3.6654045581817627
    },
    {
      "auxiliary_loss_clip": 0.01130308,
      "auxiliary_loss_mlp": 0.01030433,
      "balance_loss_clip": 1.04855824,
      "balance_loss_mlp": 1.02064586,
      "epoch": 0.3289845487885529,
      "flos": 22418708814720.0,
      "grad_norm": 4.228074683896645,
      "language_loss": 0.86455655,
      "learning_rate": 3.133465055471572e-06,
      "loss": 0.88616395,
      "num_input_tokens_seen": 58967720,
      "step": 2736,
      "time_per_iteration": 3.7835330963134766
    },
    {
      "auxiliary_loss_clip": 0.01198877,
      "auxiliary_loss_mlp": 0.01033913,
      "balance_loss_clip": 1.06136501,
      "balance_loss_mlp": 1.02395892,
      "epoch": 0.329104791679192,
      "flos": 19682603147520.0,
      "grad_norm": 2.3189767183950143,
      "language_loss": 0.66288251,
      "learning_rate": 3.1328231733413767e-06,
      "loss": 0.68521047,
      "num_input_tokens_seen": 58984360,
      "step": 2737,
      "time_per_iteration": 2.6031172275543213
    },
    {
      "auxiliary_loss_clip": 0.01187297,
      "auxiliary_loss_mlp": 0.01037829,
      "balance_loss_clip": 1.0605092,
      "balance_loss_mlp": 1.02816081,
      "epoch": 0.32922503456983104,
      "flos": 15997234803840.0,
      "grad_norm": 2.4386354795688874,
      "language_loss": 0.91269559,
      "learning_rate": 3.1321811193628067e-06,
      "loss": 0.93494684,
      "num_input_tokens_seen": 59002505,
      "step": 2738,
      "time_per_iteration": 2.6804986000061035
    },
    {
      "auxiliary_loss_clip": 0.01168619,
      "auxiliary_loss_mlp": 0.01033653,
      "balance_loss_clip": 1.05742335,
      "balance_loss_mlp": 1.02462327,
      "epoch": 0.32934527746047015,
      "flos": 26834069260800.0,
      "grad_norm": 2.572597538269773,
      "language_loss": 0.70087135,
      "learning_rate": 3.131538893633261e-06,
      "loss": 0.72289413,
      "num_input_tokens_seen": 59022065,
      "step": 2739,
      "time_per_iteration": 2.7370500564575195
    },
    {
      "auxiliary_loss_clip": 0.01195613,
      "auxiliary_loss_mlp": 0.01038319,
      "balance_loss_clip": 1.0626595,
      "balance_loss_mlp": 1.02937245,
      "epoch": 0.32946552035110926,
      "flos": 23403774372480.0,
      "grad_norm": 14.708065157740611,
      "language_loss": 0.78656036,
      "learning_rate": 3.130896496250165e-06,
      "loss": 0.80889964,
      "num_input_tokens_seen": 59041890,
      "step": 2740,
      "time_per_iteration": 4.466892957687378
    },
    {
      "auxiliary_loss_clip": 0.01133747,
      "auxiliary_loss_mlp": 0.01033204,
      "balance_loss_clip": 1.04758859,
      "balance_loss_mlp": 1.02381635,
      "epoch": 0.3295857632417483,
      "flos": 14172470029440.0,
      "grad_norm": 2.3553177492537696,
      "language_loss": 0.87056154,
      "learning_rate": 3.1302539273109693e-06,
      "loss": 0.89223111,
      "num_input_tokens_seen": 59058715,
      "step": 2741,
      "time_per_iteration": 2.731515407562256
    },
    {
      "auxiliary_loss_clip": 0.01213833,
      "auxiliary_loss_mlp": 0.01032821,
      "balance_loss_clip": 1.06439459,
      "balance_loss_mlp": 1.02321231,
      "epoch": 0.32970600613238743,
      "flos": 22196708807040.0,
      "grad_norm": 5.651938048873556,
      "language_loss": 0.80564594,
      "learning_rate": 3.1296111869131513e-06,
      "loss": 0.82811248,
      "num_input_tokens_seen": 59076140,
      "step": 2742,
      "time_per_iteration": 2.5864248275756836
    },
    {
      "auxiliary_loss_clip": 0.01130951,
      "auxiliary_loss_mlp": 0.0103158,
      "balance_loss_clip": 1.04827762,
      "balance_loss_mlp": 1.0221982,
      "epoch": 0.32982624902302654,
      "flos": 22053784590720.0,
      "grad_norm": 1.985624383383351,
      "language_loss": 0.86111951,
      "learning_rate": 3.1289682751542153e-06,
      "loss": 0.88274485,
      "num_input_tokens_seen": 59095700,
      "step": 2743,
      "time_per_iteration": 2.8442890644073486
    },
    {
      "auxiliary_loss_clip": 0.01177068,
      "auxiliary_loss_mlp": 0.01036683,
      "balance_loss_clip": 1.06024694,
      "balance_loss_mlp": 1.02761126,
      "epoch": 0.3299464919136656,
      "flos": 18661626967680.0,
      "grad_norm": 4.733973505551551,
      "language_loss": 0.71579015,
      "learning_rate": 3.1283251921316883e-06,
      "loss": 0.73792762,
      "num_input_tokens_seen": 59113445,
      "step": 2744,
      "time_per_iteration": 2.6913793087005615
    },
    {
      "auxiliary_loss_clip": 0.0115336,
      "auxiliary_loss_mlp": 0.01036053,
      "balance_loss_clip": 1.05525863,
      "balance_loss_mlp": 1.02713656,
      "epoch": 0.3300667348043047,
      "flos": 13407357404160.0,
      "grad_norm": 2.1781751064881294,
      "language_loss": 0.81083655,
      "learning_rate": 3.1276819379431277e-06,
      "loss": 0.83273077,
      "num_input_tokens_seen": 59131535,
      "step": 2745,
      "time_per_iteration": 2.755532741546631
    },
    {
      "auxiliary_loss_clip": 0.01166741,
      "auxiliary_loss_mlp": 0.01032377,
      "balance_loss_clip": 1.05516601,
      "balance_loss_mlp": 1.02280474,
      "epoch": 0.33018697769494376,
      "flos": 15742556398080.0,
      "grad_norm": 3.549209911124886,
      "language_loss": 0.75539237,
      "learning_rate": 3.1270385126861134e-06,
      "loss": 0.77738357,
      "num_input_tokens_seen": 59149520,
      "step": 2746,
      "time_per_iteration": 2.6713268756866455
    },
    {
      "auxiliary_loss_clip": 0.01110272,
      "auxiliary_loss_mlp": 0.01035321,
      "balance_loss_clip": 1.0434829,
      "balance_loss_mlp": 1.02601087,
      "epoch": 0.3303072205855829,
      "flos": 18258601392000.0,
      "grad_norm": 2.444755621239502,
      "language_loss": 0.82247734,
      "learning_rate": 3.1263949164582533e-06,
      "loss": 0.84393334,
      "num_input_tokens_seen": 59169170,
      "step": 2747,
      "time_per_iteration": 2.8276519775390625
    },
    {
      "auxiliary_loss_clip": 0.01188722,
      "auxiliary_loss_mlp": 0.01037284,
      "balance_loss_clip": 1.05685937,
      "balance_loss_mlp": 1.02719903,
      "epoch": 0.330427463476222,
      "flos": 17749424148480.0,
      "grad_norm": 4.273792443915818,
      "language_loss": 0.78414321,
      "learning_rate": 3.1257511493571797e-06,
      "loss": 0.80640334,
      "num_input_tokens_seen": 59187675,
      "step": 2748,
      "time_per_iteration": 2.6583003997802734
    },
    {
      "auxiliary_loss_clip": 0.01189991,
      "auxiliary_loss_mlp": 0.01030772,
      "balance_loss_clip": 1.05947447,
      "balance_loss_mlp": 1.02230227,
      "epoch": 0.33054770636686104,
      "flos": 27162580072320.0,
      "grad_norm": 2.400966691894998,
      "language_loss": 0.78532982,
      "learning_rate": 3.125107211480552e-06,
      "loss": 0.80753744,
      "num_input_tokens_seen": 59207610,
      "step": 2749,
      "time_per_iteration": 2.627408027648926
    },
    {
      "auxiliary_loss_clip": 0.01143561,
      "auxiliary_loss_mlp": 0.01041065,
      "balance_loss_clip": 1.05285001,
      "balance_loss_mlp": 1.03167701,
      "epoch": 0.33066794925750015,
      "flos": 20117193799680.0,
      "grad_norm": 2.3887886951162063,
      "language_loss": 0.80069596,
      "learning_rate": 3.124463102926054e-06,
      "loss": 0.82254219,
      "num_input_tokens_seen": 59226945,
      "step": 2750,
      "time_per_iteration": 2.766685724258423
    },
    {
      "auxiliary_loss_clip": 0.01086603,
      "auxiliary_loss_mlp": 0.01004089,
      "balance_loss_clip": 1.01944077,
      "balance_loss_mlp": 1.00237203,
      "epoch": 0.33078819214813926,
      "flos": 70642609718400.0,
      "grad_norm": 0.7609894560645497,
      "language_loss": 0.61565679,
      "learning_rate": 3.1238188237913984e-06,
      "loss": 0.63656372,
      "num_input_tokens_seen": 59291485,
      "step": 2751,
      "time_per_iteration": 3.2948365211486816
    },
    {
      "auxiliary_loss_clip": 0.01197759,
      "auxiliary_loss_mlp": 0.0103093,
      "balance_loss_clip": 1.06403923,
      "balance_loss_mlp": 1.02084517,
      "epoch": 0.3309084350387783,
      "flos": 21141940907520.0,
      "grad_norm": 2.360769036499667,
      "language_loss": 0.76547372,
      "learning_rate": 3.1231743741743202e-06,
      "loss": 0.78776067,
      "num_input_tokens_seen": 59310990,
      "step": 2752,
      "time_per_iteration": 2.705509662628174
    },
    {
      "auxiliary_loss_clip": 0.01171166,
      "auxiliary_loss_mlp": 0.01027938,
      "balance_loss_clip": 1.05322433,
      "balance_loss_mlp": 1.0182817,
      "epoch": 0.3310286779294174,
      "flos": 14209350318720.0,
      "grad_norm": 5.8077681718596565,
      "language_loss": 0.84158576,
      "learning_rate": 3.122529754172582e-06,
      "loss": 0.86357683,
      "num_input_tokens_seen": 59327875,
      "step": 2753,
      "time_per_iteration": 2.613194465637207
    },
    {
      "auxiliary_loss_clip": 0.01170892,
      "auxiliary_loss_mlp": 0.00763645,
      "balance_loss_clip": 1.05722058,
      "balance_loss_mlp": 1.00066495,
      "epoch": 0.33114892082005654,
      "flos": 20778130005120.0,
      "grad_norm": 2.174971439192687,
      "language_loss": 0.72852314,
      "learning_rate": 3.1218849638839736e-06,
      "loss": 0.74786854,
      "num_input_tokens_seen": 59347135,
      "step": 2754,
      "time_per_iteration": 2.6745071411132812
    },
    {
      "auxiliary_loss_clip": 0.01144919,
      "auxiliary_loss_mlp": 0.01037427,
      "balance_loss_clip": 1.05342746,
      "balance_loss_mlp": 1.02699602,
      "epoch": 0.3312691637106956,
      "flos": 17090750499840.0,
      "grad_norm": 2.441036071481381,
      "language_loss": 0.78677475,
      "learning_rate": 3.121240003406307e-06,
      "loss": 0.80859816,
      "num_input_tokens_seen": 59365985,
      "step": 2755,
      "time_per_iteration": 2.7769954204559326
    },
    {
      "auxiliary_loss_clip": 0.01138479,
      "auxiliary_loss_mlp": 0.01036749,
      "balance_loss_clip": 1.05321336,
      "balance_loss_mlp": 1.02585363,
      "epoch": 0.3313894066013347,
      "flos": 29456230008960.0,
      "grad_norm": 2.548758906557547,
      "language_loss": 0.73032135,
      "learning_rate": 3.120594872837425e-06,
      "loss": 0.75207365,
      "num_input_tokens_seen": 59384655,
      "step": 2756,
      "time_per_iteration": 2.8569867610931396
    },
    {
      "auxiliary_loss_clip": 0.01087331,
      "auxiliary_loss_mlp": 0.01003259,
      "balance_loss_clip": 1.01891768,
      "balance_loss_mlp": 1.00153065,
      "epoch": 0.3315096494919738,
      "flos": 61419242280960.0,
      "grad_norm": 0.8271922763000976,
      "language_loss": 0.62388927,
      "learning_rate": 3.1199495722751906e-06,
      "loss": 0.64479518,
      "num_input_tokens_seen": 59444185,
      "step": 2757,
      "time_per_iteration": 3.253162384033203
    },
    {
      "auxiliary_loss_clip": 0.01187951,
      "auxiliary_loss_mlp": 0.01033649,
      "balance_loss_clip": 1.06316113,
      "balance_loss_mlp": 1.02445769,
      "epoch": 0.33162989238261287,
      "flos": 21653057485440.0,
      "grad_norm": 8.80083840847275,
      "language_loss": 0.83719099,
      "learning_rate": 3.1193041018174972e-06,
      "loss": 0.85940707,
      "num_input_tokens_seen": 59464900,
      "step": 2758,
      "time_per_iteration": 2.8025786876678467
    },
    {
      "auxiliary_loss_clip": 0.01160964,
      "auxiliary_loss_mlp": 0.01037528,
      "balance_loss_clip": 1.05521798,
      "balance_loss_mlp": 1.02788985,
      "epoch": 0.331750135273252,
      "flos": 22674787850880.0,
      "grad_norm": 7.77209803180419,
      "language_loss": 0.95127857,
      "learning_rate": 3.118658461562261e-06,
      "loss": 0.97326344,
      "num_input_tokens_seen": 59481000,
      "step": 2759,
      "time_per_iteration": 2.7535061836242676
    },
    {
      "auxiliary_loss_clip": 0.01140007,
      "auxiliary_loss_mlp": 0.01039158,
      "balance_loss_clip": 1.05456686,
      "balance_loss_mlp": 1.03006792,
      "epoch": 0.33187037816389103,
      "flos": 22746896403840.0,
      "grad_norm": 2.203044174064835,
      "language_loss": 0.85017443,
      "learning_rate": 3.118012651607426e-06,
      "loss": 0.871966,
      "num_input_tokens_seen": 59502605,
      "step": 2760,
      "time_per_iteration": 2.828260898590088
    },
    {
      "auxiliary_loss_clip": 0.01143235,
      "auxiliary_loss_mlp": 0.01039574,
      "balance_loss_clip": 1.05657971,
      "balance_loss_mlp": 1.03033543,
      "epoch": 0.33199062105453014,
      "flos": 19203769918080.0,
      "grad_norm": 2.6254864426186737,
      "language_loss": 0.84038138,
      "learning_rate": 3.1173666720509603e-06,
      "loss": 0.86220944,
      "num_input_tokens_seen": 59519540,
      "step": 2761,
      "time_per_iteration": 3.6802027225494385
    },
    {
      "auxiliary_loss_clip": 0.01155967,
      "auxiliary_loss_mlp": 0.01031165,
      "balance_loss_clip": 1.05293846,
      "balance_loss_mlp": 1.02136636,
      "epoch": 0.33211086394516925,
      "flos": 31577006764800.0,
      "grad_norm": 2.234101902077234,
      "language_loss": 0.68985963,
      "learning_rate": 3.116720522990859e-06,
      "loss": 0.71173096,
      "num_input_tokens_seen": 59540415,
      "step": 2762,
      "time_per_iteration": 3.7962799072265625
    },
    {
      "auxiliary_loss_clip": 0.0118126,
      "auxiliary_loss_mlp": 0.01038733,
      "balance_loss_clip": 1.05630112,
      "balance_loss_mlp": 1.02944112,
      "epoch": 0.3322311068358083,
      "flos": 17932496791680.0,
      "grad_norm": 2.8802169701014697,
      "language_loss": 0.61964178,
      "learning_rate": 3.116074204525142e-06,
      "loss": 0.64184165,
      "num_input_tokens_seen": 59558590,
      "step": 2763,
      "time_per_iteration": 2.664306402206421
    },
    {
      "auxiliary_loss_clip": 0.01199073,
      "auxiliary_loss_mlp": 0.01032264,
      "balance_loss_clip": 1.06573951,
      "balance_loss_mlp": 1.02298331,
      "epoch": 0.3323513497264474,
      "flos": 32269831269120.0,
      "grad_norm": 1.8125680093039007,
      "language_loss": 0.83692598,
      "learning_rate": 3.1154277167518553e-06,
      "loss": 0.85923934,
      "num_input_tokens_seen": 59580205,
      "step": 2764,
      "time_per_iteration": 2.8280603885650635
    },
    {
      "auxiliary_loss_clip": 0.01041781,
      "auxiliary_loss_mlp": 0.0101082,
      "balance_loss_clip": 1.01724803,
      "balance_loss_mlp": 1.0088892,
      "epoch": 0.33247159261708653,
      "flos": 52668674588160.0,
      "grad_norm": 0.778438576338238,
      "language_loss": 0.59395981,
      "learning_rate": 3.114781059769072e-06,
      "loss": 0.61448586,
      "num_input_tokens_seen": 59631530,
      "step": 2765,
      "time_per_iteration": 4.109131813049316
    },
    {
      "auxiliary_loss_clip": 0.01148914,
      "auxiliary_loss_mlp": 0.01040228,
      "balance_loss_clip": 1.05225897,
      "balance_loss_mlp": 1.03080511,
      "epoch": 0.3325918355077256,
      "flos": 27125232906240.0,
      "grad_norm": 3.2951845078255775,
      "language_loss": 0.67547286,
      "learning_rate": 3.1141342336748874e-06,
      "loss": 0.69736421,
      "num_input_tokens_seen": 59651090,
      "step": 2766,
      "time_per_iteration": 3.7157270908355713
    },
    {
      "auxiliary_loss_clip": 0.0115924,
      "auxiliary_loss_mlp": 0.01039591,
      "balance_loss_clip": 1.05658841,
      "balance_loss_mlp": 1.03039455,
      "epoch": 0.3327120783983647,
      "flos": 23664414435840.0,
      "grad_norm": 1.603335244113829,
      "language_loss": 0.82126045,
      "learning_rate": 3.1134872385674253e-06,
      "loss": 0.84324878,
      "num_input_tokens_seen": 59675245,
      "step": 2767,
      "time_per_iteration": 2.79941987991333
    },
    {
      "auxiliary_loss_clip": 0.01210616,
      "auxiliary_loss_mlp": 0.01028903,
      "balance_loss_clip": 1.06511307,
      "balance_loss_mlp": 1.01954544,
      "epoch": 0.3328323212890038,
      "flos": 19171378828800.0,
      "grad_norm": 2.0050277948035635,
      "language_loss": 0.86064517,
      "learning_rate": 3.1128400745448353e-06,
      "loss": 0.88304037,
      "num_input_tokens_seen": 59694625,
      "step": 2768,
      "time_per_iteration": 2.6643712520599365
    },
    {
      "auxiliary_loss_clip": 0.01178718,
      "auxiliary_loss_mlp": 0.01040123,
      "balance_loss_clip": 1.06327236,
      "balance_loss_mlp": 1.03003216,
      "epoch": 0.33295256417964286,
      "flos": 37706347463040.0,
      "grad_norm": 2.184237058723349,
      "language_loss": 0.63280386,
      "learning_rate": 3.11219274170529e-06,
      "loss": 0.65499228,
      "num_input_tokens_seen": 59716435,
      "step": 2769,
      "time_per_iteration": 2.8091442584991455
    },
    {
      "auxiliary_loss_clip": 0.01153221,
      "auxiliary_loss_mlp": 0.01032275,
      "balance_loss_clip": 1.05343604,
      "balance_loss_mlp": 1.02338147,
      "epoch": 0.333072807070282,
      "flos": 26505989412480.0,
      "grad_norm": 1.8251157596550989,
      "language_loss": 0.8120029,
      "learning_rate": 3.1115452401469903e-06,
      "loss": 0.83385783,
      "num_input_tokens_seen": 59736835,
      "step": 2770,
      "time_per_iteration": 2.769167423248291
    },
    {
      "auxiliary_loss_clip": 0.01166634,
      "auxiliary_loss_mlp": 0.01045647,
      "balance_loss_clip": 1.05951643,
      "balance_loss_mlp": 1.03585434,
      "epoch": 0.3331930499609211,
      "flos": 21430913823360.0,
      "grad_norm": 1.8364711851870947,
      "language_loss": 0.86516821,
      "learning_rate": 3.1108975699681613e-06,
      "loss": 0.88729107,
      "num_input_tokens_seen": 59754230,
      "step": 2771,
      "time_per_iteration": 2.731130361557007
    },
    {
      "auxiliary_loss_clip": 0.01167593,
      "auxiliary_loss_mlp": 0.01034436,
      "balance_loss_clip": 1.06053257,
      "balance_loss_mlp": 1.02487588,
      "epoch": 0.33331329285156014,
      "flos": 20659947281280.0,
      "grad_norm": 2.205055527323958,
      "language_loss": 0.71990228,
      "learning_rate": 3.1102497312670542e-06,
      "loss": 0.74192262,
      "num_input_tokens_seen": 59772235,
      "step": 2772,
      "time_per_iteration": 2.7557451725006104
    },
    {
      "auxiliary_loss_clip": 0.01149202,
      "auxiliary_loss_mlp": 0.01041256,
      "balance_loss_clip": 1.06014633,
      "balance_loss_mlp": 1.03277993,
      "epoch": 0.33343353574219925,
      "flos": 28001596930560.0,
      "grad_norm": 3.9310680816976533,
      "language_loss": 0.80692756,
      "learning_rate": 3.109601724141946e-06,
      "loss": 0.82883215,
      "num_input_tokens_seen": 59791230,
      "step": 2773,
      "time_per_iteration": 2.8256306648254395
    },
    {
      "auxiliary_loss_clip": 0.01178203,
      "auxiliary_loss_mlp": 0.00763744,
      "balance_loss_clip": 1.05976987,
      "balance_loss_mlp": 1.00051594,
      "epoch": 0.33355377863283836,
      "flos": 23764963582080.0,
      "grad_norm": 2.08121665970231,
      "language_loss": 0.68886381,
      "learning_rate": 3.108953548691138e-06,
      "loss": 0.70828331,
      "num_input_tokens_seen": 59811315,
      "step": 2774,
      "time_per_iteration": 2.8209521770477295
    },
    {
      "auxiliary_loss_clip": 0.01195636,
      "auxiliary_loss_mlp": 0.01039532,
      "balance_loss_clip": 1.06465113,
      "balance_loss_mlp": 1.03037047,
      "epoch": 0.3336740215234774,
      "flos": 37779677078400.0,
      "grad_norm": 2.5914707782966153,
      "language_loss": 0.7248655,
      "learning_rate": 3.108305205012959e-06,
      "loss": 0.74721718,
      "num_input_tokens_seen": 59832010,
      "step": 2775,
      "time_per_iteration": 2.8085200786590576
    },
    {
      "auxiliary_loss_clip": 0.01192757,
      "auxiliary_loss_mlp": 0.01039898,
      "balance_loss_clip": 1.0629617,
      "balance_loss_mlp": 1.03114796,
      "epoch": 0.3337942644141165,
      "flos": 25519056347520.0,
      "grad_norm": 2.0428149229696118,
      "language_loss": 0.87763381,
      "learning_rate": 3.107656693205761e-06,
      "loss": 0.8999604,
      "num_input_tokens_seen": 59851450,
      "step": 2776,
      "time_per_iteration": 2.701573133468628
    },
    {
      "auxiliary_loss_clip": 0.01154104,
      "auxiliary_loss_mlp": 0.0104089,
      "balance_loss_clip": 1.05886245,
      "balance_loss_mlp": 1.03003597,
      "epoch": 0.3339145073047556,
      "flos": 25989844930560.0,
      "grad_norm": 3.7112849746656935,
      "language_loss": 0.70377207,
      "learning_rate": 3.107008013367924e-06,
      "loss": 0.72572196,
      "num_input_tokens_seen": 59870245,
      "step": 2777,
      "time_per_iteration": 2.8399128913879395
    },
    {
      "auxiliary_loss_clip": 0.01159897,
      "auxiliary_loss_mlp": 0.01042071,
      "balance_loss_clip": 1.05726898,
      "balance_loss_mlp": 1.03275478,
      "epoch": 0.3340347501953947,
      "flos": 19062569554560.0,
      "grad_norm": 3.102130773379478,
      "language_loss": 0.87109506,
      "learning_rate": 3.1063591655978507e-06,
      "loss": 0.89311481,
      "num_input_tokens_seen": 59886195,
      "step": 2778,
      "time_per_iteration": 2.734565496444702
    },
    {
      "auxiliary_loss_clip": 0.01200807,
      "auxiliary_loss_mlp": 0.01042865,
      "balance_loss_clip": 1.06346726,
      "balance_loss_mlp": 1.03334606,
      "epoch": 0.3341549930860338,
      "flos": 18109715518080.0,
      "grad_norm": 2.1909315650695156,
      "language_loss": 0.79920685,
      "learning_rate": 3.105710149993972e-06,
      "loss": 0.82164359,
      "num_input_tokens_seen": 59905525,
      "step": 2779,
      "time_per_iteration": 2.7424511909484863
    },
    {
      "auxiliary_loss_clip": 0.01144122,
      "auxiliary_loss_mlp": 0.01039005,
      "balance_loss_clip": 1.05839634,
      "balance_loss_mlp": 1.02972472,
      "epoch": 0.33427523597667286,
      "flos": 22674967418880.0,
      "grad_norm": 2.5972750109271763,
      "language_loss": 0.85609829,
      "learning_rate": 3.1050609666547427e-06,
      "loss": 0.87792957,
      "num_input_tokens_seen": 59925085,
      "step": 2780,
      "time_per_iteration": 2.792766571044922
    },
    {
      "auxiliary_loss_clip": 0.01161764,
      "auxiliary_loss_mlp": 0.01041709,
      "balance_loss_clip": 1.05567765,
      "balance_loss_mlp": 1.03158772,
      "epoch": 0.33439547886731197,
      "flos": 22638338524800.0,
      "grad_norm": 3.1445214161740256,
      "language_loss": 0.77616602,
      "learning_rate": 3.104411615678644e-06,
      "loss": 0.79820073,
      "num_input_tokens_seen": 59943935,
      "step": 2781,
      "time_per_iteration": 2.7297520637512207
    },
    {
      "auxiliary_loss_clip": 0.01143709,
      "auxiliary_loss_mlp": 0.01039512,
      "balance_loss_clip": 1.05338025,
      "balance_loss_mlp": 1.03049982,
      "epoch": 0.3345157217579511,
      "flos": 24096383395200.0,
      "grad_norm": 3.0088436055577463,
      "language_loss": 0.72991717,
      "learning_rate": 3.1037620971641803e-06,
      "loss": 0.7517494,
      "num_input_tokens_seen": 59963725,
      "step": 2782,
      "time_per_iteration": 2.7439746856689453
    },
    {
      "auxiliary_loss_clip": 0.01199447,
      "auxiliary_loss_mlp": 0.01036423,
      "balance_loss_clip": 1.06120348,
      "balance_loss_mlp": 1.02671313,
      "epoch": 0.33463596464859013,
      "flos": 18989491334400.0,
      "grad_norm": 3.1406905638724063,
      "language_loss": 0.65041935,
      "learning_rate": 3.1031124112098844e-06,
      "loss": 0.67277807,
      "num_input_tokens_seen": 59981935,
      "step": 2783,
      "time_per_iteration": 2.780062198638916
    },
    {
      "auxiliary_loss_clip": 0.0114724,
      "auxiliary_loss_mlp": 0.01047302,
      "balance_loss_clip": 1.05711985,
      "balance_loss_mlp": 1.03650761,
      "epoch": 0.33475620753922924,
      "flos": 20375607219840.0,
      "grad_norm": 1.9922421937201584,
      "language_loss": 0.72421324,
      "learning_rate": 3.1024625579143127e-06,
      "loss": 0.74615866,
      "num_input_tokens_seen": 59999455,
      "step": 2784,
      "time_per_iteration": 2.7303547859191895
    },
    {
      "auxiliary_loss_clip": 0.01209301,
      "auxiliary_loss_mlp": 0.01033719,
      "balance_loss_clip": 1.06209326,
      "balance_loss_mlp": 1.02464759,
      "epoch": 0.33487645042986836,
      "flos": 18182578256640.0,
      "grad_norm": 2.3689889551877377,
      "language_loss": 0.73048621,
      "learning_rate": 3.101812537376048e-06,
      "loss": 0.75291634,
      "num_input_tokens_seen": 60018475,
      "step": 2785,
      "time_per_iteration": 2.6643290519714355
    },
    {
      "auxiliary_loss_clip": 0.01185141,
      "auxiliary_loss_mlp": 0.01029895,
      "balance_loss_clip": 1.05971646,
      "balance_loss_mlp": 1.02132964,
      "epoch": 0.3349966933205074,
      "flos": 25848824135040.0,
      "grad_norm": 2.14510450954911,
      "language_loss": 0.84654891,
      "learning_rate": 3.1011623496936973e-06,
      "loss": 0.86869931,
      "num_input_tokens_seen": 60036770,
      "step": 2786,
      "time_per_iteration": 2.7546634674072266
    },
    {
      "auxiliary_loss_clip": 0.01210766,
      "auxiliary_loss_mlp": 0.01045155,
      "balance_loss_clip": 1.06512904,
      "balance_loss_mlp": 1.03568363,
      "epoch": 0.3351169362111465,
      "flos": 28111447699200.0,
      "grad_norm": 2.46021403615838,
      "language_loss": 0.69854623,
      "learning_rate": 3.100511994965893e-06,
      "loss": 0.72110546,
      "num_input_tokens_seen": 60056725,
      "step": 2787,
      "time_per_iteration": 3.630983591079712
    },
    {
      "auxiliary_loss_clip": 0.01182021,
      "auxiliary_loss_mlp": 0.01035591,
      "balance_loss_clip": 1.06009555,
      "balance_loss_mlp": 1.0262394,
      "epoch": 0.33523717910178563,
      "flos": 22673315393280.0,
      "grad_norm": 1.8052306162861829,
      "language_loss": 0.84710598,
      "learning_rate": 3.0998614732912947e-06,
      "loss": 0.86928207,
      "num_input_tokens_seen": 60076100,
      "step": 2788,
      "time_per_iteration": 3.6242213249206543
    },
    {
      "auxiliary_loss_clip": 0.01198486,
      "auxiliary_loss_mlp": 0.01046248,
      "balance_loss_clip": 1.06222916,
      "balance_loss_mlp": 1.03658044,
      "epoch": 0.3353574219924247,
      "flos": 15669801400320.0,
      "grad_norm": 6.445738148253758,
      "language_loss": 0.67757529,
      "learning_rate": 3.0992107847685855e-06,
      "loss": 0.70002258,
      "num_input_tokens_seen": 60093815,
      "step": 2789,
      "time_per_iteration": 2.6546545028686523
    },
    {
      "auxiliary_loss_clip": 0.01195929,
      "auxiliary_loss_mlp": 0.01036092,
      "balance_loss_clip": 1.06509972,
      "balance_loss_mlp": 1.02610171,
      "epoch": 0.3354776648830638,
      "flos": 24790644443520.0,
      "grad_norm": 2.028515186734787,
      "language_loss": 0.79523528,
      "learning_rate": 3.0985599294964736e-06,
      "loss": 0.81755543,
      "num_input_tokens_seen": 60113370,
      "step": 2790,
      "time_per_iteration": 2.744286060333252
    },
    {
      "auxiliary_loss_clip": 0.01168038,
      "auxiliary_loss_mlp": 0.01039599,
      "balance_loss_clip": 1.05683863,
      "balance_loss_mlp": 1.02969885,
      "epoch": 0.33559790777370285,
      "flos": 28694852398080.0,
      "grad_norm": 2.2618493069287733,
      "language_loss": 0.70416176,
      "learning_rate": 3.097908907573695e-06,
      "loss": 0.72623807,
      "num_input_tokens_seen": 60131350,
      "step": 2791,
      "time_per_iteration": 3.799933671951294
    },
    {
      "auxiliary_loss_clip": 0.01179741,
      "auxiliary_loss_mlp": 0.01043735,
      "balance_loss_clip": 1.05812907,
      "balance_loss_mlp": 1.03422785,
      "epoch": 0.33571815066434196,
      "flos": 22235779825920.0,
      "grad_norm": 2.0693208855370906,
      "language_loss": 0.89935964,
      "learning_rate": 3.0972577190990067e-06,
      "loss": 0.92159438,
      "num_input_tokens_seen": 60149830,
      "step": 2792,
      "time_per_iteration": 3.6383750438690186
    },
    {
      "auxiliary_loss_clip": 0.01208539,
      "auxiliary_loss_mlp": 0.01042103,
      "balance_loss_clip": 1.06346488,
      "balance_loss_mlp": 1.03239942,
      "epoch": 0.3358383935549811,
      "flos": 23842279607040.0,
      "grad_norm": 2.6610804647065844,
      "language_loss": 0.80306447,
      "learning_rate": 3.096606364171196e-06,
      "loss": 0.82557088,
      "num_input_tokens_seen": 60169620,
      "step": 2793,
      "time_per_iteration": 2.683562755584717
    },
    {
      "auxiliary_loss_clip": 0.01210148,
      "auxiliary_loss_mlp": 0.0104099,
      "balance_loss_clip": 1.06486511,
      "balance_loss_mlp": 1.03207946,
      "epoch": 0.33595863644562013,
      "flos": 22267308988800.0,
      "grad_norm": 2.3425664248337066,
      "language_loss": 0.85049069,
      "learning_rate": 3.0959548428890703e-06,
      "loss": 0.87300205,
      "num_input_tokens_seen": 60188490,
      "step": 2794,
      "time_per_iteration": 2.654627561569214
    },
    {
      "auxiliary_loss_clip": 0.01209427,
      "auxiliary_loss_mlp": 0.01030777,
      "balance_loss_clip": 1.06390965,
      "balance_loss_mlp": 1.0215497,
      "epoch": 0.33607887933625924,
      "flos": 20119779578880.0,
      "grad_norm": 2.9899830364427404,
      "language_loss": 0.84473634,
      "learning_rate": 3.095303155351468e-06,
      "loss": 0.86713839,
      "num_input_tokens_seen": 60208695,
      "step": 2795,
      "time_per_iteration": 2.608426094055176
    },
    {
      "auxiliary_loss_clip": 0.01206121,
      "auxiliary_loss_mlp": 0.01044499,
      "balance_loss_clip": 1.06200504,
      "balance_loss_mlp": 1.03534937,
      "epoch": 0.33619912222689835,
      "flos": 19318109886720.0,
      "grad_norm": 2.3959306391666098,
      "language_loss": 0.79530978,
      "learning_rate": 3.0946513016572464e-06,
      "loss": 0.81781596,
      "num_input_tokens_seen": 60227600,
      "step": 2796,
      "time_per_iteration": 2.6199114322662354
    },
    {
      "auxiliary_loss_clip": 0.0121022,
      "auxiliary_loss_mlp": 0.01039344,
      "balance_loss_clip": 1.06143415,
      "balance_loss_mlp": 1.0304625,
      "epoch": 0.3363193651175374,
      "flos": 16800664262400.0,
      "grad_norm": 2.4067152711062834,
      "language_loss": 0.77126753,
      "learning_rate": 3.0939992819052938e-06,
      "loss": 0.7937631,
      "num_input_tokens_seen": 60245110,
      "step": 2797,
      "time_per_iteration": 2.562743902206421
    },
    {
      "auxiliary_loss_clip": 0.01187338,
      "auxiliary_loss_mlp": 0.01031239,
      "balance_loss_clip": 1.05744839,
      "balance_loss_mlp": 1.02213144,
      "epoch": 0.3364396080081765,
      "flos": 23550289948800.0,
      "grad_norm": 2.1853733090409393,
      "language_loss": 0.81754255,
      "learning_rate": 3.0933470961945193e-06,
      "loss": 0.8397283,
      "num_input_tokens_seen": 60263405,
      "step": 2798,
      "time_per_iteration": 2.706291437149048
    },
    {
      "auxiliary_loss_clip": 0.0117677,
      "auxiliary_loss_mlp": 0.01036139,
      "balance_loss_clip": 1.05982232,
      "balance_loss_mlp": 1.02714467,
      "epoch": 0.3365598508988156,
      "flos": 28037902602240.0,
      "grad_norm": 2.895820593895203,
      "language_loss": 0.68453622,
      "learning_rate": 3.0926947446238597e-06,
      "loss": 0.70666528,
      "num_input_tokens_seen": 60282975,
      "step": 2799,
      "time_per_iteration": 2.7243666648864746
    },
    {
      "auxiliary_loss_clip": 0.01208267,
      "auxiliary_loss_mlp": 0.01033776,
      "balance_loss_clip": 1.06000817,
      "balance_loss_mlp": 1.02299333,
      "epoch": 0.3366800937894547,
      "flos": 16982767238400.0,
      "grad_norm": 2.6724344436603453,
      "language_loss": 0.82581484,
      "learning_rate": 3.092042227292276e-06,
      "loss": 0.84823525,
      "num_input_tokens_seen": 60299810,
      "step": 2800,
      "time_per_iteration": 2.675675392150879
    },
    {
      "auxiliary_loss_clip": 0.01178252,
      "auxiliary_loss_mlp": 0.01036995,
      "balance_loss_clip": 1.05896187,
      "balance_loss_mlp": 1.02821469,
      "epoch": 0.3368003366800938,
      "flos": 23915321913600.0,
      "grad_norm": 3.3894234836612944,
      "language_loss": 0.88563335,
      "learning_rate": 3.0913895442987557e-06,
      "loss": 0.90778583,
      "num_input_tokens_seen": 60320775,
      "step": 2801,
      "time_per_iteration": 2.7399847507476807
    },
    {
      "auxiliary_loss_clip": 0.01207935,
      "auxiliary_loss_mlp": 0.01036957,
      "balance_loss_clip": 1.06287897,
      "balance_loss_mlp": 1.02817762,
      "epoch": 0.3369205795707329,
      "flos": 24791219061120.0,
      "grad_norm": 2.862157510874502,
      "language_loss": 0.85882545,
      "learning_rate": 3.090736695742308e-06,
      "loss": 0.88127434,
      "num_input_tokens_seen": 60341905,
      "step": 2802,
      "time_per_iteration": 2.701460599899292
    },
    {
      "auxiliary_loss_clip": 0.01186694,
      "auxiliary_loss_mlp": 0.01034393,
      "balance_loss_clip": 1.05916798,
      "balance_loss_mlp": 1.02517176,
      "epoch": 0.33704082246137196,
      "flos": 17931096161280.0,
      "grad_norm": 2.416003840445918,
      "language_loss": 0.52236289,
      "learning_rate": 3.0900836817219713e-06,
      "loss": 0.54457372,
      "num_input_tokens_seen": 60358335,
      "step": 2803,
      "time_per_iteration": 2.610827684402466
    },
    {
      "auxiliary_loss_clip": 0.01205945,
      "auxiliary_loss_mlp": 0.01038156,
      "balance_loss_clip": 1.06136513,
      "balance_loss_mlp": 1.02847028,
      "epoch": 0.33716106535201107,
      "flos": 21286517149440.0,
      "grad_norm": 3.381117348921781,
      "language_loss": 0.83617961,
      "learning_rate": 3.089430502336807e-06,
      "loss": 0.85862064,
      "num_input_tokens_seen": 60378305,
      "step": 2804,
      "time_per_iteration": 2.6268720626831055
    },
    {
      "auxiliary_loss_clip": 0.01189572,
      "auxiliary_loss_mlp": 0.01033586,
      "balance_loss_clip": 1.06003654,
      "balance_loss_mlp": 1.02395952,
      "epoch": 0.3372813082426502,
      "flos": 18402962152320.0,
      "grad_norm": 4.033381198006485,
      "language_loss": 0.90785879,
      "learning_rate": 3.088777157685902e-06,
      "loss": 0.93009031,
      "num_input_tokens_seen": 60393895,
      "step": 2805,
      "time_per_iteration": 2.6459808349609375
    },
    {
      "auxiliary_loss_clip": 0.01180875,
      "auxiliary_loss_mlp": 0.01035162,
      "balance_loss_clip": 1.05668306,
      "balance_loss_mlp": 1.02612042,
      "epoch": 0.33740155113328923,
      "flos": 17201391367680.0,
      "grad_norm": 2.478144448908263,
      "language_loss": 0.85477775,
      "learning_rate": 3.088123647868367e-06,
      "loss": 0.8769381,
      "num_input_tokens_seen": 60410445,
      "step": 2806,
      "time_per_iteration": 2.662510395050049
    },
    {
      "auxiliary_loss_clip": 0.01206996,
      "auxiliary_loss_mlp": 0.01038105,
      "balance_loss_clip": 1.06140304,
      "balance_loss_mlp": 1.0295701,
      "epoch": 0.33752179402392835,
      "flos": 29058950609280.0,
      "grad_norm": 2.2386932365515904,
      "language_loss": 0.81729174,
      "learning_rate": 3.0874699729833405e-06,
      "loss": 0.83974272,
      "num_input_tokens_seen": 60431815,
      "step": 2807,
      "time_per_iteration": 2.6995649337768555
    },
    {
      "auxiliary_loss_clip": 0.01193124,
      "auxiliary_loss_mlp": 0.01033951,
      "balance_loss_clip": 1.05872011,
      "balance_loss_mlp": 1.02515924,
      "epoch": 0.3376420369145674,
      "flos": 25080730680960.0,
      "grad_norm": 2.081223944830791,
      "language_loss": 0.7976222,
      "learning_rate": 3.086816133129983e-06,
      "loss": 0.81989288,
      "num_input_tokens_seen": 60452075,
      "step": 2808,
      "time_per_iteration": 2.699061155319214
    },
    {
      "auxiliary_loss_clip": 0.01168558,
      "auxiliary_loss_mlp": 0.01039089,
      "balance_loss_clip": 1.05671167,
      "balance_loss_mlp": 1.02992833,
      "epoch": 0.3377622798052065,
      "flos": 27490624007040.0,
      "grad_norm": 1.9601335788535903,
      "language_loss": 0.7629292,
      "learning_rate": 3.0861621284074826e-06,
      "loss": 0.78500563,
      "num_input_tokens_seen": 60472600,
      "step": 2809,
      "time_per_iteration": 2.7275218963623047
    },
    {
      "auxiliary_loss_clip": 0.011895,
      "auxiliary_loss_mlp": 0.01036054,
      "balance_loss_clip": 1.05881524,
      "balance_loss_mlp": 1.02746522,
      "epoch": 0.3378825226958456,
      "flos": 21975211589760.0,
      "grad_norm": 1.8203481650649764,
      "language_loss": 0.73341107,
      "learning_rate": 3.085507958915051e-06,
      "loss": 0.75566667,
      "num_input_tokens_seen": 60491030,
      "step": 2810,
      "time_per_iteration": 2.6161375045776367
    },
    {
      "auxiliary_loss_clip": 0.01189591,
      "auxiliary_loss_mlp": 0.01034785,
      "balance_loss_clip": 1.06105447,
      "balance_loss_mlp": 1.02589226,
      "epoch": 0.3380027655864847,
      "flos": 42523189200000.0,
      "grad_norm": 3.2563861808160857,
      "language_loss": 0.71733105,
      "learning_rate": 3.084853624751925e-06,
      "loss": 0.73957479,
      "num_input_tokens_seen": 60512615,
      "step": 2811,
      "time_per_iteration": 2.87873911857605
    },
    {
      "auxiliary_loss_clip": 0.01190475,
      "auxiliary_loss_mlp": 0.0103002,
      "balance_loss_clip": 1.06148732,
      "balance_loss_mlp": 1.02182412,
      "epoch": 0.3381230084771238,
      "flos": 26725080418560.0,
      "grad_norm": 1.8782275748430508,
      "language_loss": 0.86015177,
      "learning_rate": 3.0841991260173668e-06,
      "loss": 0.8823567,
      "num_input_tokens_seen": 60532520,
      "step": 2812,
      "time_per_iteration": 2.7039365768432617
    },
    {
      "auxiliary_loss_clip": 0.01208084,
      "auxiliary_loss_mlp": 0.01034287,
      "balance_loss_clip": 1.06316948,
      "balance_loss_mlp": 1.02569795,
      "epoch": 0.3382432513677629,
      "flos": 22710375250560.0,
      "grad_norm": 3.173237547300725,
      "language_loss": 0.80537385,
      "learning_rate": 3.0835444628106634e-06,
      "loss": 0.82779753,
      "num_input_tokens_seen": 60551500,
      "step": 2813,
      "time_per_iteration": 3.676241397857666
    },
    {
      "auxiliary_loss_clip": 0.01187265,
      "auxiliary_loss_mlp": 0.01033561,
      "balance_loss_clip": 1.06019211,
      "balance_loss_mlp": 1.02464437,
      "epoch": 0.33836349425840195,
      "flos": 22122409524480.0,
      "grad_norm": 3.663522310202432,
      "language_loss": 0.82908279,
      "learning_rate": 3.082889635231126e-06,
      "loss": 0.85129106,
      "num_input_tokens_seen": 60570160,
      "step": 2814,
      "time_per_iteration": 2.7505595684051514
    },
    {
      "auxiliary_loss_clip": 0.01164582,
      "auxiliary_loss_mlp": 0.01031494,
      "balance_loss_clip": 1.05609214,
      "balance_loss_mlp": 1.02289343,
      "epoch": 0.33848373714904106,
      "flos": 27308090067840.0,
      "grad_norm": 3.473524656196039,
      "language_loss": 0.76808006,
      "learning_rate": 3.0822346433780925e-06,
      "loss": 0.79004085,
      "num_input_tokens_seen": 60590885,
      "step": 2815,
      "time_per_iteration": 2.7505276203155518
    },
    {
      "auxiliary_loss_clip": 0.01195539,
      "auxiliary_loss_mlp": 0.0103379,
      "balance_loss_clip": 1.05892301,
      "balance_loss_mlp": 1.02493894,
      "epoch": 0.3386039800396802,
      "flos": 25848716394240.0,
      "grad_norm": 2.131646762093233,
      "language_loss": 0.87018359,
      "learning_rate": 3.0815794873509237e-06,
      "loss": 0.89247692,
      "num_input_tokens_seen": 60609170,
      "step": 2816,
      "time_per_iteration": 2.792710781097412
    },
    {
      "auxiliary_loss_clip": 0.01204239,
      "auxiliary_loss_mlp": 0.01031055,
      "balance_loss_clip": 1.06157291,
      "balance_loss_mlp": 1.02259684,
      "epoch": 0.33872422293031923,
      "flos": 18880646146560.0,
      "grad_norm": 2.265361791031468,
      "language_loss": 0.72763699,
      "learning_rate": 3.0809241672490066e-06,
      "loss": 0.74998999,
      "num_input_tokens_seen": 60627340,
      "step": 2817,
      "time_per_iteration": 3.5623109340667725
    },
    {
      "auxiliary_loss_clip": 0.01160861,
      "auxiliary_loss_mlp": 0.01036311,
      "balance_loss_clip": 1.05465102,
      "balance_loss_mlp": 1.02832985,
      "epoch": 0.33884446582095834,
      "flos": 23146977064320.0,
      "grad_norm": 3.1420662094323566,
      "language_loss": 0.85277939,
      "learning_rate": 3.080268683171753e-06,
      "loss": 0.87475109,
      "num_input_tokens_seen": 60647630,
      "step": 2818,
      "time_per_iteration": 3.6404237747192383
    },
    {
      "auxiliary_loss_clip": 0.01203565,
      "auxiliary_loss_mlp": 0.01030663,
      "balance_loss_clip": 1.0623734,
      "balance_loss_mlp": 1.02249706,
      "epoch": 0.33896470871159745,
      "flos": 15997342544640.0,
      "grad_norm": 2.7990813739854508,
      "language_loss": 0.8917613,
      "learning_rate": 3.0796130352185985e-06,
      "loss": 0.91410363,
      "num_input_tokens_seen": 60664485,
      "step": 2819,
      "time_per_iteration": 2.633927583694458
    },
    {
      "auxiliary_loss_clip": 0.01204291,
      "auxiliary_loss_mlp": 0.01042991,
      "balance_loss_clip": 1.06085336,
      "balance_loss_mlp": 1.03397238,
      "epoch": 0.3390849516022365,
      "flos": 34495754112000.0,
      "grad_norm": 2.5031075797801083,
      "language_loss": 0.66495693,
      "learning_rate": 3.0789572234890057e-06,
      "loss": 0.68742979,
      "num_input_tokens_seen": 60686125,
      "step": 2820,
      "time_per_iteration": 2.710991382598877
    },
    {
      "auxiliary_loss_clip": 0.01189997,
      "auxiliary_loss_mlp": 0.01028346,
      "balance_loss_clip": 1.05880415,
      "balance_loss_mlp": 1.01964962,
      "epoch": 0.3392051944928756,
      "flos": 16180307447040.0,
      "grad_norm": 4.258048827123285,
      "language_loss": 0.77509868,
      "learning_rate": 3.0783012480824596e-06,
      "loss": 0.7972821,
      "num_input_tokens_seen": 60705270,
      "step": 2821,
      "time_per_iteration": 2.692021369934082
    },
    {
      "auxiliary_loss_clip": 0.01202327,
      "auxiliary_loss_mlp": 0.01046783,
      "balance_loss_clip": 1.0597682,
      "balance_loss_mlp": 1.03753829,
      "epoch": 0.33932543738351467,
      "flos": 17086656349440.0,
      "grad_norm": 3.8140760271565077,
      "language_loss": 0.74770117,
      "learning_rate": 3.077645109098471e-06,
      "loss": 0.77019233,
      "num_input_tokens_seen": 60721540,
      "step": 2822,
      "time_per_iteration": 2.6014606952667236
    },
    {
      "auxiliary_loss_clip": 0.01199543,
      "auxiliary_loss_mlp": 0.0102747,
      "balance_loss_clip": 1.0581888,
      "balance_loss_mlp": 1.01902449,
      "epoch": 0.3394456802741538,
      "flos": 22126970551680.0,
      "grad_norm": 5.387053153350691,
      "language_loss": 0.72499013,
      "learning_rate": 3.076988806636577e-06,
      "loss": 0.74726027,
      "num_input_tokens_seen": 60739300,
      "step": 2823,
      "time_per_iteration": 2.7016615867614746
    },
    {
      "auxiliary_loss_clip": 0.01191561,
      "auxiliary_loss_mlp": 0.01035547,
      "balance_loss_clip": 1.06170738,
      "balance_loss_mlp": 1.02676737,
      "epoch": 0.3395659231647929,
      "flos": 25226887121280.0,
      "grad_norm": 2.033187379942383,
      "language_loss": 0.88746685,
      "learning_rate": 3.0763323407963377e-06,
      "loss": 0.90973788,
      "num_input_tokens_seen": 60758910,
      "step": 2824,
      "time_per_iteration": 2.647498846054077
    },
    {
      "auxiliary_loss_clip": 0.01204032,
      "auxiliary_loss_mlp": 0.01033655,
      "balance_loss_clip": 1.06119752,
      "balance_loss_mlp": 1.02565551,
      "epoch": 0.33968616605543195,
      "flos": 29096477343360.0,
      "grad_norm": 4.69643585224092,
      "language_loss": 0.80339825,
      "learning_rate": 3.075675711677337e-06,
      "loss": 0.82577503,
      "num_input_tokens_seen": 60779005,
      "step": 2825,
      "time_per_iteration": 2.6757633686065674
    },
    {
      "auxiliary_loss_clip": 0.01187579,
      "auxiliary_loss_mlp": 0.01039067,
      "balance_loss_clip": 1.06116199,
      "balance_loss_mlp": 1.03060961,
      "epoch": 0.33980640894607106,
      "flos": 21433966479360.0,
      "grad_norm": 2.789427813475685,
      "language_loss": 0.78338599,
      "learning_rate": 3.0750189193791865e-06,
      "loss": 0.80565244,
      "num_input_tokens_seen": 60798590,
      "step": 2826,
      "time_per_iteration": 2.622628688812256
    },
    {
      "auxiliary_loss_clip": 0.01189546,
      "auxiliary_loss_mlp": 0.01036267,
      "balance_loss_clip": 1.06062269,
      "balance_loss_mlp": 1.0280416,
      "epoch": 0.33992665183671017,
      "flos": 32490035596800.0,
      "grad_norm": 2.718646208699851,
      "language_loss": 0.70760053,
      "learning_rate": 3.0743619640015203e-06,
      "loss": 0.7298587,
      "num_input_tokens_seen": 60818840,
      "step": 2827,
      "time_per_iteration": 2.721447229385376
    },
    {
      "auxiliary_loss_clip": 0.01150709,
      "auxiliary_loss_mlp": 0.01033459,
      "balance_loss_clip": 1.05514133,
      "balance_loss_mlp": 1.02481627,
      "epoch": 0.3400468947273492,
      "flos": 17055414495360.0,
      "grad_norm": 2.5363935290484783,
      "language_loss": 0.92410201,
      "learning_rate": 3.073704845643999e-06,
      "loss": 0.94594371,
      "num_input_tokens_seen": 60835965,
      "step": 2828,
      "time_per_iteration": 2.72343373298645
    },
    {
      "auxiliary_loss_clip": 0.01205102,
      "auxiliary_loss_mlp": 0.01033191,
      "balance_loss_clip": 1.06100881,
      "balance_loss_mlp": 1.02451229,
      "epoch": 0.34016713761798834,
      "flos": 16872988296960.0,
      "grad_norm": 3.853274760837107,
      "language_loss": 0.78063846,
      "learning_rate": 3.0730475644063063e-06,
      "loss": 0.80302137,
      "num_input_tokens_seen": 60851065,
      "step": 2829,
      "time_per_iteration": 2.5764429569244385
    },
    {
      "auxiliary_loss_clip": 0.01201019,
      "auxiliary_loss_mlp": 0.01023654,
      "balance_loss_clip": 1.05971217,
      "balance_loss_mlp": 1.01501083,
      "epoch": 0.34028738050862745,
      "flos": 21907161273600.0,
      "grad_norm": 2.288969909294798,
      "language_loss": 0.65139967,
      "learning_rate": 3.072390120388151e-06,
      "loss": 0.67364645,
      "num_input_tokens_seen": 60869390,
      "step": 2830,
      "time_per_iteration": 2.6251494884490967
    },
    {
      "auxiliary_loss_clip": 0.0117212,
      "auxiliary_loss_mlp": 0.0103605,
      "balance_loss_clip": 1.05714905,
      "balance_loss_mlp": 1.02701437,
      "epoch": 0.3404076233992665,
      "flos": 22746034477440.0,
      "grad_norm": 2.4792272998273206,
      "language_loss": 0.7175253,
      "learning_rate": 3.071732513689267e-06,
      "loss": 0.73960704,
      "num_input_tokens_seen": 60887925,
      "step": 2831,
      "time_per_iteration": 2.6868107318878174
    },
    {
      "auxiliary_loss_clip": 0.01207343,
      "auxiliary_loss_mlp": 0.01033255,
      "balance_loss_clip": 1.06195188,
      "balance_loss_mlp": 1.02401674,
      "epoch": 0.3405278662899056,
      "flos": 17052361839360.0,
      "grad_norm": 3.6309247330297394,
      "language_loss": 0.67866814,
      "learning_rate": 3.0710747444094134e-06,
      "loss": 0.70107412,
      "num_input_tokens_seen": 60905955,
      "step": 2832,
      "time_per_iteration": 2.618485450744629
    },
    {
      "auxiliary_loss_clip": 0.0117741,
      "auxiliary_loss_mlp": 0.01035202,
      "balance_loss_clip": 1.06098449,
      "balance_loss_mlp": 1.02576709,
      "epoch": 0.3406481091805447,
      "flos": 42813131783040.0,
      "grad_norm": 2.842805137476745,
      "language_loss": 0.65212101,
      "learning_rate": 3.070416812648372e-06,
      "loss": 0.67424715,
      "num_input_tokens_seen": 60929405,
      "step": 2833,
      "time_per_iteration": 2.862582206726074
    },
    {
      "auxiliary_loss_clip": 0.01204782,
      "auxiliary_loss_mlp": 0.0102966,
      "balance_loss_clip": 1.06282449,
      "balance_loss_mlp": 1.0211308,
      "epoch": 0.3407683520711838,
      "flos": 26761457917440.0,
      "grad_norm": 2.317750604052072,
      "language_loss": 0.65429014,
      "learning_rate": 3.069758718505951e-06,
      "loss": 0.67663455,
      "num_input_tokens_seen": 60951145,
      "step": 2834,
      "time_per_iteration": 2.7050223350524902
    },
    {
      "auxiliary_loss_clip": 0.01173478,
      "auxiliary_loss_mlp": 0.01030764,
      "balance_loss_clip": 1.05967641,
      "balance_loss_mlp": 1.02188349,
      "epoch": 0.3408885949618229,
      "flos": 28767643309440.0,
      "grad_norm": 1.9444780387719485,
      "language_loss": 0.80021936,
      "learning_rate": 3.0691004620819836e-06,
      "loss": 0.82226181,
      "num_input_tokens_seen": 60971275,
      "step": 2835,
      "time_per_iteration": 2.7192397117614746
    },
    {
      "auxiliary_loss_clip": 0.01064911,
      "auxiliary_loss_mlp": 0.01004371,
      "balance_loss_clip": 1.02060461,
      "balance_loss_mlp": 1.00265443,
      "epoch": 0.341008837852462,
      "flos": 63576252881280.0,
      "grad_norm": 0.796915057564704,
      "language_loss": 0.60139084,
      "learning_rate": 3.0684420434763254e-06,
      "loss": 0.62208366,
      "num_input_tokens_seen": 61037460,
      "step": 2836,
      "time_per_iteration": 3.3474535942077637
    },
    {
      "auxiliary_loss_clip": 0.01205724,
      "auxiliary_loss_mlp": 0.01037144,
      "balance_loss_clip": 1.06503856,
      "balance_loss_mlp": 1.02896678,
      "epoch": 0.34112908074310105,
      "flos": 20812173120000.0,
      "grad_norm": 2.2939255980511954,
      "language_loss": 0.76880562,
      "learning_rate": 3.06778346278886e-06,
      "loss": 0.79123437,
      "num_input_tokens_seen": 61056295,
      "step": 2837,
      "time_per_iteration": 2.6357336044311523
    },
    {
      "auxiliary_loss_clip": 0.01205306,
      "auxiliary_loss_mlp": 0.01032612,
      "balance_loss_clip": 1.06325364,
      "balance_loss_mlp": 1.02315831,
      "epoch": 0.34124932363374016,
      "flos": 24976446520320.0,
      "grad_norm": 1.8199262691342268,
      "language_loss": 0.79246223,
      "learning_rate": 3.0671247201194906e-06,
      "loss": 0.81484139,
      "num_input_tokens_seen": 61078430,
      "step": 2838,
      "time_per_iteration": 2.6858201026916504
    },
    {
      "auxiliary_loss_clip": 0.01154102,
      "auxiliary_loss_mlp": 0.01031176,
      "balance_loss_clip": 1.05423379,
      "balance_loss_mlp": 1.02230048,
      "epoch": 0.3413695665243792,
      "flos": 28402970480640.0,
      "grad_norm": 2.8586728204320737,
      "language_loss": 0.75597322,
      "learning_rate": 3.066465815568151e-06,
      "loss": 0.77782595,
      "num_input_tokens_seen": 61099260,
      "step": 2839,
      "time_per_iteration": 3.7603201866149902
    },
    {
      "auxiliary_loss_clip": 0.01188095,
      "auxiliary_loss_mlp": 0.01026972,
      "balance_loss_clip": 1.05868697,
      "balance_loss_mlp": 1.0182575,
      "epoch": 0.34148980941501833,
      "flos": 25302012416640.0,
      "grad_norm": 2.2343822762572367,
      "language_loss": 0.69014847,
      "learning_rate": 3.0658067492347947e-06,
      "loss": 0.71229917,
      "num_input_tokens_seen": 61121900,
      "step": 2840,
      "time_per_iteration": 3.720618486404419
    },
    {
      "auxiliary_loss_clip": 0.01186329,
      "auxiliary_loss_mlp": 0.0103194,
      "balance_loss_clip": 1.05757535,
      "balance_loss_mlp": 1.02346385,
      "epoch": 0.34161005230565744,
      "flos": 17530081747200.0,
      "grad_norm": 2.2710065073630146,
      "language_loss": 0.6699729,
      "learning_rate": 3.065147521219402e-06,
      "loss": 0.6921556,
      "num_input_tokens_seen": 61141155,
      "step": 2841,
      "time_per_iteration": 2.6447558403015137
    },
    {
      "auxiliary_loss_clip": 0.0117325,
      "auxiliary_loss_mlp": 0.01034408,
      "balance_loss_clip": 1.0595082,
      "balance_loss_mlp": 1.02540147,
      "epoch": 0.3417302951962965,
      "flos": 43650101566080.0,
      "grad_norm": 65.53906746607832,
      "language_loss": 0.74611706,
      "learning_rate": 3.064488131621977e-06,
      "loss": 0.7681936,
      "num_input_tokens_seen": 61164480,
      "step": 2842,
      "time_per_iteration": 2.9162309169769287
    },
    {
      "auxiliary_loss_clip": 0.01150248,
      "auxiliary_loss_mlp": 0.01029698,
      "balance_loss_clip": 1.05212307,
      "balance_loss_mlp": 1.02056098,
      "epoch": 0.3418505380869356,
      "flos": 30882207012480.0,
      "grad_norm": 1.7872260375203761,
      "language_loss": 0.74198532,
      "learning_rate": 3.063828580542549e-06,
      "loss": 0.76378483,
      "num_input_tokens_seen": 61185675,
      "step": 2843,
      "time_per_iteration": 3.7482950687408447
    },
    {
      "auxiliary_loss_clip": 0.0118678,
      "auxiliary_loss_mlp": 0.01026548,
      "balance_loss_clip": 1.05994678,
      "balance_loss_mlp": 1.01863801,
      "epoch": 0.3419707809775747,
      "flos": 19463871277440.0,
      "grad_norm": 3.445027011290699,
      "language_loss": 0.7374782,
      "learning_rate": 3.0631688680811706e-06,
      "loss": 0.75961149,
      "num_input_tokens_seen": 61205300,
      "step": 2844,
      "time_per_iteration": 3.708162307739258
    },
    {
      "auxiliary_loss_clip": 0.01171057,
      "auxiliary_loss_mlp": 0.01034451,
      "balance_loss_clip": 1.05686235,
      "balance_loss_mlp": 1.02595782,
      "epoch": 0.3420910238682138,
      "flos": 28727818104960.0,
      "grad_norm": 2.9177046587220405,
      "language_loss": 0.75842512,
      "learning_rate": 3.062508994337921e-06,
      "loss": 0.78048021,
      "num_input_tokens_seen": 61224905,
      "step": 2845,
      "time_per_iteration": 2.844129800796509
    },
    {
      "auxiliary_loss_clip": 0.01190672,
      "auxiliary_loss_mlp": 0.01026129,
      "balance_loss_clip": 1.05606043,
      "balance_loss_mlp": 1.01708722,
      "epoch": 0.3422112667588529,
      "flos": 21397265758080.0,
      "grad_norm": 2.236457996540503,
      "language_loss": 0.79517567,
      "learning_rate": 3.0618489594129013e-06,
      "loss": 0.81734365,
      "num_input_tokens_seen": 61243045,
      "step": 2846,
      "time_per_iteration": 2.6237993240356445
    },
    {
      "auxiliary_loss_clip": 0.01188865,
      "auxiliary_loss_mlp": 0.0103033,
      "balance_loss_clip": 1.05835509,
      "balance_loss_mlp": 1.02165794,
      "epoch": 0.342331509649492,
      "flos": 13881450038400.0,
      "grad_norm": 3.438495460756812,
      "language_loss": 0.71054387,
      "learning_rate": 3.061188763406239e-06,
      "loss": 0.73273581,
      "num_input_tokens_seen": 61259190,
      "step": 2847,
      "time_per_iteration": 2.682555913925171
    },
    {
      "auxiliary_loss_clip": 0.0120489,
      "auxiliary_loss_mlp": 0.01033294,
      "balance_loss_clip": 1.0619396,
      "balance_loss_mlp": 1.02449632,
      "epoch": 0.34245175254013105,
      "flos": 28621450955520.0,
      "grad_norm": 3.167285695683865,
      "language_loss": 0.81736422,
      "learning_rate": 3.060528406418085e-06,
      "loss": 0.839746,
      "num_input_tokens_seen": 61279040,
      "step": 2848,
      "time_per_iteration": 2.684469699859619
    },
    {
      "auxiliary_loss_clip": 0.01164499,
      "auxiliary_loss_mlp": 0.01032235,
      "balance_loss_clip": 1.05841529,
      "balance_loss_mlp": 1.02434289,
      "epoch": 0.34257199543077016,
      "flos": 34127058960000.0,
      "grad_norm": 1.6667793634933405,
      "language_loss": 0.62110519,
      "learning_rate": 3.0598678885486145e-06,
      "loss": 0.64307249,
      "num_input_tokens_seen": 61301580,
      "step": 2849,
      "time_per_iteration": 2.830700159072876
    },
    {
      "auxiliary_loss_clip": 0.01188201,
      "auxiliary_loss_mlp": 0.01033221,
      "balance_loss_clip": 1.05851281,
      "balance_loss_mlp": 1.02420282,
      "epoch": 0.34269223832140927,
      "flos": 19974018188160.0,
      "grad_norm": 3.109430287328966,
      "language_loss": 0.74692142,
      "learning_rate": 3.0592072098980282e-06,
      "loss": 0.76913565,
      "num_input_tokens_seen": 61321240,
      "step": 2850,
      "time_per_iteration": 2.6666619777679443
    },
    {
      "auxiliary_loss_clip": 0.01189874,
      "auxiliary_loss_mlp": 0.0103512,
      "balance_loss_clip": 1.06112492,
      "balance_loss_mlp": 1.02628624,
      "epoch": 0.3428124812120483,
      "flos": 27235658292480.0,
      "grad_norm": 2.25960869854751,
      "language_loss": 0.73619527,
      "learning_rate": 3.0585463705665514e-06,
      "loss": 0.75844514,
      "num_input_tokens_seen": 61341615,
      "step": 2851,
      "time_per_iteration": 2.771193027496338
    },
    {
      "auxiliary_loss_clip": 0.01191949,
      "auxiliary_loss_mlp": 0.01039971,
      "balance_loss_clip": 1.06019497,
      "balance_loss_mlp": 1.03023136,
      "epoch": 0.34293272410268744,
      "flos": 24570871079040.0,
      "grad_norm": 2.99252698190244,
      "language_loss": 0.70949531,
      "learning_rate": 3.0578853706544304e-06,
      "loss": 0.7318145,
      "num_input_tokens_seen": 61359005,
      "step": 2852,
      "time_per_iteration": 2.674161195755005
    },
    {
      "auxiliary_loss_clip": 0.01190694,
      "auxiliary_loss_mlp": 0.01031023,
      "balance_loss_clip": 1.0626353,
      "balance_loss_mlp": 1.02205276,
      "epoch": 0.34305296699332655,
      "flos": 21506865131520.0,
      "grad_norm": 3.9892423128473795,
      "language_loss": 0.66020358,
      "learning_rate": 3.0572242102619404e-06,
      "loss": 0.68242073,
      "num_input_tokens_seen": 61376160,
      "step": 2853,
      "time_per_iteration": 2.686725378036499
    },
    {
      "auxiliary_loss_clip": 0.01189894,
      "auxiliary_loss_mlp": 0.01031248,
      "balance_loss_clip": 1.06251073,
      "balance_loss_mlp": 1.02273083,
      "epoch": 0.3431732098839656,
      "flos": 24056665931520.0,
      "grad_norm": 2.9463563948178,
      "language_loss": 0.8069489,
      "learning_rate": 3.0565628894893784e-06,
      "loss": 0.82916033,
      "num_input_tokens_seen": 61396795,
      "step": 2854,
      "time_per_iteration": 2.7299115657806396
    },
    {
      "auxiliary_loss_clip": 0.0119041,
      "auxiliary_loss_mlp": 0.01034986,
      "balance_loss_clip": 1.06271458,
      "balance_loss_mlp": 1.02588415,
      "epoch": 0.3432934527746047,
      "flos": 16800879744000.0,
      "grad_norm": 1.7523151113355762,
      "language_loss": 0.74820459,
      "learning_rate": 3.0559014084370655e-06,
      "loss": 0.77045858,
      "num_input_tokens_seen": 61415320,
      "step": 2855,
      "time_per_iteration": 2.647259473800659
    },
    {
      "auxiliary_loss_clip": 0.01173565,
      "auxiliary_loss_mlp": 0.01029976,
      "balance_loss_clip": 1.06027746,
      "balance_loss_mlp": 1.02038538,
      "epoch": 0.34341369566524377,
      "flos": 23439720908160.0,
      "grad_norm": 2.3307101059916184,
      "language_loss": 0.79051566,
      "learning_rate": 3.055239767205349e-06,
      "loss": 0.81255114,
      "num_input_tokens_seen": 61437070,
      "step": 2856,
      "time_per_iteration": 2.752683639526367
    },
    {
      "auxiliary_loss_clip": 0.01180566,
      "auxiliary_loss_mlp": 0.01039766,
      "balance_loss_clip": 1.05894029,
      "balance_loss_mlp": 1.03069401,
      "epoch": 0.3435339385558829,
      "flos": 17267466435840.0,
      "grad_norm": 2.609812125813182,
      "language_loss": 0.78194129,
      "learning_rate": 3.054577965894599e-06,
      "loss": 0.80414468,
      "num_input_tokens_seen": 61453215,
      "step": 2857,
      "time_per_iteration": 2.69275164604187
    },
    {
      "auxiliary_loss_clip": 0.01194567,
      "auxiliary_loss_mlp": 0.01041823,
      "balance_loss_clip": 1.06370187,
      "balance_loss_mlp": 1.03268564,
      "epoch": 0.343654181446522,
      "flos": 22199366413440.0,
      "grad_norm": 1.7586620404615572,
      "language_loss": 0.70407975,
      "learning_rate": 3.0539160046052094e-06,
      "loss": 0.72644365,
      "num_input_tokens_seen": 61472915,
      "step": 2858,
      "time_per_iteration": 2.7594950199127197
    },
    {
      "auxiliary_loss_clip": 0.01206651,
      "auxiliary_loss_mlp": 0.01035709,
      "balance_loss_clip": 1.06038547,
      "balance_loss_mlp": 1.02619028,
      "epoch": 0.34377442433716104,
      "flos": 19901801894400.0,
      "grad_norm": 2.2799861496432943,
      "language_loss": 0.70673501,
      "learning_rate": 3.0532538834376003e-06,
      "loss": 0.72915864,
      "num_input_tokens_seen": 61492475,
      "step": 2859,
      "time_per_iteration": 2.5783116817474365
    },
    {
      "auxiliary_loss_clip": 0.0116278,
      "auxiliary_loss_mlp": 0.01034484,
      "balance_loss_clip": 1.05926251,
      "balance_loss_mlp": 1.02559078,
      "epoch": 0.34389466722780015,
      "flos": 22197678474240.0,
      "grad_norm": 3.2331835862536256,
      "language_loss": 0.78494138,
      "learning_rate": 3.0525916024922143e-06,
      "loss": 0.80691397,
      "num_input_tokens_seen": 61511660,
      "step": 2860,
      "time_per_iteration": 2.7370831966400146
    },
    {
      "auxiliary_loss_clip": 0.0117507,
      "auxiliary_loss_mlp": 0.01041749,
      "balance_loss_clip": 1.05737257,
      "balance_loss_mlp": 1.03249884,
      "epoch": 0.34401491011843927,
      "flos": 18624567110400.0,
      "grad_norm": 2.7724050260121573,
      "language_loss": 0.84220183,
      "learning_rate": 3.0519291618695193e-06,
      "loss": 0.86437005,
      "num_input_tokens_seen": 61529060,
      "step": 2861,
      "time_per_iteration": 2.633855104446411
    },
    {
      "auxiliary_loss_clip": 0.01165678,
      "auxiliary_loss_mlp": 0.01040294,
      "balance_loss_clip": 1.05499983,
      "balance_loss_mlp": 1.03145456,
      "epoch": 0.3441351530090783,
      "flos": 17858197509120.0,
      "grad_norm": 1.9987231634860263,
      "language_loss": 0.76043928,
      "learning_rate": 3.0512665616700065e-06,
      "loss": 0.78249896,
      "num_input_tokens_seen": 61548125,
      "step": 2862,
      "time_per_iteration": 2.700446367263794
    },
    {
      "auxiliary_loss_clip": 0.0118545,
      "auxiliary_loss_mlp": 0.01037147,
      "balance_loss_clip": 1.05572081,
      "balance_loss_mlp": 1.02846813,
      "epoch": 0.34425539589971743,
      "flos": 23112754381440.0,
      "grad_norm": 3.4409698830484046,
      "language_loss": 0.89387918,
      "learning_rate": 3.0506038019941933e-06,
      "loss": 0.91610515,
      "num_input_tokens_seen": 61568135,
      "step": 2863,
      "time_per_iteration": 2.6383895874023438
    },
    {
      "auxiliary_loss_clip": 0.01206963,
      "auxiliary_loss_mlp": 0.01040611,
      "balance_loss_clip": 1.06224895,
      "balance_loss_mlp": 1.03128874,
      "epoch": 0.34437563879035654,
      "flos": 21907699977600.0,
      "grad_norm": 2.4970440233087787,
      "language_loss": 0.67635185,
      "learning_rate": 3.049940882942617e-06,
      "loss": 0.69882756,
      "num_input_tokens_seen": 61586920,
      "step": 2864,
      "time_per_iteration": 2.6995062828063965
    },
    {
      "auxiliary_loss_clip": 0.01187843,
      "auxiliary_loss_mlp": 0.01031986,
      "balance_loss_clip": 1.05821061,
      "balance_loss_mlp": 1.02266383,
      "epoch": 0.3444958816809956,
      "flos": 23076915586560.0,
      "grad_norm": 2.135914072202112,
      "language_loss": 0.80830252,
      "learning_rate": 3.0492778046158448e-06,
      "loss": 0.83050084,
      "num_input_tokens_seen": 61608340,
      "step": 2865,
      "time_per_iteration": 3.582817792892456
    },
    {
      "auxiliary_loss_clip": 0.01202871,
      "auxiliary_loss_mlp": 0.01035817,
      "balance_loss_clip": 1.06171715,
      "balance_loss_mlp": 1.02712035,
      "epoch": 0.3446161245716347,
      "flos": 21908633731200.0,
      "grad_norm": 4.082216993918445,
      "language_loss": 0.77209431,
      "learning_rate": 3.0486145671144633e-06,
      "loss": 0.79448116,
      "num_input_tokens_seen": 61628130,
      "step": 2866,
      "time_per_iteration": 3.575584888458252
    },
    {
      "auxiliary_loss_clip": 0.01207945,
      "auxiliary_loss_mlp": 0.01038527,
      "balance_loss_clip": 1.06313157,
      "balance_loss_mlp": 1.02883565,
      "epoch": 0.3447363674622738,
      "flos": 25112834461440.0,
      "grad_norm": 2.8749835150209853,
      "language_loss": 0.7748071,
      "learning_rate": 3.047951170539086e-06,
      "loss": 0.79727179,
      "num_input_tokens_seen": 61647755,
      "step": 2867,
      "time_per_iteration": 2.6499805450439453
    },
    {
      "auxiliary_loss_clip": 0.0120034,
      "auxiliary_loss_mlp": 0.00762023,
      "balance_loss_clip": 1.06332493,
      "balance_loss_mlp": 1.00045311,
      "epoch": 0.3448566103529129,
      "flos": 11984684451840.0,
      "grad_norm": 2.682999363812016,
      "language_loss": 0.84905571,
      "learning_rate": 3.047287614990349e-06,
      "loss": 0.86867934,
      "num_input_tokens_seen": 61665675,
      "step": 2868,
      "time_per_iteration": 2.654209613800049
    },
    {
      "auxiliary_loss_clip": 0.01211717,
      "auxiliary_loss_mlp": 0.01038667,
      "balance_loss_clip": 1.06592512,
      "balance_loss_mlp": 1.02880228,
      "epoch": 0.344976853243552,
      "flos": 40187882465280.0,
      "grad_norm": 4.7602405485381825,
      "language_loss": 0.61881471,
      "learning_rate": 3.046623900568914e-06,
      "loss": 0.64131856,
      "num_input_tokens_seen": 61688240,
      "step": 2869,
      "time_per_iteration": 3.7507431507110596
    },
    {
      "auxiliary_loss_clip": 0.0118581,
      "auxiliary_loss_mlp": 0.01038006,
      "balance_loss_clip": 1.05795097,
      "balance_loss_mlp": 1.0283742,
      "epoch": 0.34509709613419104,
      "flos": 28723652127360.0,
      "grad_norm": 2.748397356556553,
      "language_loss": 0.69740498,
      "learning_rate": 3.045960027375465e-06,
      "loss": 0.71964312,
      "num_input_tokens_seen": 61706075,
      "step": 2870,
      "time_per_iteration": 2.7181499004364014
    },
    {
      "auxiliary_loss_clip": 0.01206524,
      "auxiliary_loss_mlp": 0.01031059,
      "balance_loss_clip": 1.06141758,
      "balance_loss_mlp": 1.02176058,
      "epoch": 0.34521733902483015,
      "flos": 29967597982080.0,
      "grad_norm": 3.081509316568568,
      "language_loss": 0.82464236,
      "learning_rate": 3.045295995510711e-06,
      "loss": 0.84701818,
      "num_input_tokens_seen": 61723045,
      "step": 2871,
      "time_per_iteration": 3.653642416000366
    },
    {
      "auxiliary_loss_clip": 0.01202117,
      "auxiliary_loss_mlp": 0.01031246,
      "balance_loss_clip": 1.06303537,
      "balance_loss_mlp": 1.02272224,
      "epoch": 0.34533758191546926,
      "flos": 27923059843200.0,
      "grad_norm": 1.8451682718029365,
      "language_loss": 0.73798198,
      "learning_rate": 3.0446318050753865e-06,
      "loss": 0.7603156,
      "num_input_tokens_seen": 61743525,
      "step": 2872,
      "time_per_iteration": 2.719212770462036
    },
    {
      "auxiliary_loss_clip": 0.01187322,
      "auxiliary_loss_mlp": 0.01035253,
      "balance_loss_clip": 1.06007242,
      "balance_loss_mlp": 1.02668762,
      "epoch": 0.3454578248061083,
      "flos": 27125879351040.0,
      "grad_norm": 2.0723711081213536,
      "language_loss": 0.77954149,
      "learning_rate": 3.0439674561702474e-06,
      "loss": 0.80176729,
      "num_input_tokens_seen": 61763025,
      "step": 2873,
      "time_per_iteration": 2.724371910095215
    },
    {
      "auxiliary_loss_clip": 0.01148966,
      "auxiliary_loss_mlp": 0.01032729,
      "balance_loss_clip": 1.05801499,
      "balance_loss_mlp": 1.02402067,
      "epoch": 0.3455780676967474,
      "flos": 19024899166080.0,
      "grad_norm": 2.231070518031152,
      "language_loss": 0.88619101,
      "learning_rate": 3.043302948896076e-06,
      "loss": 0.90800798,
      "num_input_tokens_seen": 61781630,
      "step": 2874,
      "time_per_iteration": 2.7108709812164307
    },
    {
      "auxiliary_loss_clip": 0.01208005,
      "auxiliary_loss_mlp": 0.01034131,
      "balance_loss_clip": 1.06445932,
      "balance_loss_mlp": 1.02552378,
      "epoch": 0.34569831058738654,
      "flos": 34496005507200.0,
      "grad_norm": 2.0367800599978865,
      "language_loss": 0.60457325,
      "learning_rate": 3.0426382833536756e-06,
      "loss": 0.62699461,
      "num_input_tokens_seen": 61804985,
      "step": 2875,
      "time_per_iteration": 2.808638572692871
    },
    {
      "auxiliary_loss_clip": 0.01205599,
      "auxiliary_loss_mlp": 0.00762602,
      "balance_loss_clip": 1.06282663,
      "balance_loss_mlp": 1.00041652,
      "epoch": 0.3458185534780256,
      "flos": 31138681098240.0,
      "grad_norm": 3.9488753075736684,
      "language_loss": 0.77936971,
      "learning_rate": 3.041973459643877e-06,
      "loss": 0.79905176,
      "num_input_tokens_seen": 61824440,
      "step": 2876,
      "time_per_iteration": 2.699793577194214
    },
    {
      "auxiliary_loss_clip": 0.01189045,
      "auxiliary_loss_mlp": 0.01037351,
      "balance_loss_clip": 1.05962145,
      "balance_loss_mlp": 1.02801681,
      "epoch": 0.3459387963686647,
      "flos": 32452508862720.0,
      "grad_norm": 2.274085852571922,
      "language_loss": 0.67314911,
      "learning_rate": 3.0413084778675334e-06,
      "loss": 0.69541311,
      "num_input_tokens_seen": 61845690,
      "step": 2877,
      "time_per_iteration": 2.7777819633483887
    },
    {
      "auxiliary_loss_clip": 0.0120372,
      "auxiliary_loss_mlp": 0.01030117,
      "balance_loss_clip": 1.06227994,
      "balance_loss_mlp": 1.02168262,
      "epoch": 0.3460590392593038,
      "flos": 24675658030080.0,
      "grad_norm": 2.4694754792144282,
      "language_loss": 0.84174073,
      "learning_rate": 3.0406433381255214e-06,
      "loss": 0.86407912,
      "num_input_tokens_seen": 61863725,
      "step": 2878,
      "time_per_iteration": 2.6992247104644775
    },
    {
      "auxiliary_loss_clip": 0.01203107,
      "auxiliary_loss_mlp": 0.01030908,
      "balance_loss_clip": 1.06206894,
      "balance_loss_mlp": 1.02256036,
      "epoch": 0.34617928214994287,
      "flos": 18807316531200.0,
      "grad_norm": 4.2429165400776,
      "language_loss": 0.82559979,
      "learning_rate": 3.0399780405187425e-06,
      "loss": 0.84793997,
      "num_input_tokens_seen": 61882720,
      "step": 2879,
      "time_per_iteration": 2.6266210079193115
    },
    {
      "auxiliary_loss_clip": 0.01204682,
      "auxiliary_loss_mlp": 0.0103014,
      "balance_loss_clip": 1.06379855,
      "balance_loss_mlp": 1.02188444,
      "epoch": 0.346299525040582,
      "flos": 24857653265280.0,
      "grad_norm": 7.074286247810877,
      "language_loss": 0.78840238,
      "learning_rate": 3.0393125851481216e-06,
      "loss": 0.8107506,
      "num_input_tokens_seen": 61902595,
      "step": 2880,
      "time_per_iteration": 2.617142677307129
    },
    {
      "auxiliary_loss_clip": 0.01153962,
      "auxiliary_loss_mlp": 0.010391,
      "balance_loss_clip": 1.05620372,
      "balance_loss_mlp": 1.03052282,
      "epoch": 0.3464197679312211,
      "flos": 16434914025600.0,
      "grad_norm": 3.0048891768456736,
      "language_loss": 0.86354738,
      "learning_rate": 3.038646972114608e-06,
      "loss": 0.88547796,
      "num_input_tokens_seen": 61918920,
      "step": 2881,
      "time_per_iteration": 2.843235492706299
    },
    {
      "auxiliary_loss_clip": 0.01189076,
      "auxiliary_loss_mlp": 0.0103388,
      "balance_loss_clip": 1.06144512,
      "balance_loss_mlp": 1.02516007,
      "epoch": 0.34654001082186014,
      "flos": 22382474970240.0,
      "grad_norm": 2.2855457428787695,
      "language_loss": 0.67486584,
      "learning_rate": 3.037981201519174e-06,
      "loss": 0.69709545,
      "num_input_tokens_seen": 61939520,
      "step": 2882,
      "time_per_iteration": 2.780407428741455
    },
    {
      "auxiliary_loss_clip": 0.01184663,
      "auxiliary_loss_mlp": 0.01044525,
      "balance_loss_clip": 1.06206656,
      "balance_loss_mlp": 1.03573346,
      "epoch": 0.34666025371249926,
      "flos": 19573901614080.0,
      "grad_norm": 2.912329363611193,
      "language_loss": 0.71548891,
      "learning_rate": 3.0373152734628175e-06,
      "loss": 0.73778081,
      "num_input_tokens_seen": 61957800,
      "step": 2883,
      "time_per_iteration": 2.66568660736084
    },
    {
      "auxiliary_loss_clip": 0.01202407,
      "auxiliary_loss_mlp": 0.01031749,
      "balance_loss_clip": 1.06055903,
      "balance_loss_mlp": 1.02316594,
      "epoch": 0.34678049660313837,
      "flos": 15267637751040.0,
      "grad_norm": 3.563456568097466,
      "language_loss": 0.76390535,
      "learning_rate": 3.0366491880465584e-06,
      "loss": 0.7862469,
      "num_input_tokens_seen": 61975820,
      "step": 2884,
      "time_per_iteration": 2.626483201980591
    },
    {
      "auxiliary_loss_clip": 0.0119317,
      "auxiliary_loss_mlp": 0.01042823,
      "balance_loss_clip": 1.06608295,
      "balance_loss_mlp": 1.03341758,
      "epoch": 0.3469007394937774,
      "flos": 21181550630400.0,
      "grad_norm": 2.456748856222078,
      "language_loss": 0.8242985,
      "learning_rate": 3.035982945371443e-06,
      "loss": 0.84665835,
      "num_input_tokens_seen": 61997515,
      "step": 2885,
      "time_per_iteration": 2.6949119567871094
    },
    {
      "auxiliary_loss_clip": 0.01194771,
      "auxiliary_loss_mlp": 0.01041869,
      "balance_loss_clip": 1.06687438,
      "balance_loss_mlp": 1.03297639,
      "epoch": 0.34702098238441653,
      "flos": 22375471818240.0,
      "grad_norm": 3.130341570031837,
      "language_loss": 0.85554427,
      "learning_rate": 3.035316545538537e-06,
      "loss": 0.87791061,
      "num_input_tokens_seen": 62016310,
      "step": 2886,
      "time_per_iteration": 2.7079591751098633
    },
    {
      "auxiliary_loss_clip": 0.01180621,
      "auxiliary_loss_mlp": 0.01042928,
      "balance_loss_clip": 1.06551623,
      "balance_loss_mlp": 1.03481019,
      "epoch": 0.3471412252750556,
      "flos": 22929430343040.0,
      "grad_norm": 3.1356789116451114,
      "language_loss": 0.79573435,
      "learning_rate": 3.034649988648935e-06,
      "loss": 0.8179698,
      "num_input_tokens_seen": 62036075,
      "step": 2887,
      "time_per_iteration": 2.6875078678131104
    },
    {
      "auxiliary_loss_clip": 0.01193108,
      "auxiliary_loss_mlp": 0.01032791,
      "balance_loss_clip": 1.06249762,
      "balance_loss_mlp": 1.02410626,
      "epoch": 0.3472614681656947,
      "flos": 21324259365120.0,
      "grad_norm": 2.050248244888865,
      "language_loss": 0.80745506,
      "learning_rate": 3.033983274803752e-06,
      "loss": 0.82971406,
      "num_input_tokens_seen": 62055865,
      "step": 2888,
      "time_per_iteration": 2.6858742237091064
    },
    {
      "auxiliary_loss_clip": 0.01190013,
      "auxiliary_loss_mlp": 0.01038945,
      "balance_loss_clip": 1.06258321,
      "balance_loss_mlp": 1.03064775,
      "epoch": 0.3473817110563338,
      "flos": 23475739271040.0,
      "grad_norm": 3.4444739640498043,
      "language_loss": 0.73173696,
      "learning_rate": 3.0333164041041283e-06,
      "loss": 0.75402653,
      "num_input_tokens_seen": 62072180,
      "step": 2889,
      "time_per_iteration": 2.664411783218384
    },
    {
      "auxiliary_loss_clip": 0.01205678,
      "auxiliary_loss_mlp": 0.01039429,
      "balance_loss_clip": 1.06571126,
      "balance_loss_mlp": 1.03005958,
      "epoch": 0.34750195394697286,
      "flos": 22346025644160.0,
      "grad_norm": 2.315587916134687,
      "language_loss": 0.7206986,
      "learning_rate": 3.032649376651228e-06,
      "loss": 0.74314964,
      "num_input_tokens_seen": 62091600,
      "step": 2890,
      "time_per_iteration": 2.689354658126831
    },
    {
      "auxiliary_loss_clip": 0.01207549,
      "auxiliary_loss_mlp": 0.00763211,
      "balance_loss_clip": 1.06399775,
      "balance_loss_mlp": 1.00041401,
      "epoch": 0.347622196837612,
      "flos": 29095004885760.0,
      "grad_norm": 3.678720365130533,
      "language_loss": 0.75935721,
      "learning_rate": 3.031982192546238e-06,
      "loss": 0.77906477,
      "num_input_tokens_seen": 62114695,
      "step": 2891,
      "time_per_iteration": 2.664144515991211
    },
    {
      "auxiliary_loss_clip": 0.01192496,
      "auxiliary_loss_mlp": 0.01029387,
      "balance_loss_clip": 1.06199837,
      "balance_loss_mlp": 1.0209229,
      "epoch": 0.3477424397282511,
      "flos": 22455732758400.0,
      "grad_norm": 2.409358232944214,
      "language_loss": 0.94907159,
      "learning_rate": 3.0313148518903696e-06,
      "loss": 0.97129047,
      "num_input_tokens_seen": 62134520,
      "step": 2892,
      "time_per_iteration": 3.654705762863159
    },
    {
      "auxiliary_loss_clip": 0.01207032,
      "auxiliary_loss_mlp": 0.00762298,
      "balance_loss_clip": 1.06457341,
      "balance_loss_mlp": 1.00033307,
      "epoch": 0.34786268261889014,
      "flos": 15778790242560.0,
      "grad_norm": 3.2143810557715717,
      "language_loss": 0.81549722,
      "learning_rate": 3.030647354784859e-06,
      "loss": 0.83519053,
      "num_input_tokens_seen": 62151560,
      "step": 2893,
      "time_per_iteration": 2.6316726207733154
    },
    {
      "auxiliary_loss_clip": 0.01176469,
      "auxiliary_loss_mlp": 0.01032996,
      "balance_loss_clip": 1.05857205,
      "balance_loss_mlp": 1.02407885,
      "epoch": 0.34798292550952925,
      "flos": 20777627214720.0,
      "grad_norm": 3.4179250026131114,
      "language_loss": 0.77677554,
      "learning_rate": 3.029979701330964e-06,
      "loss": 0.79887021,
      "num_input_tokens_seen": 62170985,
      "step": 2894,
      "time_per_iteration": 2.7167704105377197
    },
    {
      "auxiliary_loss_clip": 0.01191303,
      "auxiliary_loss_mlp": 0.01032294,
      "balance_loss_clip": 1.06464171,
      "balance_loss_mlp": 1.02361596,
      "epoch": 0.34810316840016836,
      "flos": 19937820257280.0,
      "grad_norm": 15.719209958761187,
      "language_loss": 0.80242586,
      "learning_rate": 3.029311891629966e-06,
      "loss": 0.82466185,
      "num_input_tokens_seen": 62189440,
      "step": 2895,
      "time_per_iteration": 2.7214884757995605
    },
    {
      "auxiliary_loss_clip": 0.01186316,
      "auxiliary_loss_mlp": 0.01036406,
      "balance_loss_clip": 1.058532,
      "balance_loss_mlp": 1.02788234,
      "epoch": 0.3482234112908074,
      "flos": 23623296341760.0,
      "grad_norm": 2.4521584668651246,
      "language_loss": 0.74834388,
      "learning_rate": 3.0286439257831744e-06,
      "loss": 0.77057111,
      "num_input_tokens_seen": 62208910,
      "step": 2896,
      "time_per_iteration": 3.5794076919555664
    },
    {
      "auxiliary_loss_clip": 0.01196355,
      "auxiliary_loss_mlp": 0.01036594,
      "balance_loss_clip": 1.06192231,
      "balance_loss_mlp": 1.02605617,
      "epoch": 0.3483436541814465,
      "flos": 23986712194560.0,
      "grad_norm": 2.0727482665549686,
      "language_loss": 0.7156536,
      "learning_rate": 3.0279758038919156e-06,
      "loss": 0.73798311,
      "num_input_tokens_seen": 62227135,
      "step": 2897,
      "time_per_iteration": 3.5634517669677734
    },
    {
      "auxiliary_loss_clip": 0.01154345,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.05795515,
      "balance_loss_mlp": 1.02258492,
      "epoch": 0.34846389707208564,
      "flos": 22638338524800.0,
      "grad_norm": 2.2276514264854037,
      "language_loss": 0.78444773,
      "learning_rate": 3.0273075260575455e-06,
      "loss": 0.80630612,
      "num_input_tokens_seen": 62246035,
      "step": 2898,
      "time_per_iteration": 2.7915878295898438
    },
    {
      "auxiliary_loss_clip": 0.01192443,
      "auxiliary_loss_mlp": 0.00763071,
      "balance_loss_clip": 1.06115556,
      "balance_loss_mlp": 1.00037408,
      "epoch": 0.3485841399627247,
      "flos": 21792857218560.0,
      "grad_norm": 3.3240415197015287,
      "language_loss": 0.80677652,
      "learning_rate": 3.0266390923814396e-06,
      "loss": 0.82633162,
      "num_input_tokens_seen": 62264095,
      "step": 2899,
      "time_per_iteration": 2.707016706466675
    },
    {
      "auxiliary_loss_clip": 0.01190794,
      "auxiliary_loss_mlp": 0.01041371,
      "balance_loss_clip": 1.06141031,
      "balance_loss_mlp": 1.03244257,
      "epoch": 0.3487043828533638,
      "flos": 17019036996480.0,
      "grad_norm": 3.5576549884974638,
      "language_loss": 0.82471555,
      "learning_rate": 3.0259705029650008e-06,
      "loss": 0.8470372,
      "num_input_tokens_seen": 62282025,
      "step": 2900,
      "time_per_iteration": 2.8498189449310303
    },
    {
      "auxiliary_loss_clip": 0.01191879,
      "auxiliary_loss_mlp": 0.01033456,
      "balance_loss_clip": 1.06012106,
      "balance_loss_mlp": 1.02495027,
      "epoch": 0.34882462574400286,
      "flos": 22601135013120.0,
      "grad_norm": 3.888068950227685,
      "language_loss": 0.72559637,
      "learning_rate": 3.025301757909652e-06,
      "loss": 0.7478497,
      "num_input_tokens_seen": 62302220,
      "step": 2901,
      "time_per_iteration": 2.657891035079956
    },
    {
      "auxiliary_loss_clip": 0.01208376,
      "auxiliary_loss_mlp": 0.01032588,
      "balance_loss_clip": 1.06221497,
      "balance_loss_mlp": 1.02350438,
      "epoch": 0.34894486863464197,
      "flos": 29861518141440.0,
      "grad_norm": 1.8532114065922625,
      "language_loss": 0.80672312,
      "learning_rate": 3.024632857316842e-06,
      "loss": 0.82913268,
      "num_input_tokens_seen": 62323535,
      "step": 2902,
      "time_per_iteration": 2.707066059112549
    },
    {
      "auxiliary_loss_clip": 0.01183726,
      "auxiliary_loss_mlp": 0.01036724,
      "balance_loss_clip": 1.06418145,
      "balance_loss_mlp": 1.02752757,
      "epoch": 0.3490651115252811,
      "flos": 22122265870080.0,
      "grad_norm": 4.132011667279632,
      "language_loss": 0.77884382,
      "learning_rate": 3.0239638012880412e-06,
      "loss": 0.80104834,
      "num_input_tokens_seen": 62343430,
      "step": 2903,
      "time_per_iteration": 2.7324535846710205
    },
    {
      "auxiliary_loss_clip": 0.01206173,
      "auxiliary_loss_mlp": 0.01033684,
      "balance_loss_clip": 1.06246543,
      "balance_loss_mlp": 1.02373028,
      "epoch": 0.34918535441592014,
      "flos": 12676682943360.0,
      "grad_norm": 3.7385195453644493,
      "language_loss": 0.81480896,
      "learning_rate": 3.0232945899247466e-06,
      "loss": 0.83720756,
      "num_input_tokens_seen": 62360365,
      "step": 2904,
      "time_per_iteration": 2.7595949172973633
    },
    {
      "auxiliary_loss_clip": 0.01195879,
      "auxiliary_loss_mlp": 0.00763638,
      "balance_loss_clip": 1.06154251,
      "balance_loss_mlp": 1.00038028,
      "epoch": 0.34930559730655925,
      "flos": 23185617120000.0,
      "grad_norm": 2.4648173465277106,
      "language_loss": 0.77332377,
      "learning_rate": 3.022625223328476e-06,
      "loss": 0.79291898,
      "num_input_tokens_seen": 62382105,
      "step": 2905,
      "time_per_iteration": 2.7989368438720703
    },
    {
      "auxiliary_loss_clip": 0.011784,
      "auxiliary_loss_mlp": 0.01031041,
      "balance_loss_clip": 1.05937135,
      "balance_loss_mlp": 1.02188563,
      "epoch": 0.34942584019719836,
      "flos": 22855023319680.0,
      "grad_norm": 1.7361462893403035,
      "language_loss": 0.69468731,
      "learning_rate": 3.0219557016007723e-06,
      "loss": 0.71678174,
      "num_input_tokens_seen": 62402235,
      "step": 2906,
      "time_per_iteration": 2.8164846897125244
    },
    {
      "auxiliary_loss_clip": 0.01199826,
      "auxiliary_loss_mlp": 0.01032102,
      "balance_loss_clip": 1.06020284,
      "balance_loss_mlp": 1.02242196,
      "epoch": 0.3495460830878374,
      "flos": 24426043441920.0,
      "grad_norm": 2.0953891015552757,
      "language_loss": 0.69873029,
      "learning_rate": 3.021286024843202e-06,
      "loss": 0.72104955,
      "num_input_tokens_seen": 62420430,
      "step": 2907,
      "time_per_iteration": 2.698709487915039
    },
    {
      "auxiliary_loss_clip": 0.01074869,
      "auxiliary_loss_mlp": 0.01013931,
      "balance_loss_clip": 1.0272131,
      "balance_loss_mlp": 1.01208353,
      "epoch": 0.3496663259784765,
      "flos": 70008749389440.0,
      "grad_norm": 1.0734419189172384,
      "language_loss": 0.64701998,
      "learning_rate": 3.0206161931573526e-06,
      "loss": 0.66790795,
      "num_input_tokens_seen": 62472980,
      "step": 2908,
      "time_per_iteration": 3.163165807723999
    },
    {
      "auxiliary_loss_clip": 0.0118723,
      "auxiliary_loss_mlp": 0.01033347,
      "balance_loss_clip": 1.05814278,
      "balance_loss_mlp": 1.0244416,
      "epoch": 0.34978656886911563,
      "flos": 28692805322880.0,
      "grad_norm": 1.709863230306198,
      "language_loss": 0.92888141,
      "learning_rate": 3.0199462066448388e-06,
      "loss": 0.95108712,
      "num_input_tokens_seen": 62495175,
      "step": 2909,
      "time_per_iteration": 2.7781269550323486
    },
    {
      "auxiliary_loss_clip": 0.01206705,
      "auxiliary_loss_mlp": 0.01028887,
      "balance_loss_clip": 1.06446958,
      "balance_loss_mlp": 1.01893926,
      "epoch": 0.3499068117597547,
      "flos": 21142156389120.0,
      "grad_norm": 1.9874425840310483,
      "language_loss": 0.69329035,
      "learning_rate": 3.019276065407296e-06,
      "loss": 0.71564627,
      "num_input_tokens_seen": 62514295,
      "step": 2910,
      "time_per_iteration": 2.63798451423645
    },
    {
      "auxiliary_loss_clip": 0.01182301,
      "auxiliary_loss_mlp": 0.01035799,
      "balance_loss_clip": 1.05895472,
      "balance_loss_mlp": 1.02639902,
      "epoch": 0.3500270546503938,
      "flos": 22782699285120.0,
      "grad_norm": 1.854547387693281,
      "language_loss": 0.80680364,
      "learning_rate": 3.018605769546385e-06,
      "loss": 0.82898462,
      "num_input_tokens_seen": 62534850,
      "step": 2911,
      "time_per_iteration": 2.770711898803711
    },
    {
      "auxiliary_loss_clip": 0.01197011,
      "auxiliary_loss_mlp": 0.01034333,
      "balance_loss_clip": 1.06225729,
      "balance_loss_mlp": 1.02508867,
      "epoch": 0.3501472975410329,
      "flos": 22894058424960.0,
      "grad_norm": 2.674432100736347,
      "language_loss": 0.79943871,
      "learning_rate": 3.017935319163788e-06,
      "loss": 0.82175219,
      "num_input_tokens_seen": 62553810,
      "step": 2912,
      "time_per_iteration": 2.6708271503448486
    },
    {
      "auxiliary_loss_clip": 0.0120488,
      "auxiliary_loss_mlp": 0.01031904,
      "balance_loss_clip": 1.06032646,
      "balance_loss_mlp": 1.02259398,
      "epoch": 0.35026754043167196,
      "flos": 25446588658560.0,
      "grad_norm": 2.5477059865167875,
      "language_loss": 0.70454931,
      "learning_rate": 3.017264714361213e-06,
      "loss": 0.72691709,
      "num_input_tokens_seen": 62573460,
      "step": 2913,
      "time_per_iteration": 2.657379388809204
    },
    {
      "auxiliary_loss_clip": 0.0117837,
      "auxiliary_loss_mlp": 0.01031351,
      "balance_loss_clip": 1.05960357,
      "balance_loss_mlp": 1.02250612,
      "epoch": 0.3503877833223111,
      "flos": 19573757959680.0,
      "grad_norm": 2.2462050587872673,
      "language_loss": 0.8260802,
      "learning_rate": 3.016593955240389e-06,
      "loss": 0.84817737,
      "num_input_tokens_seen": 62592150,
      "step": 2914,
      "time_per_iteration": 2.6593008041381836
    },
    {
      "auxiliary_loss_clip": 0.01091991,
      "auxiliary_loss_mlp": 0.01008599,
      "balance_loss_clip": 1.0241431,
      "balance_loss_mlp": 1.00678658,
      "epoch": 0.3505080262129502,
      "flos": 65072075880960.0,
      "grad_norm": 0.8231461901838294,
      "language_loss": 0.6367358,
      "learning_rate": 3.015923041903071e-06,
      "loss": 0.65774167,
      "num_input_tokens_seen": 62658275,
      "step": 2915,
      "time_per_iteration": 3.28428316116333
    },
    {
      "auxiliary_loss_clip": 0.01203439,
      "auxiliary_loss_mlp": 0.01038977,
      "balance_loss_clip": 1.06153786,
      "balance_loss_mlp": 1.03072155,
      "epoch": 0.35062826910358924,
      "flos": 29314562768640.0,
      "grad_norm": 2.413078836080311,
      "language_loss": 0.83967024,
      "learning_rate": 3.0152519744510347e-06,
      "loss": 0.8620944,
      "num_input_tokens_seen": 62678075,
      "step": 2916,
      "time_per_iteration": 2.687824249267578
    },
    {
      "auxiliary_loss_clip": 0.01182349,
      "auxiliary_loss_mlp": 0.01030544,
      "balance_loss_clip": 1.05626535,
      "balance_loss_mlp": 1.02183604,
      "epoch": 0.35074851199422835,
      "flos": 23987717775360.0,
      "grad_norm": 2.2055312311038846,
      "language_loss": 0.82721466,
      "learning_rate": 3.014580752986081e-06,
      "loss": 0.84934354,
      "num_input_tokens_seen": 62696950,
      "step": 2917,
      "time_per_iteration": 2.6422181129455566
    },
    {
      "auxiliary_loss_clip": 0.0120329,
      "auxiliary_loss_mlp": 0.01036778,
      "balance_loss_clip": 1.06208587,
      "balance_loss_mlp": 1.02811098,
      "epoch": 0.3508687548848674,
      "flos": 15224436668160.0,
      "grad_norm": 3.5334873498529267,
      "language_loss": 0.78831494,
      "learning_rate": 3.0139093776100345e-06,
      "loss": 0.81071556,
      "num_input_tokens_seen": 62713540,
      "step": 2918,
      "time_per_iteration": 3.5326156616210938
    },
    {
      "auxiliary_loss_clip": 0.0118837,
      "auxiliary_loss_mlp": 0.01028785,
      "balance_loss_clip": 1.05904758,
      "balance_loss_mlp": 1.01921272,
      "epoch": 0.3509889977755065,
      "flos": 21361750185600.0,
      "grad_norm": 2.1353347363991353,
      "language_loss": 0.7611841,
      "learning_rate": 3.013237848424741e-06,
      "loss": 0.78335571,
      "num_input_tokens_seen": 62732925,
      "step": 2919,
      "time_per_iteration": 2.6435208320617676
    },
    {
      "auxiliary_loss_clip": 0.01189428,
      "auxiliary_loss_mlp": 0.01033358,
      "balance_loss_clip": 1.06061006,
      "balance_loss_mlp": 1.02372038,
      "epoch": 0.35110924066614563,
      "flos": 19135360465920.0,
      "grad_norm": 3.6333871616813385,
      "language_loss": 0.75510073,
      "learning_rate": 3.012566165532072e-06,
      "loss": 0.77732861,
      "num_input_tokens_seen": 62751715,
      "step": 2920,
      "time_per_iteration": 2.6977434158325195
    },
    {
      "auxiliary_loss_clip": 0.0118731,
      "auxiliary_loss_mlp": 0.01031098,
      "balance_loss_clip": 1.05943322,
      "balance_loss_mlp": 1.02272356,
      "epoch": 0.3512294835567847,
      "flos": 21980885938560.0,
      "grad_norm": 2.3739570620051222,
      "language_loss": 0.77255964,
      "learning_rate": 3.0118943290339207e-06,
      "loss": 0.79474372,
      "num_input_tokens_seen": 62771925,
      "step": 2921,
      "time_per_iteration": 2.664954662322998
    },
    {
      "auxiliary_loss_clip": 0.0119099,
      "auxiliary_loss_mlp": 0.00762913,
      "balance_loss_clip": 1.06162131,
      "balance_loss_mlp": 1.00043905,
      "epoch": 0.3513497264474238,
      "flos": 17817294896640.0,
      "grad_norm": 1.9021349816410247,
      "language_loss": 0.6839233,
      "learning_rate": 3.011222339032204e-06,
      "loss": 0.70346236,
      "num_input_tokens_seen": 62790075,
      "step": 2922,
      "time_per_iteration": 3.682722806930542
    },
    {
      "auxiliary_loss_clip": 0.01168169,
      "auxiliary_loss_mlp": 0.01042229,
      "balance_loss_clip": 1.05088758,
      "balance_loss_mlp": 1.03361666,
      "epoch": 0.3514699693380629,
      "flos": 26943417239040.0,
      "grad_norm": 2.129602954005827,
      "language_loss": 0.69594646,
      "learning_rate": 3.0105501956288626e-06,
      "loss": 0.71805048,
      "num_input_tokens_seen": 62810545,
      "step": 2923,
      "time_per_iteration": 3.617091178894043
    },
    {
      "auxiliary_loss_clip": 0.01193661,
      "auxiliary_loss_mlp": 0.00763597,
      "balance_loss_clip": 1.0581789,
      "balance_loss_mlp": 1.00033545,
      "epoch": 0.35159021222870196,
      "flos": 15267565923840.0,
      "grad_norm": 2.140463625283783,
      "language_loss": 0.72755051,
      "learning_rate": 3.0098778989258602e-06,
      "loss": 0.74712312,
      "num_input_tokens_seen": 62829155,
      "step": 2924,
      "time_per_iteration": 2.5927820205688477
    },
    {
      "auxiliary_loss_clip": 0.01188415,
      "auxiliary_loss_mlp": 0.01043546,
      "balance_loss_clip": 1.06178951,
      "balance_loss_mlp": 1.03481364,
      "epoch": 0.35171045511934107,
      "flos": 13984154000640.0,
      "grad_norm": 2.359132851548591,
      "language_loss": 0.88558662,
      "learning_rate": 3.009205449025183e-06,
      "loss": 0.90790629,
      "num_input_tokens_seen": 62845350,
      "step": 2925,
      "time_per_iteration": 2.6882436275482178
    },
    {
      "auxiliary_loss_clip": 0.01181484,
      "auxiliary_loss_mlp": 0.01039439,
      "balance_loss_clip": 1.05472374,
      "balance_loss_mlp": 1.02962828,
      "epoch": 0.3518306980099802,
      "flos": 14283434119680.0,
      "grad_norm": 2.216905377051595,
      "language_loss": 0.62964368,
      "learning_rate": 3.008532846028842e-06,
      "loss": 0.65185291,
      "num_input_tokens_seen": 62862110,
      "step": 2926,
      "time_per_iteration": 2.61690616607666
    },
    {
      "auxiliary_loss_clip": 0.01159607,
      "auxiliary_loss_mlp": 0.01036879,
      "balance_loss_clip": 1.05611515,
      "balance_loss_mlp": 1.02699089,
      "epoch": 0.35195094090061924,
      "flos": 27052872958080.0,
      "grad_norm": 2.821332883446634,
      "language_loss": 0.72277749,
      "learning_rate": 3.0078600900388694e-06,
      "loss": 0.74474233,
      "num_input_tokens_seen": 62882415,
      "step": 2927,
      "time_per_iteration": 2.8019766807556152
    },
    {
      "auxiliary_loss_clip": 0.01184037,
      "auxiliary_loss_mlp": 0.01032346,
      "balance_loss_clip": 1.05575621,
      "balance_loss_mlp": 1.02360189,
      "epoch": 0.35207118379125835,
      "flos": 25629266252160.0,
      "grad_norm": 8.17007805929718,
      "language_loss": 0.74508542,
      "learning_rate": 3.007187181157323e-06,
      "loss": 0.76724923,
      "num_input_tokens_seen": 62902425,
      "step": 2928,
      "time_per_iteration": 2.7485971450805664
    },
    {
      "auxiliary_loss_clip": 0.01192765,
      "auxiliary_loss_mlp": 0.01033027,
      "balance_loss_clip": 1.05962038,
      "balance_loss_mlp": 1.02393675,
      "epoch": 0.35219142668189746,
      "flos": 18004713085440.0,
      "grad_norm": 4.655752822587067,
      "language_loss": 0.68673432,
      "learning_rate": 3.006514119486282e-06,
      "loss": 0.70899218,
      "num_input_tokens_seen": 62919255,
      "step": 2929,
      "time_per_iteration": 2.642591953277588
    },
    {
      "auxiliary_loss_clip": 0.01173942,
      "auxiliary_loss_mlp": 0.01036356,
      "balance_loss_clip": 1.0560298,
      "balance_loss_mlp": 1.02732015,
      "epoch": 0.3523116695725365,
      "flos": 14028109269120.0,
      "grad_norm": 2.0842613332249353,
      "language_loss": 0.70058459,
      "learning_rate": 3.005840905127849e-06,
      "loss": 0.7226876,
      "num_input_tokens_seen": 62936160,
      "step": 2930,
      "time_per_iteration": 2.7077107429504395
    },
    {
      "auxiliary_loss_clip": 0.01189706,
      "auxiliary_loss_mlp": 0.01033182,
      "balance_loss_clip": 1.06043184,
      "balance_loss_mlp": 1.02365124,
      "epoch": 0.3524319124631756,
      "flos": 21433966479360.0,
      "grad_norm": 3.5937253951779335,
      "language_loss": 0.87536943,
      "learning_rate": 3.0051675381841516e-06,
      "loss": 0.89759839,
      "num_input_tokens_seen": 62953470,
      "step": 2931,
      "time_per_iteration": 2.634821653366089
    },
    {
      "auxiliary_loss_clip": 0.01180864,
      "auxiliary_loss_mlp": 0.01036724,
      "balance_loss_clip": 1.05831718,
      "balance_loss_mlp": 1.02793217,
      "epoch": 0.3525521553538147,
      "flos": 26322773114880.0,
      "grad_norm": 1.641354104294742,
      "language_loss": 0.77104527,
      "learning_rate": 3.0044940187573363e-06,
      "loss": 0.79322112,
      "num_input_tokens_seen": 62974480,
      "step": 2932,
      "time_per_iteration": 2.7572238445281982
    },
    {
      "auxiliary_loss_clip": 0.011776,
      "auxiliary_loss_mlp": 0.01026977,
      "balance_loss_clip": 1.06120849,
      "balance_loss_mlp": 1.01710045,
      "epoch": 0.3526723982444538,
      "flos": 21543314457600.0,
      "grad_norm": 2.7727325086237773,
      "language_loss": 0.65698147,
      "learning_rate": 3.003820346949578e-06,
      "loss": 0.67902732,
      "num_input_tokens_seen": 62992560,
      "step": 2933,
      "time_per_iteration": 2.7614715099334717
    },
    {
      "auxiliary_loss_clip": 0.01203388,
      "auxiliary_loss_mlp": 0.01035376,
      "balance_loss_clip": 1.05899739,
      "balance_loss_mlp": 1.02579105,
      "epoch": 0.3527926411350929,
      "flos": 23733649900800.0,
      "grad_norm": 2.138450177644929,
      "language_loss": 0.79796541,
      "learning_rate": 3.003146522863071e-06,
      "loss": 0.82035303,
      "num_input_tokens_seen": 63013445,
      "step": 2934,
      "time_per_iteration": 2.670621156692505
    },
    {
      "auxiliary_loss_clip": 0.01203774,
      "auxiliary_loss_mlp": 0.01041664,
      "balance_loss_clip": 1.06129205,
      "balance_loss_mlp": 1.03302717,
      "epoch": 0.35291288402573195,
      "flos": 30445461544320.0,
      "grad_norm": 2.1349572528973826,
      "language_loss": 0.86173433,
      "learning_rate": 3.0024725466000345e-06,
      "loss": 0.88418865,
      "num_input_tokens_seen": 63033400,
      "step": 2935,
      "time_per_iteration": 2.7043135166168213
    },
    {
      "auxiliary_loss_clip": 0.01153867,
      "auxiliary_loss_mlp": 0.01036912,
      "balance_loss_clip": 1.05423379,
      "balance_loss_mlp": 1.02841234,
      "epoch": 0.35303312691637107,
      "flos": 23112179763840.0,
      "grad_norm": 2.2986468045719817,
      "language_loss": 0.79386055,
      "learning_rate": 3.0017984182627087e-06,
      "loss": 0.81576836,
      "num_input_tokens_seen": 63052725,
      "step": 2936,
      "time_per_iteration": 2.7761054039001465
    },
    {
      "auxiliary_loss_clip": 0.01201568,
      "auxiliary_loss_mlp": 0.01037077,
      "balance_loss_clip": 1.05923784,
      "balance_loss_mlp": 1.02810073,
      "epoch": 0.3531533698070102,
      "flos": 21835699165440.0,
      "grad_norm": 2.5197883861764434,
      "language_loss": 0.81827706,
      "learning_rate": 3.00112413795336e-06,
      "loss": 0.84066349,
      "num_input_tokens_seen": 63072560,
      "step": 2937,
      "time_per_iteration": 2.62784743309021
    },
    {
      "auxiliary_loss_clip": 0.0117936,
      "auxiliary_loss_mlp": 0.01038195,
      "balance_loss_clip": 1.05521631,
      "balance_loss_mlp": 1.02886724,
      "epoch": 0.35327361269764923,
      "flos": 15778969810560.0,
      "grad_norm": 2.6588983893032037,
      "language_loss": 0.79936916,
      "learning_rate": 3.000449705774275e-06,
      "loss": 0.82154477,
      "num_input_tokens_seen": 63090800,
      "step": 2938,
      "time_per_iteration": 2.6990740299224854
    },
    {
      "auxiliary_loss_clip": 0.01188424,
      "auxiliary_loss_mlp": 0.01028038,
      "balance_loss_clip": 1.05922937,
      "balance_loss_mlp": 1.01916289,
      "epoch": 0.35339385558828834,
      "flos": 22090413484800.0,
      "grad_norm": 3.3066397751160075,
      "language_loss": 0.71731275,
      "learning_rate": 2.9997751218277654e-06,
      "loss": 0.7394774,
      "num_input_tokens_seen": 63108955,
      "step": 2939,
      "time_per_iteration": 2.708916664123535
    },
    {
      "auxiliary_loss_clip": 0.01155802,
      "auxiliary_loss_mlp": 0.01033926,
      "balance_loss_clip": 1.05162632,
      "balance_loss_mlp": 1.02474737,
      "epoch": 0.35351409847892745,
      "flos": 24165008328960.0,
      "grad_norm": 2.572939147577485,
      "language_loss": 0.78264058,
      "learning_rate": 2.999100386216166e-06,
      "loss": 0.80453789,
      "num_input_tokens_seen": 63127895,
      "step": 2940,
      "time_per_iteration": 2.796391248703003
    },
    {
      "auxiliary_loss_clip": 0.01201214,
      "auxiliary_loss_mlp": 0.01035661,
      "balance_loss_clip": 1.05942941,
      "balance_loss_mlp": 1.02692938,
      "epoch": 0.3536343413695665,
      "flos": 27052298340480.0,
      "grad_norm": 1.9746478614776222,
      "language_loss": 0.7501061,
      "learning_rate": 2.998425499041831e-06,
      "loss": 0.77247483,
      "num_input_tokens_seen": 63148410,
      "step": 2941,
      "time_per_iteration": 2.7708375453948975
    },
    {
      "auxiliary_loss_clip": 0.01095482,
      "auxiliary_loss_mlp": 0.01003431,
      "balance_loss_clip": 1.01990259,
      "balance_loss_mlp": 1.00179791,
      "epoch": 0.3537545842602056,
      "flos": 65991066370560.0,
      "grad_norm": 1.2590108997254954,
      "language_loss": 0.64589202,
      "learning_rate": 2.997750460407142e-06,
      "loss": 0.66688108,
      "num_input_tokens_seen": 63209765,
      "step": 2942,
      "time_per_iteration": 3.225189208984375
    },
    {
      "auxiliary_loss_clip": 0.01207855,
      "auxiliary_loss_mlp": 0.01040394,
      "balance_loss_clip": 1.06147325,
      "balance_loss_mlp": 1.03070188,
      "epoch": 0.35387482715084473,
      "flos": 18436897526400.0,
      "grad_norm": 3.758606014960876,
      "language_loss": 0.70148778,
      "learning_rate": 2.997075270414501e-06,
      "loss": 0.72397029,
      "num_input_tokens_seen": 63226980,
      "step": 2943,
      "time_per_iteration": 2.7274932861328125
    },
    {
      "auxiliary_loss_clip": 0.01071345,
      "auxiliary_loss_mlp": 0.01004526,
      "balance_loss_clip": 1.0190084,
      "balance_loss_mlp": 1.00282145,
      "epoch": 0.3539950700414838,
      "flos": 65588579498880.0,
      "grad_norm": 0.711802112056976,
      "language_loss": 0.57699805,
      "learning_rate": 2.9963999291663347e-06,
      "loss": 0.5977568,
      "num_input_tokens_seen": 63292760,
      "step": 2944,
      "time_per_iteration": 5.190213441848755
    },
    {
      "auxiliary_loss_clip": 0.0118458,
      "auxiliary_loss_mlp": 0.01044217,
      "balance_loss_clip": 1.05589724,
      "balance_loss_mlp": 1.03481126,
      "epoch": 0.3541153129321229,
      "flos": 20521655919360.0,
      "grad_norm": 2.5946460143498147,
      "language_loss": 0.74147427,
      "learning_rate": 2.9957244367650915e-06,
      "loss": 0.76376224,
      "num_input_tokens_seen": 63309005,
      "step": 2945,
      "time_per_iteration": 2.681328773498535
    },
    {
      "auxiliary_loss_clip": 0.01203168,
      "auxiliary_loss_mlp": 0.01039081,
      "balance_loss_clip": 1.06164253,
      "balance_loss_mlp": 1.03024209,
      "epoch": 0.354235555822762,
      "flos": 19573578391680.0,
      "grad_norm": 2.4035804719157086,
      "language_loss": 0.84151876,
      "learning_rate": 2.9950487933132425e-06,
      "loss": 0.86394131,
      "num_input_tokens_seen": 63326420,
      "step": 2946,
      "time_per_iteration": 2.685471296310425
    },
    {
      "auxiliary_loss_clip": 0.01203749,
      "auxiliary_loss_mlp": 0.01032566,
      "balance_loss_clip": 1.06024277,
      "balance_loss_mlp": 1.02304745,
      "epoch": 0.35435579871340106,
      "flos": 20777268078720.0,
      "grad_norm": 2.332754976701996,
      "language_loss": 0.71453589,
      "learning_rate": 2.994372998913283e-06,
      "loss": 0.73689908,
      "num_input_tokens_seen": 63344925,
      "step": 2947,
      "time_per_iteration": 3.583641529083252
    },
    {
      "auxiliary_loss_clip": 0.01187184,
      "auxiliary_loss_mlp": 0.010426,
      "balance_loss_clip": 1.05996597,
      "balance_loss_mlp": 1.03314114,
      "epoch": 0.35447604160404017,
      "flos": 23951807153280.0,
      "grad_norm": 2.88878855060419,
      "language_loss": 0.63138932,
      "learning_rate": 2.99369705366773e-06,
      "loss": 0.65368724,
      "num_input_tokens_seen": 63365170,
      "step": 2948,
      "time_per_iteration": 2.601062059402466
    },
    {
      "auxiliary_loss_clip": 0.0118476,
      "auxiliary_loss_mlp": 0.01027187,
      "balance_loss_clip": 1.05596912,
      "balance_loss_mlp": 1.01798403,
      "epoch": 0.3545962844946792,
      "flos": 23435662671360.0,
      "grad_norm": 2.2048824718576236,
      "language_loss": 0.82771027,
      "learning_rate": 2.9930209576791244e-06,
      "loss": 0.84982979,
      "num_input_tokens_seen": 63383645,
      "step": 2949,
      "time_per_iteration": 3.624845504760742
    },
    {
      "auxiliary_loss_clip": 0.01201377,
      "auxiliary_loss_mlp": 0.01033356,
      "balance_loss_clip": 1.05899227,
      "balance_loss_mlp": 1.02456474,
      "epoch": 0.35471652738531834,
      "flos": 22085134185600.0,
      "grad_norm": 3.4548424301710523,
      "language_loss": 0.64309216,
      "learning_rate": 2.9923447110500285e-06,
      "loss": 0.66543949,
      "num_input_tokens_seen": 63402390,
      "step": 2950,
      "time_per_iteration": 2.593778371810913
    },
    {
      "auxiliary_loss_clip": 0.01148967,
      "auxiliary_loss_mlp": 0.01030491,
      "balance_loss_clip": 1.05212951,
      "balance_loss_mlp": 1.0220511,
      "epoch": 0.35483677027595745,
      "flos": 27341881787520.0,
      "grad_norm": 1.9755597976932342,
      "language_loss": 0.75636387,
      "learning_rate": 2.9916683138830295e-06,
      "loss": 0.77815849,
      "num_input_tokens_seen": 63423055,
      "step": 2951,
      "time_per_iteration": 2.7348644733428955
    },
    {
      "auxiliary_loss_clip": 0.01163773,
      "auxiliary_loss_mlp": 0.01035287,
      "balance_loss_clip": 1.05494142,
      "balance_loss_mlp": 1.0255537,
      "epoch": 0.3549570131665965,
      "flos": 13516166678400.0,
      "grad_norm": 2.662147493377751,
      "language_loss": 0.81857324,
      "learning_rate": 2.9909917662807353e-06,
      "loss": 0.84056389,
      "num_input_tokens_seen": 63440855,
      "step": 2952,
      "time_per_iteration": 2.65332293510437
    },
    {
      "auxiliary_loss_clip": 0.01168257,
      "auxiliary_loss_mlp": 0.01029239,
      "balance_loss_clip": 1.05133164,
      "balance_loss_mlp": 1.02015531,
      "epoch": 0.3550772560572356,
      "flos": 20887549810560.0,
      "grad_norm": 2.6010966984944845,
      "language_loss": 0.69990474,
      "learning_rate": 2.9903150683457783e-06,
      "loss": 0.72187972,
      "num_input_tokens_seen": 63459400,
      "step": 2953,
      "time_per_iteration": 2.6649277210235596
    },
    {
      "auxiliary_loss_clip": 0.01160697,
      "auxiliary_loss_mlp": 0.01030911,
      "balance_loss_clip": 1.0524708,
      "balance_loss_mlp": 1.02209544,
      "epoch": 0.3551974989478747,
      "flos": 20194042947840.0,
      "grad_norm": 2.4117974642088775,
      "language_loss": 0.65231878,
      "learning_rate": 2.9896382201808126e-06,
      "loss": 0.67423487,
      "num_input_tokens_seen": 63476800,
      "step": 2954,
      "time_per_iteration": 2.710305690765381
    },
    {
      "auxiliary_loss_clip": 0.01178351,
      "auxiliary_loss_mlp": 0.01030542,
      "balance_loss_clip": 1.06005335,
      "balance_loss_mlp": 1.0212791,
      "epoch": 0.3553177418385138,
      "flos": 19828831415040.0,
      "grad_norm": 2.6339066410280623,
      "language_loss": 0.81312525,
      "learning_rate": 2.988961221888516e-06,
      "loss": 0.83521414,
      "num_input_tokens_seen": 63493475,
      "step": 2955,
      "time_per_iteration": 2.6995060443878174
    },
    {
      "auxiliary_loss_clip": 0.01201932,
      "auxiliary_loss_mlp": 0.01037632,
      "balance_loss_clip": 1.05909228,
      "balance_loss_mlp": 1.02832806,
      "epoch": 0.3554379847291529,
      "flos": 14829132516480.0,
      "grad_norm": 2.9753093831780095,
      "language_loss": 0.79512161,
      "learning_rate": 2.988284073571589e-06,
      "loss": 0.81751722,
      "num_input_tokens_seen": 63509560,
      "step": 2956,
      "time_per_iteration": 2.6745214462280273
    },
    {
      "auxiliary_loss_clip": 0.01140784,
      "auxiliary_loss_mlp": 0.01033052,
      "balance_loss_clip": 1.05101585,
      "balance_loss_mlp": 1.02364612,
      "epoch": 0.355558227619792,
      "flos": 20485350247680.0,
      "grad_norm": 3.78019737725807,
      "language_loss": 0.73478353,
      "learning_rate": 2.9876067753327528e-06,
      "loss": 0.75652188,
      "num_input_tokens_seen": 63527290,
      "step": 2957,
      "time_per_iteration": 2.7401745319366455
    },
    {
      "auxiliary_loss_clip": 0.01201865,
      "auxiliary_loss_mlp": 0.01037418,
      "balance_loss_clip": 1.05783296,
      "balance_loss_mlp": 1.02741623,
      "epoch": 0.35567847051043106,
      "flos": 37663613256960.0,
      "grad_norm": 2.1227695535789164,
      "language_loss": 0.8055988,
      "learning_rate": 2.986929327274754e-06,
      "loss": 0.8279916,
      "num_input_tokens_seen": 63547870,
      "step": 2958,
      "time_per_iteration": 2.781376600265503
    },
    {
      "auxiliary_loss_clip": 0.01177574,
      "auxiliary_loss_mlp": 0.0103581,
      "balance_loss_clip": 1.0558238,
      "balance_loss_mlp": 1.0269109,
      "epoch": 0.35579871340107017,
      "flos": 26943058103040.0,
      "grad_norm": 5.518051475913736,
      "language_loss": 0.79277468,
      "learning_rate": 2.9862517295003617e-06,
      "loss": 0.8149085,
      "num_input_tokens_seen": 63568285,
      "step": 2959,
      "time_per_iteration": 2.6797144412994385
    },
    {
      "auxiliary_loss_clip": 0.01174426,
      "auxiliary_loss_mlp": 0.01032356,
      "balance_loss_clip": 1.05683947,
      "balance_loss_mlp": 1.02253366,
      "epoch": 0.3559189562917093,
      "flos": 28293335193600.0,
      "grad_norm": 1.615599362086459,
      "language_loss": 0.7287038,
      "learning_rate": 2.9855739821123654e-06,
      "loss": 0.75077164,
      "num_input_tokens_seen": 63589865,
      "step": 2960,
      "time_per_iteration": 2.749600410461426
    },
    {
      "auxiliary_loss_clip": 0.01205097,
      "auxiliary_loss_mlp": 0.01035662,
      "balance_loss_clip": 1.06207943,
      "balance_loss_mlp": 1.02641118,
      "epoch": 0.35603919918234833,
      "flos": 25664063552640.0,
      "grad_norm": 2.21927817446453,
      "language_loss": 0.82538718,
      "learning_rate": 2.98489608521358e-06,
      "loss": 0.84779477,
      "num_input_tokens_seen": 63609805,
      "step": 2961,
      "time_per_iteration": 2.68131947517395
    },
    {
      "auxiliary_loss_clip": 0.01196478,
      "auxiliary_loss_mlp": 0.01042157,
      "balance_loss_clip": 1.06187487,
      "balance_loss_mlp": 1.03243566,
      "epoch": 0.35615944207298744,
      "flos": 23000856537600.0,
      "grad_norm": 3.097788614868147,
      "language_loss": 0.80148023,
      "learning_rate": 2.9842180389068425e-06,
      "loss": 0.82386661,
      "num_input_tokens_seen": 63627115,
      "step": 2962,
      "time_per_iteration": 2.6862173080444336
    },
    {
      "auxiliary_loss_clip": 0.01077693,
      "auxiliary_loss_mlp": 0.01014952,
      "balance_loss_clip": 1.02067423,
      "balance_loss_mlp": 1.01319981,
      "epoch": 0.35627968496362655,
      "flos": 68251283723520.0,
      "grad_norm": 0.7679872158707403,
      "language_loss": 0.59171003,
      "learning_rate": 2.98353984329501e-06,
      "loss": 0.61263651,
      "num_input_tokens_seen": 63691460,
      "step": 2963,
      "time_per_iteration": 3.252253532409668
    },
    {
      "auxiliary_loss_clip": 0.01205512,
      "auxiliary_loss_mlp": 0.01046805,
      "balance_loss_clip": 1.06106746,
      "balance_loss_mlp": 1.03628445,
      "epoch": 0.3563999278542656,
      "flos": 22641714403200.0,
      "grad_norm": 4.118885325351382,
      "language_loss": 0.70445657,
      "learning_rate": 2.982861498480965e-06,
      "loss": 0.72697973,
      "num_input_tokens_seen": 63713840,
      "step": 2964,
      "time_per_iteration": 2.6512863636016846
    },
    {
      "auxiliary_loss_clip": 0.01202467,
      "auxiliary_loss_mlp": 0.01038199,
      "balance_loss_clip": 1.06057894,
      "balance_loss_mlp": 1.02900207,
      "epoch": 0.3565201707449047,
      "flos": 25952533678080.0,
      "grad_norm": 2.51232071509396,
      "language_loss": 0.82848883,
      "learning_rate": 2.9821830045676122e-06,
      "loss": 0.85089552,
      "num_input_tokens_seen": 63733540,
      "step": 2965,
      "time_per_iteration": 2.701486110687256
    },
    {
      "auxiliary_loss_clip": 0.01205621,
      "auxiliary_loss_mlp": 0.01043704,
      "balance_loss_clip": 1.06232476,
      "balance_loss_mlp": 1.03453696,
      "epoch": 0.3566404136355438,
      "flos": 28475725478400.0,
      "grad_norm": 3.5832318326816806,
      "language_loss": 0.73136908,
      "learning_rate": 2.9815043616578793e-06,
      "loss": 0.75386226,
      "num_input_tokens_seen": 63754335,
      "step": 2966,
      "time_per_iteration": 2.7008044719696045
    },
    {
      "auxiliary_loss_clip": 0.01208467,
      "auxiliary_loss_mlp": 0.01034296,
      "balance_loss_clip": 1.06399083,
      "balance_loss_mlp": 1.02533793,
      "epoch": 0.3567606565261829,
      "flos": 38363117690880.0,
      "grad_norm": 11.902723692350305,
      "language_loss": 0.77114308,
      "learning_rate": 2.9808255698547145e-06,
      "loss": 0.79357076,
      "num_input_tokens_seen": 63777135,
      "step": 2967,
      "time_per_iteration": 2.836261749267578
    },
    {
      "auxiliary_loss_clip": 0.01150175,
      "auxiliary_loss_mlp": 0.01036639,
      "balance_loss_clip": 1.05349874,
      "balance_loss_mlp": 1.0272572,
      "epoch": 0.356880899416822,
      "flos": 21981029592960.0,
      "grad_norm": 2.0451870386669118,
      "language_loss": 0.79929239,
      "learning_rate": 2.9801466292610913e-06,
      "loss": 0.82116055,
      "num_input_tokens_seen": 63797020,
      "step": 2968,
      "time_per_iteration": 2.7583670616149902
    },
    {
      "auxiliary_loss_clip": 0.0116824,
      "auxiliary_loss_mlp": 0.01037436,
      "balance_loss_clip": 1.05729461,
      "balance_loss_mlp": 1.02924025,
      "epoch": 0.35700114230746105,
      "flos": 18989132198400.0,
      "grad_norm": 2.8678659840933447,
      "language_loss": 0.81078631,
      "learning_rate": 2.979467539980003e-06,
      "loss": 0.83284307,
      "num_input_tokens_seen": 63813810,
      "step": 2969,
      "time_per_iteration": 2.7203781604766846
    },
    {
      "auxiliary_loss_clip": 0.01204899,
      "auxiliary_loss_mlp": 0.01038323,
      "balance_loss_clip": 1.0611608,
      "balance_loss_mlp": 1.02882802,
      "epoch": 0.35712138519810016,
      "flos": 19756112330880.0,
      "grad_norm": 2.1825807135630586,
      "language_loss": 0.76950765,
      "learning_rate": 2.978788302114468e-06,
      "loss": 0.79193985,
      "num_input_tokens_seen": 63830925,
      "step": 2970,
      "time_per_iteration": 3.520707607269287
    },
    {
      "auxiliary_loss_clip": 0.01191116,
      "auxiliary_loss_mlp": 0.01033662,
      "balance_loss_clip": 1.05918109,
      "balance_loss_mlp": 1.0251025,
      "epoch": 0.35724162808873927,
      "flos": 35183012008320.0,
      "grad_norm": 3.49469640401563,
      "language_loss": 0.83489645,
      "learning_rate": 2.9781089157675255e-06,
      "loss": 0.85714424,
      "num_input_tokens_seen": 63849385,
      "step": 2971,
      "time_per_iteration": 3.7156407833099365
    },
    {
      "auxiliary_loss_clip": 0.01204778,
      "auxiliary_loss_mlp": 0.01038565,
      "balance_loss_clip": 1.06259537,
      "balance_loss_mlp": 1.02952254,
      "epoch": 0.3573618709793783,
      "flos": 25556726736000.0,
      "grad_norm": 1.7683312413465369,
      "language_loss": 0.88541645,
      "learning_rate": 2.977429381042238e-06,
      "loss": 0.90784991,
      "num_input_tokens_seen": 63870060,
      "step": 2972,
      "time_per_iteration": 2.676293134689331
    },
    {
      "auxiliary_loss_clip": 0.01203619,
      "auxiliary_loss_mlp": 0.01035462,
      "balance_loss_clip": 1.06339741,
      "balance_loss_mlp": 1.02754641,
      "epoch": 0.35748211387001744,
      "flos": 29132352051840.0,
      "grad_norm": 2.197748549523962,
      "language_loss": 0.8914541,
      "learning_rate": 2.9767496980416913e-06,
      "loss": 0.91384488,
      "num_input_tokens_seen": 63889355,
      "step": 2973,
      "time_per_iteration": 2.7192490100860596
    },
    {
      "auxiliary_loss_clip": 0.01163602,
      "auxiliary_loss_mlp": 0.01035835,
      "balance_loss_clip": 1.0561893,
      "balance_loss_mlp": 1.02625084,
      "epoch": 0.35760235676065655,
      "flos": 13954169122560.0,
      "grad_norm": 3.218969813616586,
      "language_loss": 0.81354856,
      "learning_rate": 2.9760698668689914e-06,
      "loss": 0.83554298,
      "num_input_tokens_seen": 63905580,
      "step": 2974,
      "time_per_iteration": 3.6555240154266357
    },
    {
      "auxiliary_loss_clip": 0.01166923,
      "auxiliary_loss_mlp": 0.01035773,
      "balance_loss_clip": 1.05231023,
      "balance_loss_mlp": 1.02600932,
      "epoch": 0.3577225996512956,
      "flos": 44018688977280.0,
      "grad_norm": 1.8338655570900289,
      "language_loss": 0.71470708,
      "learning_rate": 2.975389887627269e-06,
      "loss": 0.73673403,
      "num_input_tokens_seen": 63928180,
      "step": 2975,
      "time_per_iteration": 2.9098589420318604
    },
    {
      "auxiliary_loss_clip": 0.01200674,
      "auxiliary_loss_mlp": 0.01033051,
      "balance_loss_clip": 1.05712414,
      "balance_loss_mlp": 1.02418756,
      "epoch": 0.3578428425419347,
      "flos": 17055199013760.0,
      "grad_norm": 2.7269892282198924,
      "language_loss": 0.90001178,
      "learning_rate": 2.9747097604196764e-06,
      "loss": 0.92234904,
      "num_input_tokens_seen": 63944825,
      "step": 2976,
      "time_per_iteration": 2.574915885925293
    },
    {
      "auxiliary_loss_clip": 0.01062132,
      "auxiliary_loss_mlp": 0.01004431,
      "balance_loss_clip": 1.01972866,
      "balance_loss_mlp": 1.00248766,
      "epoch": 0.3579630854325738,
      "flos": 71676550707840.0,
      "grad_norm": 0.6782692148094343,
      "language_loss": 0.56542253,
      "learning_rate": 2.9740294853493875e-06,
      "loss": 0.58608818,
      "num_input_tokens_seen": 64016385,
      "step": 2977,
      "time_per_iteration": 3.468125343322754
    },
    {
      "auxiliary_loss_clip": 0.0119259,
      "auxiliary_loss_mlp": 0.01032784,
      "balance_loss_clip": 1.06045461,
      "balance_loss_mlp": 1.02362299,
      "epoch": 0.3580833283232129,
      "flos": 25046651652480.0,
      "grad_norm": 2.6602524198550115,
      "language_loss": 0.67426318,
      "learning_rate": 2.9733490625196008e-06,
      "loss": 0.69651693,
      "num_input_tokens_seen": 64036245,
      "step": 2978,
      "time_per_iteration": 2.646747350692749
    },
    {
      "auxiliary_loss_clip": 0.01182007,
      "auxiliary_loss_mlp": 0.01035142,
      "balance_loss_clip": 1.05807424,
      "balance_loss_mlp": 1.02614129,
      "epoch": 0.358203571213852,
      "flos": 13953127628160.0,
      "grad_norm": 4.707743121987474,
      "language_loss": 0.76231611,
      "learning_rate": 2.9726684920335353e-06,
      "loss": 0.78448761,
      "num_input_tokens_seen": 64054110,
      "step": 2979,
      "time_per_iteration": 2.7094995975494385
    },
    {
      "auxiliary_loss_clip": 0.01168503,
      "auxiliary_loss_mlp": 0.01035421,
      "balance_loss_clip": 1.054986,
      "balance_loss_mlp": 1.02494299,
      "epoch": 0.35832381410449105,
      "flos": 20302457172480.0,
      "grad_norm": 2.735150909816514,
      "language_loss": 0.81852877,
      "learning_rate": 2.971987773994432e-06,
      "loss": 0.84056807,
      "num_input_tokens_seen": 64070295,
      "step": 2980,
      "time_per_iteration": 2.765484571456909
    },
    {
      "auxiliary_loss_clip": 0.01169964,
      "auxiliary_loss_mlp": 0.01035141,
      "balance_loss_clip": 1.05858719,
      "balance_loss_mlp": 1.0258069,
      "epoch": 0.35844405699513016,
      "flos": 16983234115200.0,
      "grad_norm": 2.1564797625359007,
      "language_loss": 0.83279693,
      "learning_rate": 2.9713069085055566e-06,
      "loss": 0.85484797,
      "num_input_tokens_seen": 64088605,
      "step": 2981,
      "time_per_iteration": 2.6803300380706787
    },
    {
      "auxiliary_loss_clip": 0.0120382,
      "auxiliary_loss_mlp": 0.01034035,
      "balance_loss_clip": 1.0612092,
      "balance_loss_mlp": 1.0248971,
      "epoch": 0.35856429988576927,
      "flos": 23216858974080.0,
      "grad_norm": 1.816250338927545,
      "language_loss": 0.79238683,
      "learning_rate": 2.9706258956701958e-06,
      "loss": 0.81476533,
      "num_input_tokens_seen": 64108595,
      "step": 2982,
      "time_per_iteration": 2.682177782058716
    },
    {
      "auxiliary_loss_clip": 0.01180686,
      "auxiliary_loss_mlp": 0.01032801,
      "balance_loss_clip": 1.05595636,
      "balance_loss_mlp": 1.02337158,
      "epoch": 0.3586845427764083,
      "flos": 23034576430080.0,
      "grad_norm": 2.410255378883406,
      "language_loss": 0.77394712,
      "learning_rate": 2.9699447355916575e-06,
      "loss": 0.79608196,
      "num_input_tokens_seen": 64127405,
      "step": 2983,
      "time_per_iteration": 2.6532509326934814
    },
    {
      "auxiliary_loss_clip": 0.01201646,
      "auxiliary_loss_mlp": 0.01036208,
      "balance_loss_clip": 1.05964756,
      "balance_loss_mlp": 1.02712381,
      "epoch": 0.35880478566704743,
      "flos": 20010682995840.0,
      "grad_norm": 2.0231231299030066,
      "language_loss": 0.73829079,
      "learning_rate": 2.969263428373275e-06,
      "loss": 0.76066935,
      "num_input_tokens_seen": 64145755,
      "step": 2984,
      "time_per_iteration": 2.6212382316589355
    },
    {
      "auxiliary_loss_clip": 0.01204996,
      "auxiliary_loss_mlp": 0.01036393,
      "balance_loss_clip": 1.06279373,
      "balance_loss_mlp": 1.02696955,
      "epoch": 0.35892502855768654,
      "flos": 13699095667200.0,
      "grad_norm": 3.0069015448976035,
      "language_loss": 0.79589999,
      "learning_rate": 2.9685819741184007e-06,
      "loss": 0.8183139,
      "num_input_tokens_seen": 64164195,
      "step": 2985,
      "time_per_iteration": 2.6339056491851807
    },
    {
      "auxiliary_loss_clip": 0.01186591,
      "auxiliary_loss_mlp": 0.01033652,
      "balance_loss_clip": 1.05901718,
      "balance_loss_mlp": 1.02449131,
      "epoch": 0.3590452714483256,
      "flos": 18114096977280.0,
      "grad_norm": 3.95616526449339,
      "language_loss": 0.69001418,
      "learning_rate": 2.967900372930411e-06,
      "loss": 0.71221656,
      "num_input_tokens_seen": 64182705,
      "step": 2986,
      "time_per_iteration": 2.6917855739593506
    },
    {
      "auxiliary_loss_clip": 0.01195458,
      "auxiliary_loss_mlp": 0.010404,
      "balance_loss_clip": 1.06262827,
      "balance_loss_mlp": 1.03086281,
      "epoch": 0.3591655143389647,
      "flos": 17749352321280.0,
      "grad_norm": 2.4875225527792826,
      "language_loss": 0.79305947,
      "learning_rate": 2.9672186249127046e-06,
      "loss": 0.815418,
      "num_input_tokens_seen": 64202170,
      "step": 2987,
      "time_per_iteration": 2.629960060119629
    },
    {
      "auxiliary_loss_clip": 0.01203781,
      "auxiliary_loss_mlp": 0.01034876,
      "balance_loss_clip": 1.06354165,
      "balance_loss_mlp": 1.02542901,
      "epoch": 0.3592857572296038,
      "flos": 25224409082880.0,
      "grad_norm": 2.878857150927443,
      "language_loss": 0.78987455,
      "learning_rate": 2.9665367301687014e-06,
      "loss": 0.8122611,
      "num_input_tokens_seen": 64220415,
      "step": 2988,
      "time_per_iteration": 2.6843903064727783
    },
    {
      "auxiliary_loss_clip": 0.01129168,
      "auxiliary_loss_mlp": 0.01037866,
      "balance_loss_clip": 1.05068719,
      "balance_loss_mlp": 1.0280726,
      "epoch": 0.3594060001202429,
      "flos": 29384408764800.0,
      "grad_norm": 2.811061785554603,
      "language_loss": 0.77051365,
      "learning_rate": 2.965854688801845e-06,
      "loss": 0.79218394,
      "num_input_tokens_seen": 64242475,
      "step": 2989,
      "time_per_iteration": 2.8463308811187744
    },
    {
      "auxiliary_loss_clip": 0.01182454,
      "auxiliary_loss_mlp": 0.01033923,
      "balance_loss_clip": 1.05720425,
      "balance_loss_mlp": 1.02472568,
      "epoch": 0.359526243010882,
      "flos": 17052900543360.0,
      "grad_norm": 2.082877188936741,
      "language_loss": 0.76467204,
      "learning_rate": 2.9651725009156005e-06,
      "loss": 0.78683579,
      "num_input_tokens_seen": 64260220,
      "step": 2990,
      "time_per_iteration": 2.630620241165161
    },
    {
      "auxiliary_loss_clip": 0.01204424,
      "auxiliary_loss_mlp": 0.01035661,
      "balance_loss_clip": 1.0592711,
      "balance_loss_mlp": 1.02682734,
      "epoch": 0.3596464859015211,
      "flos": 22965089569920.0,
      "grad_norm": 1.9592808622372067,
      "language_loss": 0.74532896,
      "learning_rate": 2.964490166613454e-06,
      "loss": 0.76772982,
      "num_input_tokens_seen": 64280145,
      "step": 2991,
      "time_per_iteration": 2.600118637084961
    },
    {
      "auxiliary_loss_clip": 0.01100842,
      "auxiliary_loss_mlp": 0.0100256,
      "balance_loss_clip": 1.02578712,
      "balance_loss_mlp": 1.00081909,
      "epoch": 0.35976672879216015,
      "flos": 54739462590720.0,
      "grad_norm": 0.8168567024923865,
      "language_loss": 0.57711625,
      "learning_rate": 2.963807685998917e-06,
      "loss": 0.59815037,
      "num_input_tokens_seen": 64336010,
      "step": 2992,
      "time_per_iteration": 3.0451252460479736
    },
    {
      "auxiliary_loss_clip": 0.01200038,
      "auxiliary_loss_mlp": 0.01027669,
      "balance_loss_clip": 1.05951858,
      "balance_loss_mlp": 1.01844239,
      "epoch": 0.35988697168279926,
      "flos": 43139020901760.0,
      "grad_norm": 2.062359216914623,
      "language_loss": 0.78293455,
      "learning_rate": 2.9631250591755196e-06,
      "loss": 0.8052116,
      "num_input_tokens_seen": 64358725,
      "step": 2993,
      "time_per_iteration": 2.806490421295166
    },
    {
      "auxiliary_loss_clip": 0.01173994,
      "auxiliary_loss_mlp": 0.01036788,
      "balance_loss_clip": 1.05946672,
      "balance_loss_mlp": 1.02749574,
      "epoch": 0.36000721457343837,
      "flos": 35845600239360.0,
      "grad_norm": 1.846837046379138,
      "language_loss": 0.57796144,
      "learning_rate": 2.962442286246817e-06,
      "loss": 0.60006928,
      "num_input_tokens_seen": 64381555,
      "step": 2994,
      "time_per_iteration": 2.7818071842193604
    },
    {
      "auxiliary_loss_clip": 0.01202496,
      "auxiliary_loss_mlp": 0.01030179,
      "balance_loss_clip": 1.06049347,
      "balance_loss_mlp": 1.0200882,
      "epoch": 0.3601274574640774,
      "flos": 18291100222080.0,
      "grad_norm": 3.03186294220308,
      "language_loss": 0.69847488,
      "learning_rate": 2.9617593673163853e-06,
      "loss": 0.72080165,
      "num_input_tokens_seen": 64400375,
      "step": 2995,
      "time_per_iteration": 3.59586501121521
    },
    {
      "auxiliary_loss_clip": 0.01186829,
      "auxiliary_loss_mlp": 0.01033634,
      "balance_loss_clip": 1.05998015,
      "balance_loss_mlp": 1.02470481,
      "epoch": 0.36024770035471654,
      "flos": 13333955961600.0,
      "grad_norm": 10.124814737446533,
      "language_loss": 0.77286732,
      "learning_rate": 2.9610763024878216e-06,
      "loss": 0.79507202,
      "num_input_tokens_seen": 64415880,
      "step": 2996,
      "time_per_iteration": 3.6792283058166504
    },
    {
      "auxiliary_loss_clip": 0.01187608,
      "auxiliary_loss_mlp": 0.0103399,
      "balance_loss_clip": 1.06089783,
      "balance_loss_mlp": 1.02443588,
      "epoch": 0.3603679432453556,
      "flos": 20267013427200.0,
      "grad_norm": 2.5159297645446808,
      "language_loss": 0.91602468,
      "learning_rate": 2.960393091864747e-06,
      "loss": 0.93824065,
      "num_input_tokens_seen": 64434260,
      "step": 2997,
      "time_per_iteration": 2.641275644302368
    },
    {
      "auxiliary_loss_clip": 0.01193635,
      "auxiliary_loss_mlp": 0.01039032,
      "balance_loss_clip": 1.06077468,
      "balance_loss_mlp": 1.0287981,
      "epoch": 0.3604881861359947,
      "flos": 22451135817600.0,
      "grad_norm": 2.4559362154298983,
      "language_loss": 0.74926031,
      "learning_rate": 2.959709735550804e-06,
      "loss": 0.77158695,
      "num_input_tokens_seen": 64453855,
      "step": 2998,
      "time_per_iteration": 2.7189605236053467
    },
    {
      "auxiliary_loss_clip": 0.01206194,
      "auxiliary_loss_mlp": 0.010316,
      "balance_loss_clip": 1.06251383,
      "balance_loss_mlp": 1.02223039,
      "epoch": 0.3606084290266338,
      "flos": 22054251467520.0,
      "grad_norm": 2.782221181417371,
      "language_loss": 0.7586903,
      "learning_rate": 2.9590262336496575e-06,
      "loss": 0.78106821,
      "num_input_tokens_seen": 64473585,
      "step": 2999,
      "time_per_iteration": 2.8163516521453857
    },
    {
      "auxiliary_loss_clip": 0.01166916,
      "auxiliary_loss_mlp": 0.01037916,
      "balance_loss_clip": 1.05805755,
      "balance_loss_mlp": 1.02780747,
      "epoch": 0.36072867191727287,
      "flos": 15632921111040.0,
      "grad_norm": 2.9352016584138942,
      "language_loss": 0.8589319,
      "learning_rate": 2.9583425862649936e-06,
      "loss": 0.88098025,
      "num_input_tokens_seen": 64491720,
      "step": 3000,
      "time_per_iteration": 3.6217198371887207
    },
    {
      "auxiliary_loss_clip": 0.01192901,
      "auxiliary_loss_mlp": 0.01039298,
      "balance_loss_clip": 1.06062889,
      "balance_loss_mlp": 1.02931452,
      "epoch": 0.360848914807912,
      "flos": 19677000625920.0,
      "grad_norm": 4.220574561615475,
      "language_loss": 0.74309707,
      "learning_rate": 2.9576587935005215e-06,
      "loss": 0.76541913,
      "num_input_tokens_seen": 64509800,
      "step": 3001,
      "time_per_iteration": 3.586101770401001
    },
    {
      "auxiliary_loss_clip": 0.01207187,
      "auxiliary_loss_mlp": 0.01036466,
      "balance_loss_clip": 1.06239748,
      "balance_loss_mlp": 1.02622557,
      "epoch": 0.3609691576985511,
      "flos": 18877808972160.0,
      "grad_norm": 3.1173890620197784,
      "language_loss": 0.7239939,
      "learning_rate": 2.9569748554599713e-06,
      "loss": 0.74643046,
      "num_input_tokens_seen": 64525410,
      "step": 3002,
      "time_per_iteration": 2.5700764656066895
    },
    {
      "auxiliary_loss_clip": 0.01203849,
      "auxiliary_loss_mlp": 0.01036556,
      "balance_loss_clip": 1.06370175,
      "balance_loss_mlp": 1.02773786,
      "epoch": 0.36108940058919015,
      "flos": 42224088648960.0,
      "grad_norm": 2.0328581098714813,
      "language_loss": 0.73661172,
      "learning_rate": 2.956290772247097e-06,
      "loss": 0.7590158,
      "num_input_tokens_seen": 64544085,
      "step": 3003,
      "time_per_iteration": 2.7777059078216553
    },
    {
      "auxiliary_loss_clip": 0.01205024,
      "auxiliary_loss_mlp": 0.0103092,
      "balance_loss_clip": 1.06262314,
      "balance_loss_mlp": 1.0210439,
      "epoch": 0.36120964347982926,
      "flos": 23185150243200.0,
      "grad_norm": 2.6665249043426233,
      "language_loss": 0.73601866,
      "learning_rate": 2.9556065439656724e-06,
      "loss": 0.75837815,
      "num_input_tokens_seen": 64563135,
      "step": 3004,
      "time_per_iteration": 2.6278414726257324
    },
    {
      "auxiliary_loss_clip": 0.01174036,
      "auxiliary_loss_mlp": 0.01038584,
      "balance_loss_clip": 1.05747557,
      "balance_loss_mlp": 1.03004813,
      "epoch": 0.36132988637046837,
      "flos": 18113055482880.0,
      "grad_norm": 1.919579742486563,
      "language_loss": 0.81785762,
      "learning_rate": 2.9549221707194952e-06,
      "loss": 0.83998382,
      "num_input_tokens_seen": 64581985,
      "step": 3005,
      "time_per_iteration": 2.7783570289611816
    },
    {
      "auxiliary_loss_clip": 0.012074,
      "auxiliary_loss_mlp": 0.01031208,
      "balance_loss_clip": 1.0628705,
      "balance_loss_mlp": 1.02210617,
      "epoch": 0.3614501292611074,
      "flos": 27813101333760.0,
      "grad_norm": 2.2646319689711105,
      "language_loss": 0.73026347,
      "learning_rate": 2.954237652612384e-06,
      "loss": 0.75264955,
      "num_input_tokens_seen": 64601035,
      "step": 3006,
      "time_per_iteration": 2.6067559719085693
    },
    {
      "auxiliary_loss_clip": 0.01176422,
      "auxiliary_loss_mlp": 0.01036425,
      "balance_loss_clip": 1.05382228,
      "balance_loss_mlp": 1.02788341,
      "epoch": 0.36157037215174653,
      "flos": 22634926732800.0,
      "grad_norm": 2.2751669435345545,
      "language_loss": 0.84515065,
      "learning_rate": 2.9535529897481796e-06,
      "loss": 0.86727905,
      "num_input_tokens_seen": 64618580,
      "step": 3007,
      "time_per_iteration": 2.73610520362854
    },
    {
      "auxiliary_loss_clip": 0.01191155,
      "auxiliary_loss_mlp": 0.01036166,
      "balance_loss_clip": 1.05692303,
      "balance_loss_mlp": 1.02631927,
      "epoch": 0.36169061504238564,
      "flos": 12600839376000.0,
      "grad_norm": 2.2841478922248255,
      "language_loss": 0.77326751,
      "learning_rate": 2.9528681822307446e-06,
      "loss": 0.79554069,
      "num_input_tokens_seen": 64635430,
      "step": 3008,
      "time_per_iteration": 2.6347098350524902
    },
    {
      "auxiliary_loss_clip": 0.01171444,
      "auxiliary_loss_mlp": 0.01028978,
      "balance_loss_clip": 1.05599785,
      "balance_loss_mlp": 1.02013326,
      "epoch": 0.3618108579330247,
      "flos": 26684644682880.0,
      "grad_norm": 2.4720608649185407,
      "language_loss": 0.8260808,
      "learning_rate": 2.952183230163964e-06,
      "loss": 0.84808505,
      "num_input_tokens_seen": 64655005,
      "step": 3009,
      "time_per_iteration": 2.720327138900757
    },
    {
      "auxiliary_loss_clip": 0.01201404,
      "auxiliary_loss_mlp": 0.01034876,
      "balance_loss_clip": 1.05955696,
      "balance_loss_mlp": 1.02556574,
      "epoch": 0.3619311008236638,
      "flos": 22817029708800.0,
      "grad_norm": 2.5115745552278157,
      "language_loss": 0.72952336,
      "learning_rate": 2.9514981336517448e-06,
      "loss": 0.75188613,
      "num_input_tokens_seen": 64674775,
      "step": 3010,
      "time_per_iteration": 2.6505861282348633
    },
    {
      "auxiliary_loss_clip": 0.01164646,
      "auxiliary_loss_mlp": 0.01039685,
      "balance_loss_clip": 1.05664515,
      "balance_loss_mlp": 1.02993321,
      "epoch": 0.36205134371430286,
      "flos": 25919603884800.0,
      "grad_norm": 3.9796629863764994,
      "language_loss": 0.81601191,
      "learning_rate": 2.950812892798015e-06,
      "loss": 0.83805519,
      "num_input_tokens_seen": 64695670,
      "step": 3011,
      "time_per_iteration": 2.8298327922821045
    },
    {
      "auxiliary_loss_clip": 0.01207031,
      "auxiliary_loss_mlp": 0.01037249,
      "balance_loss_clip": 1.06479371,
      "balance_loss_mlp": 1.02885687,
      "epoch": 0.362171586604942,
      "flos": 26139592730880.0,
      "grad_norm": 1.950018981146389,
      "language_loss": 0.87314785,
      "learning_rate": 2.9501275077067256e-06,
      "loss": 0.89559066,
      "num_input_tokens_seen": 64716290,
      "step": 3012,
      "time_per_iteration": 2.667379140853882
    },
    {
      "auxiliary_loss_clip": 0.01203578,
      "auxiliary_loss_mlp": 0.01033724,
      "balance_loss_clip": 1.06233692,
      "balance_loss_mlp": 1.02527761,
      "epoch": 0.3622918294955811,
      "flos": 28074208273920.0,
      "grad_norm": 1.9241983322166472,
      "language_loss": 0.88565344,
      "learning_rate": 2.949441978481848e-06,
      "loss": 0.90802646,
      "num_input_tokens_seen": 64737190,
      "step": 3013,
      "time_per_iteration": 2.7146155834198
    },
    {
      "auxiliary_loss_clip": 0.01205021,
      "auxiliary_loss_mlp": 0.01031576,
      "balance_loss_clip": 1.06083548,
      "balance_loss_mlp": 1.02218866,
      "epoch": 0.36241207238622014,
      "flos": 19828005402240.0,
      "grad_norm": 3.587963081440332,
      "language_loss": 0.80218172,
      "learning_rate": 2.9487563052273778e-06,
      "loss": 0.82454765,
      "num_input_tokens_seen": 64753950,
      "step": 3014,
      "time_per_iteration": 2.6600582599639893
    },
    {
      "auxiliary_loss_clip": 0.01202916,
      "auxiliary_loss_mlp": 0.01034577,
      "balance_loss_clip": 1.06084633,
      "balance_loss_mlp": 1.0253799,
      "epoch": 0.36253231527685925,
      "flos": 21397158017280.0,
      "grad_norm": 3.0634177326133862,
      "language_loss": 0.8601951,
      "learning_rate": 2.94807048804733e-06,
      "loss": 0.88257003,
      "num_input_tokens_seen": 64773570,
      "step": 3015,
      "time_per_iteration": 2.6295390129089355
    },
    {
      "auxiliary_loss_clip": 0.01195305,
      "auxiliary_loss_mlp": 0.01040811,
      "balance_loss_clip": 1.06023645,
      "balance_loss_mlp": 1.03122628,
      "epoch": 0.36265255816749836,
      "flos": 18362885552640.0,
      "grad_norm": 1.8421009372677901,
      "language_loss": 0.90329599,
      "learning_rate": 2.9473845270457434e-06,
      "loss": 0.92565715,
      "num_input_tokens_seen": 64790385,
      "step": 3016,
      "time_per_iteration": 2.7594590187072754
    },
    {
      "auxiliary_loss_clip": 0.01162447,
      "auxiliary_loss_mlp": 0.01036048,
      "balance_loss_clip": 1.05187666,
      "balance_loss_mlp": 1.02674377,
      "epoch": 0.3627728010581374,
      "flos": 18660046769280.0,
      "grad_norm": 2.4700650674043665,
      "language_loss": 0.70709306,
      "learning_rate": 2.946698422326677e-06,
      "loss": 0.72907805,
      "num_input_tokens_seen": 64807845,
      "step": 3017,
      "time_per_iteration": 2.7349421977996826
    },
    {
      "auxiliary_loss_clip": 0.01203394,
      "auxiliary_loss_mlp": 0.01030565,
      "balance_loss_clip": 1.05923235,
      "balance_loss_mlp": 1.02120709,
      "epoch": 0.36289304394877653,
      "flos": 27524272072320.0,
      "grad_norm": 3.554050894403776,
      "language_loss": 0.79921377,
      "learning_rate": 2.946012173994213e-06,
      "loss": 0.82155335,
      "num_input_tokens_seen": 64827630,
      "step": 3018,
      "time_per_iteration": 2.76169753074646
    },
    {
      "auxiliary_loss_clip": 0.0119114,
      "auxiliary_loss_mlp": 0.01032893,
      "balance_loss_clip": 1.06028283,
      "balance_loss_mlp": 1.02376199,
      "epoch": 0.36301328683941564,
      "flos": 34533244932480.0,
      "grad_norm": 1.670312118586647,
      "language_loss": 0.67810476,
      "learning_rate": 2.945325782152454e-06,
      "loss": 0.7003451,
      "num_input_tokens_seen": 64850665,
      "step": 3019,
      "time_per_iteration": 2.779832363128662
    },
    {
      "auxiliary_loss_clip": 0.01162791,
      "auxiliary_loss_mlp": 0.01034358,
      "balance_loss_clip": 1.05771744,
      "balance_loss_mlp": 1.02566719,
      "epoch": 0.3631335297300547,
      "flos": 19025976574080.0,
      "grad_norm": 2.4463859801245,
      "language_loss": 0.78984451,
      "learning_rate": 2.9446392469055257e-06,
      "loss": 0.81181598,
      "num_input_tokens_seen": 64868700,
      "step": 3020,
      "time_per_iteration": 2.7176573276519775
    },
    {
      "auxiliary_loss_clip": 0.01166799,
      "auxiliary_loss_mlp": 0.01036588,
      "balance_loss_clip": 1.05537271,
      "balance_loss_mlp": 1.02791584,
      "epoch": 0.3632537726206938,
      "flos": 19536769929600.0,
      "grad_norm": 1.8798672366690905,
      "language_loss": 0.8037411,
      "learning_rate": 2.9439525683575745e-06,
      "loss": 0.82577497,
      "num_input_tokens_seen": 64887620,
      "step": 3021,
      "time_per_iteration": 2.666577100753784
    },
    {
      "auxiliary_loss_clip": 0.01192118,
      "auxiliary_loss_mlp": 0.01042383,
      "balance_loss_clip": 1.06007564,
      "balance_loss_mlp": 1.03203607,
      "epoch": 0.3633740155113329,
      "flos": 21068611292160.0,
      "grad_norm": 10.17708707959518,
      "language_loss": 0.75453687,
      "learning_rate": 2.9432657466127694e-06,
      "loss": 0.77688193,
      "num_input_tokens_seen": 64907190,
      "step": 3022,
      "time_per_iteration": 4.511340856552124
    },
    {
      "auxiliary_loss_clip": 0.01209012,
      "auxiliary_loss_mlp": 0.0104251,
      "balance_loss_clip": 1.06531882,
      "balance_loss_mlp": 1.03262794,
      "epoch": 0.36349425840197197,
      "flos": 20298722158080.0,
      "grad_norm": 1.948235571532719,
      "language_loss": 0.76715851,
      "learning_rate": 2.9425787817753007e-06,
      "loss": 0.78967375,
      "num_input_tokens_seen": 64925850,
      "step": 3023,
      "time_per_iteration": 2.6496779918670654
    },
    {
      "auxiliary_loss_clip": 0.01203108,
      "auxiliary_loss_mlp": 0.01033842,
      "balance_loss_clip": 1.05906224,
      "balance_loss_mlp": 1.02459788,
      "epoch": 0.3636145012926111,
      "flos": 29716762331520.0,
      "grad_norm": 2.5396227955738753,
      "language_loss": 0.71627361,
      "learning_rate": 2.94189167394938e-06,
      "loss": 0.73864317,
      "num_input_tokens_seen": 64948285,
      "step": 3024,
      "time_per_iteration": 2.7011098861694336
    },
    {
      "auxiliary_loss_clip": 0.01197275,
      "auxiliary_loss_mlp": 0.01042413,
      "balance_loss_clip": 1.06462955,
      "balance_loss_mlp": 1.03278077,
      "epoch": 0.3637347441832502,
      "flos": 21431847576960.0,
      "grad_norm": 9.36237129245091,
      "language_loss": 0.81737101,
      "learning_rate": 2.941204423239241e-06,
      "loss": 0.83976781,
      "num_input_tokens_seen": 64967160,
      "step": 3025,
      "time_per_iteration": 2.6745808124542236
    },
    {
      "auxiliary_loss_clip": 0.01187124,
      "auxiliary_loss_mlp": 0.01036707,
      "balance_loss_clip": 1.05968547,
      "balance_loss_mlp": 1.02771878,
      "epoch": 0.36385498707388925,
      "flos": 29533941083520.0,
      "grad_norm": 6.002639827544448,
      "language_loss": 0.75850511,
      "learning_rate": 2.9405170297491395e-06,
      "loss": 0.78074348,
      "num_input_tokens_seen": 64987155,
      "step": 3026,
      "time_per_iteration": 3.7106335163116455
    },
    {
      "auxiliary_loss_clip": 0.01203991,
      "auxiliary_loss_mlp": 0.0104156,
      "balance_loss_clip": 1.06179917,
      "balance_loss_mlp": 1.0315876,
      "epoch": 0.36397522996452836,
      "flos": 22236569925120.0,
      "grad_norm": 15.041588130471942,
      "language_loss": 0.80443406,
      "learning_rate": 2.939829493583353e-06,
      "loss": 0.82688963,
      "num_input_tokens_seen": 65003800,
      "step": 3027,
      "time_per_iteration": 3.596261978149414
    },
    {
      "auxiliary_loss_clip": 0.01157221,
      "auxiliary_loss_mlp": 0.01027628,
      "balance_loss_clip": 1.05138278,
      "balance_loss_mlp": 1.01860428,
      "epoch": 0.3640954728551674,
      "flos": 21506505995520.0,
      "grad_norm": 2.530286849919935,
      "language_loss": 0.83732402,
      "learning_rate": 2.939141814846179e-06,
      "loss": 0.85917258,
      "num_input_tokens_seen": 65021215,
      "step": 3028,
      "time_per_iteration": 2.722818374633789
    },
    {
      "auxiliary_loss_clip": 0.01202911,
      "auxiliary_loss_mlp": 0.01035628,
      "balance_loss_clip": 1.06108212,
      "balance_loss_mlp": 1.02632928,
      "epoch": 0.3642157157458065,
      "flos": 17712867081600.0,
      "grad_norm": 1.7311182053759804,
      "language_loss": 0.82881439,
      "learning_rate": 2.938453993641938e-06,
      "loss": 0.85119987,
      "num_input_tokens_seen": 65039590,
      "step": 3029,
      "time_per_iteration": 2.710679769515991
    },
    {
      "auxiliary_loss_clip": 0.01191393,
      "auxiliary_loss_mlp": 0.01036464,
      "balance_loss_clip": 1.06390238,
      "balance_loss_mlp": 1.02675438,
      "epoch": 0.36433595863644563,
      "flos": 17639537466240.0,
      "grad_norm": 2.5673444796175637,
      "language_loss": 0.70510691,
      "learning_rate": 2.937766030074973e-06,
      "loss": 0.72738552,
      "num_input_tokens_seen": 65056845,
      "step": 3030,
      "time_per_iteration": 2.6016616821289062
    },
    {
      "auxiliary_loss_clip": 0.01170595,
      "auxiliary_loss_mlp": 0.01033305,
      "balance_loss_clip": 1.05740166,
      "balance_loss_mlp": 1.02342224,
      "epoch": 0.3644562015270847,
      "flos": 26833279161600.0,
      "grad_norm": 1.995594926632856,
      "language_loss": 0.8278656,
      "learning_rate": 2.937077924249646e-06,
      "loss": 0.84990466,
      "num_input_tokens_seen": 65079435,
      "step": 3031,
      "time_per_iteration": 2.753634214401245
    },
    {
      "auxiliary_loss_clip": 0.01205276,
      "auxiliary_loss_mlp": 0.01037472,
      "balance_loss_clip": 1.06089759,
      "balance_loss_mlp": 1.02691031,
      "epoch": 0.3645764444177238,
      "flos": 14282715847680.0,
      "grad_norm": 2.4439258797047865,
      "language_loss": 0.75532532,
      "learning_rate": 2.9363896762703443e-06,
      "loss": 0.77775282,
      "num_input_tokens_seen": 65096500,
      "step": 3032,
      "time_per_iteration": 2.6220951080322266
    },
    {
      "auxiliary_loss_clip": 0.01189687,
      "auxiliary_loss_mlp": 0.01035121,
      "balance_loss_clip": 1.0597403,
      "balance_loss_mlp": 1.02573371,
      "epoch": 0.3646966873083629,
      "flos": 20667489137280.0,
      "grad_norm": 2.0317028979223113,
      "language_loss": 0.84678012,
      "learning_rate": 2.9357012862414725e-06,
      "loss": 0.86902821,
      "num_input_tokens_seen": 65115860,
      "step": 3033,
      "time_per_iteration": 2.679145336151123
    },
    {
      "auxiliary_loss_clip": 0.01198785,
      "auxiliary_loss_mlp": 0.01038101,
      "balance_loss_clip": 1.05918717,
      "balance_loss_mlp": 1.02836776,
      "epoch": 0.36481693019900197,
      "flos": 27782613665280.0,
      "grad_norm": 2.242607749161222,
      "language_loss": 0.71698636,
      "learning_rate": 2.9350127542674593e-06,
      "loss": 0.73935521,
      "num_input_tokens_seen": 65138070,
      "step": 3034,
      "time_per_iteration": 2.6302597522735596
    },
    {
      "auxiliary_loss_clip": 0.01190214,
      "auxiliary_loss_mlp": 0.01041162,
      "balance_loss_clip": 1.05806375,
      "balance_loss_mlp": 1.0319531,
      "epoch": 0.3649371730896411,
      "flos": 19712588025600.0,
      "grad_norm": 1.9368593468968893,
      "language_loss": 0.76691079,
      "learning_rate": 2.934324080452755e-06,
      "loss": 0.78922457,
      "num_input_tokens_seen": 65155860,
      "step": 3035,
      "time_per_iteration": 2.6333882808685303
    },
    {
      "auxiliary_loss_clip": 0.01139473,
      "auxiliary_loss_mlp": 0.01033218,
      "balance_loss_clip": 1.05066276,
      "balance_loss_mlp": 1.02378297,
      "epoch": 0.3650574159802802,
      "flos": 24750496016640.0,
      "grad_norm": 1.7526065619892992,
      "language_loss": 0.7874887,
      "learning_rate": 2.9336352649018307e-06,
      "loss": 0.80921555,
      "num_input_tokens_seen": 65175930,
      "step": 3036,
      "time_per_iteration": 2.678643226623535
    },
    {
      "auxiliary_loss_clip": 0.01192264,
      "auxiliary_loss_mlp": 0.01049888,
      "balance_loss_clip": 1.06300163,
      "balance_loss_mlp": 1.03980887,
      "epoch": 0.36517765887091924,
      "flos": 32853487363200.0,
      "grad_norm": 2.420871458526115,
      "language_loss": 0.7029146,
      "learning_rate": 2.9329463077191783e-06,
      "loss": 0.72533619,
      "num_input_tokens_seen": 65199305,
      "step": 3037,
      "time_per_iteration": 2.76062273979187
    },
    {
      "auxiliary_loss_clip": 0.01189575,
      "auxiliary_loss_mlp": 0.01040631,
      "balance_loss_clip": 1.06160617,
      "balance_loss_mlp": 1.03120136,
      "epoch": 0.36529790176155835,
      "flos": 20120318282880.0,
      "grad_norm": 3.9216264902334634,
      "language_loss": 0.64497292,
      "learning_rate": 2.9322572090093135e-06,
      "loss": 0.66727495,
      "num_input_tokens_seen": 65218010,
      "step": 3038,
      "time_per_iteration": 2.5952773094177246
    },
    {
      "auxiliary_loss_clip": 0.01149501,
      "auxiliary_loss_mlp": 0.01033547,
      "balance_loss_clip": 1.05157661,
      "balance_loss_mlp": 1.0237422,
      "epoch": 0.36541814465219746,
      "flos": 17639573379840.0,
      "grad_norm": 4.016445283026886,
      "language_loss": 0.76766169,
      "learning_rate": 2.9315679688767713e-06,
      "loss": 0.78949219,
      "num_input_tokens_seen": 65236020,
      "step": 3039,
      "time_per_iteration": 2.7038497924804688
    },
    {
      "auxiliary_loss_clip": 0.01183545,
      "auxiliary_loss_mlp": 0.0102779,
      "balance_loss_clip": 1.0592978,
      "balance_loss_mlp": 1.01923704,
      "epoch": 0.3655383875428365,
      "flos": 22674356887680.0,
      "grad_norm": 1.685608790311072,
      "language_loss": 0.66864818,
      "learning_rate": 2.9308785874261085e-06,
      "loss": 0.69076151,
      "num_input_tokens_seen": 65256210,
      "step": 3040,
      "time_per_iteration": 2.6504855155944824
    },
    {
      "auxiliary_loss_clip": 0.01190532,
      "auxiliary_loss_mlp": 0.01033371,
      "balance_loss_clip": 1.06189752,
      "balance_loss_mlp": 1.02389431,
      "epoch": 0.36565863043347563,
      "flos": 21981173247360.0,
      "grad_norm": 2.5057956154953143,
      "language_loss": 0.81661558,
      "learning_rate": 2.9301890647619045e-06,
      "loss": 0.83885455,
      "num_input_tokens_seen": 65275505,
      "step": 3041,
      "time_per_iteration": 2.709911584854126
    },
    {
      "auxiliary_loss_clip": 0.01177082,
      "auxiliary_loss_mlp": 0.01037946,
      "balance_loss_clip": 1.05638766,
      "balance_loss_mlp": 1.02811682,
      "epoch": 0.36577887332411474,
      "flos": 24827632473600.0,
      "grad_norm": 2.3724257689151647,
      "language_loss": 0.80389994,
      "learning_rate": 2.929499400988759e-06,
      "loss": 0.82605022,
      "num_input_tokens_seen": 65296665,
      "step": 3042,
      "time_per_iteration": 2.7667388916015625
    },
    {
      "auxiliary_loss_clip": 0.01204793,
      "auxiliary_loss_mlp": 0.0103337,
      "balance_loss_clip": 1.06080461,
      "balance_loss_mlp": 1.0236845,
      "epoch": 0.3658991162147538,
      "flos": 28293191539200.0,
      "grad_norm": 2.4117240504743944,
      "language_loss": 0.6567148,
      "learning_rate": 2.9288095962112927e-06,
      "loss": 0.67909646,
      "num_input_tokens_seen": 65317370,
      "step": 3043,
      "time_per_iteration": 2.6941545009613037
    },
    {
      "auxiliary_loss_clip": 0.01170441,
      "auxiliary_loss_mlp": 0.01035427,
      "balance_loss_clip": 1.05656612,
      "balance_loss_mlp": 1.02657557,
      "epoch": 0.3660193591053929,
      "flos": 17785550252160.0,
      "grad_norm": 3.148073377591863,
      "language_loss": 0.85340118,
      "learning_rate": 2.9281196505341503e-06,
      "loss": 0.87545991,
      "num_input_tokens_seen": 65334540,
      "step": 3044,
      "time_per_iteration": 2.6504130363464355
    },
    {
      "auxiliary_loss_clip": 0.01179011,
      "auxiliary_loss_mlp": 0.01032163,
      "balance_loss_clip": 1.0564853,
      "balance_loss_mlp": 1.02355647,
      "epoch": 0.36613960199603196,
      "flos": 10342776839040.0,
      "grad_norm": 5.3926516440645,
      "language_loss": 0.78946757,
      "learning_rate": 2.9274295640619946e-06,
      "loss": 0.81157935,
      "num_input_tokens_seen": 65351670,
      "step": 3045,
      "time_per_iteration": 2.679187536239624
    },
    {
      "auxiliary_loss_clip": 0.01137401,
      "auxiliary_loss_mlp": 0.01033258,
      "balance_loss_clip": 1.04776692,
      "balance_loss_mlp": 1.02429998,
      "epoch": 0.36625984488667107,
      "flos": 19755609540480.0,
      "grad_norm": 2.0868933946844677,
      "language_loss": 0.78729606,
      "learning_rate": 2.9267393368995103e-06,
      "loss": 0.80900264,
      "num_input_tokens_seen": 65370900,
      "step": 3046,
      "time_per_iteration": 2.6930222511291504
    },
    {
      "auxiliary_loss_clip": 0.01190365,
      "auxiliary_loss_mlp": 0.01037432,
      "balance_loss_clip": 1.05767345,
      "balance_loss_mlp": 1.02798486,
      "epoch": 0.3663800877773102,
      "flos": 17674262939520.0,
      "grad_norm": 2.4786541531925765,
      "language_loss": 0.74516964,
      "learning_rate": 2.926048969151407e-06,
      "loss": 0.76744759,
      "num_input_tokens_seen": 65388185,
      "step": 3047,
      "time_per_iteration": 2.6011900901794434
    },
    {
      "auxiliary_loss_clip": 0.01179889,
      "auxiliary_loss_mlp": 0.01035936,
      "balance_loss_clip": 1.05798149,
      "balance_loss_mlp": 1.02647042,
      "epoch": 0.36650033066794924,
      "flos": 20303606407680.0,
      "grad_norm": 1.9692511499655023,
      "language_loss": 0.68850273,
      "learning_rate": 2.92535846092241e-06,
      "loss": 0.71066093,
      "num_input_tokens_seen": 65407200,
      "step": 3048,
      "time_per_iteration": 3.6421761512756348
    },
    {
      "auxiliary_loss_clip": 0.01205319,
      "auxiliary_loss_mlp": 0.0103501,
      "balance_loss_clip": 1.06062651,
      "balance_loss_mlp": 1.02546179,
      "epoch": 0.36662057355858835,
      "flos": 24716237420160.0,
      "grad_norm": 58.45955070040644,
      "language_loss": 0.82724416,
      "learning_rate": 2.9246678123172704e-06,
      "loss": 0.84964752,
      "num_input_tokens_seen": 65427290,
      "step": 3049,
      "time_per_iteration": 3.594863176345825
    },
    {
      "auxiliary_loss_clip": 0.01204849,
      "auxiliary_loss_mlp": 0.01035646,
      "balance_loss_clip": 1.06006265,
      "balance_loss_mlp": 1.02609754,
      "epoch": 0.36674081644922746,
      "flos": 12385267902720.0,
      "grad_norm": 2.9465292945363175,
      "language_loss": 0.75075525,
      "learning_rate": 2.9239770234407596e-06,
      "loss": 0.77316022,
      "num_input_tokens_seen": 65445595,
      "step": 3050,
      "time_per_iteration": 2.59509015083313
    },
    {
      "auxiliary_loss_clip": 0.01174116,
      "auxiliary_loss_mlp": 0.01030745,
      "balance_loss_clip": 1.05955434,
      "balance_loss_mlp": 1.02147651,
      "epoch": 0.3668610593398665,
      "flos": 21105922544640.0,
      "grad_norm": 3.1068777767926608,
      "language_loss": 0.68734986,
      "learning_rate": 2.9232860943976686e-06,
      "loss": 0.70939845,
      "num_input_tokens_seen": 65466330,
      "step": 3051,
      "time_per_iteration": 2.71980357170105
    },
    {
      "auxiliary_loss_clip": 0.01190019,
      "auxiliary_loss_mlp": 0.01030992,
      "balance_loss_clip": 1.05854416,
      "balance_loss_mlp": 1.02217066,
      "epoch": 0.3669813022305056,
      "flos": 26758082039040.0,
      "grad_norm": 1.7396766104391734,
      "language_loss": 0.84255248,
      "learning_rate": 2.9225950252928115e-06,
      "loss": 0.8647626,
      "num_input_tokens_seen": 65487180,
      "step": 3052,
      "time_per_iteration": 3.578528881072998
    },
    {
      "auxiliary_loss_clip": 0.01165814,
      "auxiliary_loss_mlp": 0.01039067,
      "balance_loss_clip": 1.05528712,
      "balance_loss_mlp": 1.02880263,
      "epoch": 0.36710154512114473,
      "flos": 19099521671040.0,
      "grad_norm": 2.455053988978204,
      "language_loss": 0.81854826,
      "learning_rate": 2.9219038162310217e-06,
      "loss": 0.84059703,
      "num_input_tokens_seen": 65505380,
      "step": 3053,
      "time_per_iteration": 2.6650078296661377
    },
    {
      "auxiliary_loss_clip": 0.01177888,
      "auxiliary_loss_mlp": 0.01033887,
      "balance_loss_clip": 1.05440545,
      "balance_loss_mlp": 1.02424252,
      "epoch": 0.3672217880117838,
      "flos": 20812029465600.0,
      "grad_norm": 5.757965390421106,
      "language_loss": 0.83259284,
      "learning_rate": 2.921212467317157e-06,
      "loss": 0.85471058,
      "num_input_tokens_seen": 65524825,
      "step": 3054,
      "time_per_iteration": 3.619676351547241
    },
    {
      "auxiliary_loss_clip": 0.012039,
      "auxiliary_loss_mlp": 0.01040873,
      "balance_loss_clip": 1.05996823,
      "balance_loss_mlp": 1.03117514,
      "epoch": 0.3673420309024229,
      "flos": 13590394133760.0,
      "grad_norm": 2.3718879769124066,
      "language_loss": 0.80279523,
      "learning_rate": 2.920520978656093e-06,
      "loss": 0.825243,
      "num_input_tokens_seen": 65541790,
      "step": 3055,
      "time_per_iteration": 2.5799853801727295
    },
    {
      "auxiliary_loss_clip": 0.01199838,
      "auxiliary_loss_mlp": 0.01030189,
      "balance_loss_clip": 1.05941367,
      "balance_loss_mlp": 1.02149916,
      "epoch": 0.367462273793062,
      "flos": 28986877969920.0,
      "grad_norm": 1.9377701618655816,
      "language_loss": 0.76960987,
      "learning_rate": 2.919829350352729e-06,
      "loss": 0.79191017,
      "num_input_tokens_seen": 65563395,
      "step": 3056,
      "time_per_iteration": 2.7030646800994873
    },
    {
      "auxiliary_loss_clip": 0.01075939,
      "auxiliary_loss_mlp": 0.01001983,
      "balance_loss_clip": 1.01954889,
      "balance_loss_mlp": 1.00047517,
      "epoch": 0.36758251668370107,
      "flos": 62643148346880.0,
      "grad_norm": 0.8990228408838644,
      "language_loss": 0.59975308,
      "learning_rate": 2.919137582511983e-06,
      "loss": 0.62053233,
      "num_input_tokens_seen": 65619835,
      "step": 3057,
      "time_per_iteration": 3.1889798641204834
    },
    {
      "auxiliary_loss_clip": 0.01169414,
      "auxiliary_loss_mlp": 0.01037135,
      "balance_loss_clip": 1.06021833,
      "balance_loss_mlp": 1.02788401,
      "epoch": 0.3677027595743402,
      "flos": 12713886455040.0,
      "grad_norm": 2.913442753780036,
      "language_loss": 0.6417768,
      "learning_rate": 2.918445675238797e-06,
      "loss": 0.66384226,
      "num_input_tokens_seen": 65636760,
      "step": 3058,
      "time_per_iteration": 2.7684640884399414
    },
    {
      "auxiliary_loss_clip": 0.01203253,
      "auxiliary_loss_mlp": 0.01033591,
      "balance_loss_clip": 1.0600915,
      "balance_loss_mlp": 1.02395284,
      "epoch": 0.36782300246497923,
      "flos": 25046579825280.0,
      "grad_norm": 7.807432856704515,
      "language_loss": 0.70168763,
      "learning_rate": 2.917753628638132e-06,
      "loss": 0.72405612,
      "num_input_tokens_seen": 65657065,
      "step": 3059,
      "time_per_iteration": 2.6440842151641846
    },
    {
      "auxiliary_loss_clip": 0.01203359,
      "auxiliary_loss_mlp": 0.01043523,
      "balance_loss_clip": 1.06097031,
      "balance_loss_mlp": 1.03358078,
      "epoch": 0.36794324535561834,
      "flos": 17419512706560.0,
      "grad_norm": 2.451941931440969,
      "language_loss": 0.70517218,
      "learning_rate": 2.9170614428149716e-06,
      "loss": 0.72764099,
      "num_input_tokens_seen": 65675400,
      "step": 3060,
      "time_per_iteration": 2.6914546489715576
    },
    {
      "auxiliary_loss_clip": 0.01180584,
      "auxiliary_loss_mlp": 0.0103419,
      "balance_loss_clip": 1.06270552,
      "balance_loss_mlp": 1.02437901,
      "epoch": 0.36806348824625745,
      "flos": 24089128848000.0,
      "grad_norm": 2.4474310983943393,
      "language_loss": 0.86717284,
      "learning_rate": 2.9163691178743195e-06,
      "loss": 0.88932061,
      "num_input_tokens_seen": 65694050,
      "step": 3061,
      "time_per_iteration": 2.7261719703674316
    },
    {
      "auxiliary_loss_clip": 0.01183289,
      "auxiliary_loss_mlp": 0.01035306,
      "balance_loss_clip": 1.0584774,
      "balance_loss_mlp": 1.02594233,
      "epoch": 0.3681837311368965,
      "flos": 20521871400960.0,
      "grad_norm": 2.1479490167620576,
      "language_loss": 0.77999139,
      "learning_rate": 2.9156766539212006e-06,
      "loss": 0.80217743,
      "num_input_tokens_seen": 65711695,
      "step": 3062,
      "time_per_iteration": 2.604088306427002
    },
    {
      "auxiliary_loss_clip": 0.01174905,
      "auxiliary_loss_mlp": 0.01035806,
      "balance_loss_clip": 1.05227828,
      "balance_loss_mlp": 1.02710986,
      "epoch": 0.3683039740275356,
      "flos": 21466644877440.0,
      "grad_norm": 8.166787307071964,
      "language_loss": 0.7183699,
      "learning_rate": 2.9149840510606614e-06,
      "loss": 0.74047703,
      "num_input_tokens_seen": 65730350,
      "step": 3063,
      "time_per_iteration": 2.730886459350586
    },
    {
      "auxiliary_loss_clip": 0.01093734,
      "auxiliary_loss_mlp": 0.01001673,
      "balance_loss_clip": 1.01941919,
      "balance_loss_mlp": 1.00013483,
      "epoch": 0.36842421691817473,
      "flos": 70380999987840.0,
      "grad_norm": 1.0319145277952964,
      "language_loss": 0.64208317,
      "learning_rate": 2.914291309397769e-06,
      "loss": 0.66303718,
      "num_input_tokens_seen": 65787820,
      "step": 3064,
      "time_per_iteration": 3.2707369327545166
    },
    {
      "auxiliary_loss_clip": 0.01204335,
      "auxiliary_loss_mlp": 0.01034064,
      "balance_loss_clip": 1.06051385,
      "balance_loss_mlp": 1.02440786,
      "epoch": 0.3685444598088138,
      "flos": 23331378510720.0,
      "grad_norm": 2.3507607036462987,
      "language_loss": 0.78458118,
      "learning_rate": 2.9135984290376117e-06,
      "loss": 0.80696517,
      "num_input_tokens_seen": 65806685,
      "step": 3065,
      "time_per_iteration": 2.6965198516845703
    },
    {
      "auxiliary_loss_clip": 0.01185129,
      "auxiliary_loss_mlp": 0.0104033,
      "balance_loss_clip": 1.0591228,
      "balance_loss_mlp": 1.03074586,
      "epoch": 0.3686647026994529,
      "flos": 23070271570560.0,
      "grad_norm": 1.905916799102708,
      "language_loss": 0.83105963,
      "learning_rate": 2.9129054100853e-06,
      "loss": 0.85331428,
      "num_input_tokens_seen": 65825525,
      "step": 3066,
      "time_per_iteration": 2.649864912033081
    },
    {
      "auxiliary_loss_clip": 0.01203181,
      "auxiliary_loss_mlp": 0.01043827,
      "balance_loss_clip": 1.05885017,
      "balance_loss_mlp": 1.03442097,
      "epoch": 0.368784945590092,
      "flos": 25119909440640.0,
      "grad_norm": 4.768805784783694,
      "language_loss": 0.76151776,
      "learning_rate": 2.912212252645963e-06,
      "loss": 0.78398782,
      "num_input_tokens_seen": 65848110,
      "step": 3067,
      "time_per_iteration": 2.7594668865203857
    },
    {
      "auxiliary_loss_clip": 0.01196849,
      "auxiliary_loss_mlp": 0.0103605,
      "balance_loss_clip": 1.05906773,
      "balance_loss_mlp": 1.02499914,
      "epoch": 0.36890518848073106,
      "flos": 18442284566400.0,
      "grad_norm": 2.600705698299164,
      "language_loss": 0.76745337,
      "learning_rate": 2.9115189568247523e-06,
      "loss": 0.78978235,
      "num_input_tokens_seen": 65865670,
      "step": 3068,
      "time_per_iteration": 2.5712649822235107
    },
    {
      "auxiliary_loss_clip": 0.01191447,
      "auxiliary_loss_mlp": 0.01032277,
      "balance_loss_clip": 1.05916846,
      "balance_loss_mlp": 1.02257967,
      "epoch": 0.36902543137137017,
      "flos": 16362446336640.0,
      "grad_norm": 2.0777758356223592,
      "language_loss": 0.92333722,
      "learning_rate": 2.910825522726841e-06,
      "loss": 0.94557446,
      "num_input_tokens_seen": 65883195,
      "step": 3069,
      "time_per_iteration": 2.6875133514404297
    },
    {
      "auxiliary_loss_clip": 0.0120128,
      "auxiliary_loss_mlp": 0.01036235,
      "balance_loss_clip": 1.06049287,
      "balance_loss_mlp": 1.02625751,
      "epoch": 0.3691456742620093,
      "flos": 12275596702080.0,
      "grad_norm": 3.010366560660707,
      "language_loss": 0.77785146,
      "learning_rate": 2.9101319504574215e-06,
      "loss": 0.80022663,
      "num_input_tokens_seen": 65899635,
      "step": 3070,
      "time_per_iteration": 2.5917930603027344
    },
    {
      "auxiliary_loss_clip": 0.01203249,
      "auxiliary_loss_mlp": 0.01034253,
      "balance_loss_clip": 1.05990839,
      "balance_loss_mlp": 1.02488303,
      "epoch": 0.36926591715264834,
      "flos": 17786412178560.0,
      "grad_norm": 2.546468239804844,
      "language_loss": 0.76644492,
      "learning_rate": 2.909438240121709e-06,
      "loss": 0.78881991,
      "num_input_tokens_seen": 65919910,
      "step": 3071,
      "time_per_iteration": 2.653843641281128
    },
    {
      "auxiliary_loss_clip": 0.01201976,
      "auxiliary_loss_mlp": 0.01031372,
      "balance_loss_clip": 1.06052804,
      "balance_loss_mlp": 1.02208543,
      "epoch": 0.36938616004328745,
      "flos": 28948309741440.0,
      "grad_norm": 5.5197820319801,
      "language_loss": 0.70711929,
      "learning_rate": 2.908744391824939e-06,
      "loss": 0.72945279,
      "num_input_tokens_seen": 65940930,
      "step": 3072,
      "time_per_iteration": 2.6979916095733643
    },
    {
      "auxiliary_loss_clip": 0.01183664,
      "auxiliary_loss_mlp": 0.01035517,
      "balance_loss_clip": 1.05638313,
      "balance_loss_mlp": 1.02596784,
      "epoch": 0.36950640293392656,
      "flos": 29205394358400.0,
      "grad_norm": 2.8939061462746447,
      "language_loss": 0.78944886,
      "learning_rate": 2.908050405672367e-06,
      "loss": 0.81164068,
      "num_input_tokens_seen": 65960475,
      "step": 3073,
      "time_per_iteration": 2.713315725326538
    },
    {
      "auxiliary_loss_clip": 0.01172431,
      "auxiliary_loss_mlp": 0.01035446,
      "balance_loss_clip": 1.05498004,
      "balance_loss_mlp": 1.02577257,
      "epoch": 0.3696266458245656,
      "flos": 24827776128000.0,
      "grad_norm": 1.8285165636744674,
      "language_loss": 0.79326904,
      "learning_rate": 2.9073562817692703e-06,
      "loss": 0.81534779,
      "num_input_tokens_seen": 65979160,
      "step": 3074,
      "time_per_iteration": 3.641700506210327
    },
    {
      "auxiliary_loss_clip": 0.01092815,
      "auxiliary_loss_mlp": 0.01002053,
      "balance_loss_clip": 1.01857042,
      "balance_loss_mlp": 1.00053883,
      "epoch": 0.3697468887152047,
      "flos": 59887257264000.0,
      "grad_norm": 0.7217092504443654,
      "language_loss": 0.56446826,
      "learning_rate": 2.9066620202209468e-06,
      "loss": 0.58541691,
      "num_input_tokens_seen": 66041650,
      "step": 3075,
      "time_per_iteration": 3.1721880435943604
    },
    {
      "auxiliary_loss_clip": 0.01188291,
      "auxiliary_loss_mlp": 0.01035213,
      "balance_loss_clip": 1.05927885,
      "balance_loss_mlp": 1.02662432,
      "epoch": 0.3698671316058438,
      "flos": 26137581569280.0,
      "grad_norm": 3.374828673696694,
      "language_loss": 0.78203893,
      "learning_rate": 2.905967621132716e-06,
      "loss": 0.80427396,
      "num_input_tokens_seen": 66059260,
      "step": 3076,
      "time_per_iteration": 3.648524045944214
    },
    {
      "auxiliary_loss_clip": 0.01190059,
      "auxiliary_loss_mlp": 0.01034191,
      "balance_loss_clip": 1.05604792,
      "balance_loss_mlp": 1.02399838,
      "epoch": 0.3699873744964829,
      "flos": 24607464059520.0,
      "grad_norm": 3.1187033821694747,
      "language_loss": 0.75500345,
      "learning_rate": 2.9052730846099172e-06,
      "loss": 0.777246,
      "num_input_tokens_seen": 66080605,
      "step": 3077,
      "time_per_iteration": 2.7003281116485596
    },
    {
      "auxiliary_loss_clip": 0.01080537,
      "auxiliary_loss_mlp": 0.01006782,
      "balance_loss_clip": 1.01726246,
      "balance_loss_mlp": 1.00516069,
      "epoch": 0.370107617387122,
      "flos": 64885340050560.0,
      "grad_norm": 0.8829197701770297,
      "language_loss": 0.60865355,
      "learning_rate": 2.9045784107579123e-06,
      "loss": 0.62952673,
      "num_input_tokens_seen": 66140710,
      "step": 3078,
      "time_per_iteration": 3.226168632507324
    },
    {
      "auxiliary_loss_clip": 0.01189016,
      "auxiliary_loss_mlp": 0.01035603,
      "balance_loss_clip": 1.05779397,
      "balance_loss_mlp": 1.02639413,
      "epoch": 0.37022786027776106,
      "flos": 15961683317760.0,
      "grad_norm": 2.009277266507001,
      "language_loss": 0.67354584,
      "learning_rate": 2.9038835996820807e-06,
      "loss": 0.69579208,
      "num_input_tokens_seen": 66158320,
      "step": 3079,
      "time_per_iteration": 3.6004695892333984
    },
    {
      "auxiliary_loss_clip": 0.01188612,
      "auxiliary_loss_mlp": 0.01035145,
      "balance_loss_clip": 1.06094956,
      "balance_loss_mlp": 1.02655602,
      "epoch": 0.37034810316840017,
      "flos": 18546927863040.0,
      "grad_norm": 1.9555551092512646,
      "language_loss": 0.79848057,
      "learning_rate": 2.903188651487826e-06,
      "loss": 0.82071817,
      "num_input_tokens_seen": 66176875,
      "step": 3080,
      "time_per_iteration": 3.867990016937256
    },
    {
      "auxiliary_loss_clip": 0.0118432,
      "auxiliary_loss_mlp": 0.01030106,
      "balance_loss_clip": 1.05860329,
      "balance_loss_mlp": 1.02102864,
      "epoch": 0.3704683460590393,
      "flos": 17821927751040.0,
      "grad_norm": 2.1748639520327626,
      "language_loss": 0.86512566,
      "learning_rate": 2.902493566280571e-06,
      "loss": 0.88726997,
      "num_input_tokens_seen": 66194980,
      "step": 3081,
      "time_per_iteration": 2.628866195678711
    },
    {
      "auxiliary_loss_clip": 0.0118777,
      "auxiliary_loss_mlp": 0.01036481,
      "balance_loss_clip": 1.0536226,
      "balance_loss_mlp": 1.02663994,
      "epoch": 0.37058858894967833,
      "flos": 14134081368960.0,
      "grad_norm": 2.5792679193964214,
      "language_loss": 0.81532788,
      "learning_rate": 2.9017983441657595e-06,
      "loss": 0.83757043,
      "num_input_tokens_seen": 66212310,
      "step": 3082,
      "time_per_iteration": 2.668565511703491
    },
    {
      "auxiliary_loss_clip": 0.01183106,
      "auxiliary_loss_mlp": 0.01034614,
      "balance_loss_clip": 1.05796921,
      "balance_loss_mlp": 1.02498221,
      "epoch": 0.37070883184031744,
      "flos": 13954492344960.0,
      "grad_norm": 2.1444556453117274,
      "language_loss": 0.7623229,
      "learning_rate": 2.9011029852488564e-06,
      "loss": 0.78450012,
      "num_input_tokens_seen": 66229545,
      "step": 3083,
      "time_per_iteration": 2.6097331047058105
    },
    {
      "auxiliary_loss_clip": 0.01091175,
      "auxiliary_loss_mlp": 0.01000739,
      "balance_loss_clip": 1.01712203,
      "balance_loss_mlp": 0.99927324,
      "epoch": 0.37082907473095655,
      "flos": 52315419306240.0,
      "grad_norm": 1.0838377356955138,
      "language_loss": 0.6246177,
      "learning_rate": 2.9004074896353465e-06,
      "loss": 0.64553684,
      "num_input_tokens_seen": 66283545,
      "step": 3084,
      "time_per_iteration": 3.104247808456421
    },
    {
      "auxiliary_loss_clip": 0.0119421,
      "auxiliary_loss_mlp": 0.01038035,
      "balance_loss_clip": 1.06312442,
      "balance_loss_mlp": 1.02814043,
      "epoch": 0.3709493176215956,
      "flos": 15998096730240.0,
      "grad_norm": 1.9518756325216196,
      "language_loss": 0.81824565,
      "learning_rate": 2.8997118574307362e-06,
      "loss": 0.84056807,
      "num_input_tokens_seen": 66300500,
      "step": 3085,
      "time_per_iteration": 2.6494033336639404
    },
    {
      "auxiliary_loss_clip": 0.01180357,
      "auxiliary_loss_mlp": 0.01033689,
      "balance_loss_clip": 1.05717194,
      "balance_loss_mlp": 1.02455699,
      "epoch": 0.3710695605122347,
      "flos": 20959837931520.0,
      "grad_norm": 1.9386465173238479,
      "language_loss": 0.74630553,
      "learning_rate": 2.899016088740553e-06,
      "loss": 0.76844597,
      "num_input_tokens_seen": 66318610,
      "step": 3086,
      "time_per_iteration": 2.6261441707611084
    },
    {
      "auxiliary_loss_clip": 0.01184176,
      "auxiliary_loss_mlp": 0.01032473,
      "balance_loss_clip": 1.05706692,
      "balance_loss_mlp": 1.02330613,
      "epoch": 0.37118980340287383,
      "flos": 14355578586240.0,
      "grad_norm": 2.808005474780545,
      "language_loss": 0.79613465,
      "learning_rate": 2.898320183670344e-06,
      "loss": 0.81830108,
      "num_input_tokens_seen": 66336025,
      "step": 3087,
      "time_per_iteration": 2.7474019527435303
    },
    {
      "auxiliary_loss_clip": 0.01189021,
      "auxiliary_loss_mlp": 0.01034707,
      "balance_loss_clip": 1.06074882,
      "balance_loss_mlp": 1.02497935,
      "epoch": 0.3713100462935129,
      "flos": 25885381201920.0,
      "grad_norm": 1.6326656807738633,
      "language_loss": 0.89204651,
      "learning_rate": 2.8976241423256767e-06,
      "loss": 0.91428375,
      "num_input_tokens_seen": 66356120,
      "step": 3088,
      "time_per_iteration": 2.7046899795532227
    },
    {
      "auxiliary_loss_clip": 0.01184991,
      "auxiliary_loss_mlp": 0.00763434,
      "balance_loss_clip": 1.06155574,
      "balance_loss_mlp": 1.00143707,
      "epoch": 0.371430289184152,
      "flos": 30518934814080.0,
      "grad_norm": 2.579877246015885,
      "language_loss": 0.684448,
      "learning_rate": 2.896927964812142e-06,
      "loss": 0.70393229,
      "num_input_tokens_seen": 66376685,
      "step": 3089,
      "time_per_iteration": 2.7845118045806885
    },
    {
      "auxiliary_loss_clip": 0.0115169,
      "auxiliary_loss_mlp": 0.01029355,
      "balance_loss_clip": 1.0562799,
      "balance_loss_mlp": 1.01999688,
      "epoch": 0.37155053207479105,
      "flos": 15742233175680.0,
      "grad_norm": 2.550241558049245,
      "language_loss": 0.7462824,
      "learning_rate": 2.8962316512353465e-06,
      "loss": 0.76809287,
      "num_input_tokens_seen": 66394230,
      "step": 3090,
      "time_per_iteration": 2.6485931873321533
    },
    {
      "auxiliary_loss_clip": 0.01173806,
      "auxiliary_loss_mlp": 0.01034764,
      "balance_loss_clip": 1.05970347,
      "balance_loss_mlp": 1.02548313,
      "epoch": 0.37167077496543016,
      "flos": 23404061681280.0,
      "grad_norm": 1.9706259393912344,
      "language_loss": 0.7502104,
      "learning_rate": 2.8955352017009233e-06,
      "loss": 0.77229607,
      "num_input_tokens_seen": 66413475,
      "step": 3091,
      "time_per_iteration": 2.735974073410034
    },
    {
      "auxiliary_loss_clip": 0.01202398,
      "auxiliary_loss_mlp": 0.0103505,
      "balance_loss_clip": 1.0596931,
      "balance_loss_mlp": 1.02650237,
      "epoch": 0.3717910178560693,
      "flos": 22088653718400.0,
      "grad_norm": 2.193859777651357,
      "language_loss": 0.77257907,
      "learning_rate": 2.8948386163145212e-06,
      "loss": 0.79495353,
      "num_input_tokens_seen": 66432685,
      "step": 3092,
      "time_per_iteration": 2.6085877418518066
    },
    {
      "auxiliary_loss_clip": 0.01168621,
      "auxiliary_loss_mlp": 0.01034229,
      "balance_loss_clip": 1.05268168,
      "balance_loss_mlp": 1.02560401,
      "epoch": 0.3719112607467083,
      "flos": 26939969533440.0,
      "grad_norm": 2.5105090109102948,
      "language_loss": 0.79726636,
      "learning_rate": 2.8941418951818135e-06,
      "loss": 0.81929481,
      "num_input_tokens_seen": 66452245,
      "step": 3093,
      "time_per_iteration": 2.7475571632385254
    },
    {
      "auxiliary_loss_clip": 0.0118432,
      "auxiliary_loss_mlp": 0.01036701,
      "balance_loss_clip": 1.05719447,
      "balance_loss_mlp": 1.02783179,
      "epoch": 0.37203150363734744,
      "flos": 12166500119040.0,
      "grad_norm": 2.5174312515302404,
      "language_loss": 0.71725333,
      "learning_rate": 2.8934450384084903e-06,
      "loss": 0.73946357,
      "num_input_tokens_seen": 66469760,
      "step": 3094,
      "time_per_iteration": 2.5896689891815186
    },
    {
      "auxiliary_loss_clip": 0.01201764,
      "auxiliary_loss_mlp": 0.01035006,
      "balance_loss_clip": 1.05834603,
      "balance_loss_mlp": 1.02601171,
      "epoch": 0.37215174652798655,
      "flos": 23697595624320.0,
      "grad_norm": 2.0822421430858107,
      "language_loss": 0.702564,
      "learning_rate": 2.8927480461002653e-06,
      "loss": 0.72493172,
      "num_input_tokens_seen": 66489730,
      "step": 3095,
      "time_per_iteration": 2.6707520484924316
    },
    {
      "auxiliary_loss_clip": 0.01169973,
      "auxiliary_loss_mlp": 0.01033594,
      "balance_loss_clip": 1.05022001,
      "balance_loss_mlp": 1.02391458,
      "epoch": 0.3722719894186256,
      "flos": 17887751424000.0,
      "grad_norm": 2.3990621988611345,
      "language_loss": 0.86836725,
      "learning_rate": 2.892050918362872e-06,
      "loss": 0.89040291,
      "num_input_tokens_seen": 66504785,
      "step": 3096,
      "time_per_iteration": 2.6842854022979736
    },
    {
      "auxiliary_loss_clip": 0.01071421,
      "auxiliary_loss_mlp": 0.01003095,
      "balance_loss_clip": 1.01659441,
      "balance_loss_mlp": 1.00155759,
      "epoch": 0.3723922323092647,
      "flos": 62419891363200.0,
      "grad_norm": 0.9421945309971662,
      "language_loss": 0.55842149,
      "learning_rate": 2.8913536553020626e-06,
      "loss": 0.57916665,
      "num_input_tokens_seen": 66558840,
      "step": 3097,
      "time_per_iteration": 3.239769220352173
    },
    {
      "auxiliary_loss_clip": 0.01174656,
      "auxiliary_loss_mlp": 0.01034421,
      "balance_loss_clip": 1.05758834,
      "balance_loss_mlp": 1.02565312,
      "epoch": 0.3725124751999038,
      "flos": 23039747988480.0,
      "grad_norm": 2.853230082430737,
      "language_loss": 0.85183978,
      "learning_rate": 2.8906562570236137e-06,
      "loss": 0.87393051,
      "num_input_tokens_seen": 66576750,
      "step": 3098,
      "time_per_iteration": 2.7287511825561523
    },
    {
      "auxiliary_loss_clip": 0.01200022,
      "auxiliary_loss_mlp": 0.01032797,
      "balance_loss_clip": 1.05910563,
      "balance_loss_mlp": 1.02418423,
      "epoch": 0.3726327180905429,
      "flos": 20920551431040.0,
      "grad_norm": 1.5407170957182215,
      "language_loss": 0.7652899,
      "learning_rate": 2.889958723633318e-06,
      "loss": 0.7876181,
      "num_input_tokens_seen": 66595690,
      "step": 3099,
      "time_per_iteration": 2.591759443283081
    },
    {
      "auxiliary_loss_clip": 0.01187476,
      "auxiliary_loss_mlp": 0.01037382,
      "balance_loss_clip": 1.0592103,
      "balance_loss_mlp": 1.02823877,
      "epoch": 0.372752960981182,
      "flos": 30592156688640.0,
      "grad_norm": 2.8966196255203016,
      "language_loss": 0.73793775,
      "learning_rate": 2.889261055236992e-06,
      "loss": 0.76018631,
      "num_input_tokens_seen": 66617905,
      "step": 3100,
      "time_per_iteration": 3.7097249031066895
    },
    {
      "auxiliary_loss_clip": 0.01162869,
      "auxiliary_loss_mlp": 0.01038178,
      "balance_loss_clip": 1.05557966,
      "balance_loss_mlp": 1.02893329,
      "epoch": 0.3728732038718211,
      "flos": 25116749043840.0,
      "grad_norm": 3.875929121476214,
      "language_loss": 0.83175421,
      "learning_rate": 2.8885632519404704e-06,
      "loss": 0.85376465,
      "num_input_tokens_seen": 66638175,
      "step": 3101,
      "time_per_iteration": 2.7186827659606934
    },
    {
      "auxiliary_loss_clip": 0.01156886,
      "auxiliary_loss_mlp": 0.01036519,
      "balance_loss_clip": 1.05452013,
      "balance_loss_mlp": 1.02779269,
      "epoch": 0.37299344676246016,
      "flos": 25302048330240.0,
      "grad_norm": 2.2922260455925145,
      "language_loss": 0.76052004,
      "learning_rate": 2.8878653138496107e-06,
      "loss": 0.78245413,
      "num_input_tokens_seen": 66658670,
      "step": 3102,
      "time_per_iteration": 3.646294116973877
    },
    {
      "auxiliary_loss_clip": 0.01205416,
      "auxiliary_loss_mlp": 0.0103054,
      "balance_loss_clip": 1.06168413,
      "balance_loss_mlp": 1.02058625,
      "epoch": 0.37311368965309927,
      "flos": 23842531002240.0,
      "grad_norm": 3.4956022111535563,
      "language_loss": 0.77197003,
      "learning_rate": 2.8871672410702878e-06,
      "loss": 0.79432958,
      "num_input_tokens_seen": 66676030,
      "step": 3103,
      "time_per_iteration": 2.6891167163848877
    },
    {
      "auxiliary_loss_clip": 0.01186182,
      "auxiliary_loss_mlp": 0.01033419,
      "balance_loss_clip": 1.05895352,
      "balance_loss_mlp": 1.0232271,
      "epoch": 0.3732339325437384,
      "flos": 25811943845760.0,
      "grad_norm": 2.196487123056005,
      "language_loss": 0.8237856,
      "learning_rate": 2.8864690337084008e-06,
      "loss": 0.8459816,
      "num_input_tokens_seen": 66695305,
      "step": 3104,
      "time_per_iteration": 2.661904811859131
    },
    {
      "auxiliary_loss_clip": 0.0117989,
      "auxiliary_loss_mlp": 0.0103472,
      "balance_loss_clip": 1.05907381,
      "balance_loss_mlp": 1.02532077,
      "epoch": 0.37335417543437743,
      "flos": 26208433146240.0,
      "grad_norm": 2.7151619501931266,
      "language_loss": 0.78229892,
      "learning_rate": 2.885770691869866e-06,
      "loss": 0.80444503,
      "num_input_tokens_seen": 66716185,
      "step": 3105,
      "time_per_iteration": 3.671217441558838
    },
    {
      "auxiliary_loss_clip": 0.01198309,
      "auxiliary_loss_mlp": 0.01035832,
      "balance_loss_clip": 1.05962253,
      "balance_loss_mlp": 1.02721953,
      "epoch": 0.37347441832501654,
      "flos": 24023879792640.0,
      "grad_norm": 3.0395629269163407,
      "language_loss": 0.747639,
      "learning_rate": 2.8850722156606207e-06,
      "loss": 0.76998043,
      "num_input_tokens_seen": 66734575,
      "step": 3106,
      "time_per_iteration": 3.6400349140167236
    },
    {
      "auxiliary_loss_clip": 0.01198374,
      "auxiliary_loss_mlp": 0.01034484,
      "balance_loss_clip": 1.05902982,
      "balance_loss_mlp": 1.02528095,
      "epoch": 0.3735946612156556,
      "flos": 19714922409600.0,
      "grad_norm": 2.8606282005361243,
      "language_loss": 0.67365527,
      "learning_rate": 2.8843736051866252e-06,
      "loss": 0.69598383,
      "num_input_tokens_seen": 66753500,
      "step": 3107,
      "time_per_iteration": 2.6536898612976074
    },
    {
      "auxiliary_loss_clip": 0.01181337,
      "auxiliary_loss_mlp": 0.01033633,
      "balance_loss_clip": 1.05566216,
      "balance_loss_mlp": 1.02405465,
      "epoch": 0.3737149041062947,
      "flos": 23039604334080.0,
      "grad_norm": 1.7828073951327483,
      "language_loss": 0.69519877,
      "learning_rate": 2.8836748605538557e-06,
      "loss": 0.71734846,
      "num_input_tokens_seen": 66775140,
      "step": 3108,
      "time_per_iteration": 2.6750645637512207
    },
    {
      "auxiliary_loss_clip": 0.0117345,
      "auxiliary_loss_mlp": 0.01026093,
      "balance_loss_clip": 1.0565896,
      "balance_loss_mlp": 1.01712871,
      "epoch": 0.3738351469969338,
      "flos": 34678108483200.0,
      "grad_norm": 2.8876406938134833,
      "language_loss": 0.63511807,
      "learning_rate": 2.882975981868313e-06,
      "loss": 0.65711349,
      "num_input_tokens_seen": 66795525,
      "step": 3109,
      "time_per_iteration": 2.7534468173980713
    },
    {
      "auxiliary_loss_clip": 0.01195768,
      "auxiliary_loss_mlp": 0.01033109,
      "balance_loss_clip": 1.06106949,
      "balance_loss_mlp": 1.02351272,
      "epoch": 0.3739553898875729,
      "flos": 43507967448960.0,
      "grad_norm": 2.66994103965097,
      "language_loss": 0.69104248,
      "learning_rate": 2.882276969236016e-06,
      "loss": 0.71333128,
      "num_input_tokens_seen": 66816885,
      "step": 3110,
      "time_per_iteration": 2.9725701808929443
    },
    {
      "auxiliary_loss_clip": 0.011919,
      "auxiliary_loss_mlp": 0.01031736,
      "balance_loss_clip": 1.06039548,
      "balance_loss_mlp": 1.02215731,
      "epoch": 0.374075632778212,
      "flos": 12856487448960.0,
      "grad_norm": 2.866064360223247,
      "language_loss": 0.76441491,
      "learning_rate": 2.881577822763005e-06,
      "loss": 0.78665125,
      "num_input_tokens_seen": 66834835,
      "step": 3111,
      "time_per_iteration": 2.6370575428009033
    },
    {
      "auxiliary_loss_clip": 0.01201058,
      "auxiliary_loss_mlp": 0.01034792,
      "balance_loss_clip": 1.05861568,
      "balance_loss_mlp": 1.0255177,
      "epoch": 0.3741958756688511,
      "flos": 26024031699840.0,
      "grad_norm": 2.1933340160034898,
      "language_loss": 0.87683845,
      "learning_rate": 2.880878542555338e-06,
      "loss": 0.89919698,
      "num_input_tokens_seen": 66852600,
      "step": 3112,
      "time_per_iteration": 2.623384952545166
    },
    {
      "auxiliary_loss_clip": 0.01175942,
      "auxiliary_loss_mlp": 0.01038436,
      "balance_loss_clip": 1.05799341,
      "balance_loss_mlp": 1.02928722,
      "epoch": 0.37431611855949015,
      "flos": 21433894652160.0,
      "grad_norm": 2.143309571703776,
      "language_loss": 0.80634642,
      "learning_rate": 2.8801791287190976e-06,
      "loss": 0.82849026,
      "num_input_tokens_seen": 66870595,
      "step": 3113,
      "time_per_iteration": 2.689140796661377
    },
    {
      "auxiliary_loss_clip": 0.01187401,
      "auxiliary_loss_mlp": 0.01034418,
      "balance_loss_clip": 1.05951118,
      "balance_loss_mlp": 1.02504849,
      "epoch": 0.37443636145012926,
      "flos": 24207096090240.0,
      "grad_norm": 3.5014345900370065,
      "language_loss": 0.85918915,
      "learning_rate": 2.8794795813603817e-06,
      "loss": 0.88140738,
      "num_input_tokens_seen": 66886060,
      "step": 3114,
      "time_per_iteration": 2.6673195362091064
    },
    {
      "auxiliary_loss_clip": 0.01173885,
      "auxiliary_loss_mlp": 0.01033242,
      "balance_loss_clip": 1.05702972,
      "balance_loss_mlp": 1.02363944,
      "epoch": 0.3745566043407684,
      "flos": 15378601841280.0,
      "grad_norm": 1.784141656089687,
      "language_loss": 0.82013941,
      "learning_rate": 2.878779900585314e-06,
      "loss": 0.84221065,
      "num_input_tokens_seen": 66903900,
      "step": 3115,
      "time_per_iteration": 2.7292685508728027
    },
    {
      "auxiliary_loss_clip": 0.01159532,
      "auxiliary_loss_mlp": 0.00763948,
      "balance_loss_clip": 1.05543649,
      "balance_loss_mlp": 1.00136566,
      "epoch": 0.37467684723140743,
      "flos": 24608218245120.0,
      "grad_norm": 1.5963493474482042,
      "language_loss": 0.75423765,
      "learning_rate": 2.8780800865000336e-06,
      "loss": 0.77347243,
      "num_input_tokens_seen": 66925210,
      "step": 3116,
      "time_per_iteration": 2.7574551105499268
    },
    {
      "auxiliary_loss_clip": 0.01054988,
      "auxiliary_loss_mlp": 0.01007029,
      "balance_loss_clip": 1.01603389,
      "balance_loss_mlp": 1.00536013,
      "epoch": 0.37479709012204654,
      "flos": 64377491610240.0,
      "grad_norm": 0.9806073503536763,
      "language_loss": 0.59187818,
      "learning_rate": 2.877380139210702e-06,
      "loss": 0.61249828,
      "num_input_tokens_seen": 66983880,
      "step": 3117,
      "time_per_iteration": 3.2765822410583496
    },
    {
      "auxiliary_loss_clip": 0.01191704,
      "auxiliary_loss_mlp": 0.01038378,
      "balance_loss_clip": 1.05698001,
      "balance_loss_mlp": 1.02843034,
      "epoch": 0.37491733301268565,
      "flos": 23803962773760.0,
      "grad_norm": 1.8871660665764176,
      "language_loss": 0.76448613,
      "learning_rate": 2.876680058823501e-06,
      "loss": 0.78678697,
      "num_input_tokens_seen": 67004280,
      "step": 3118,
      "time_per_iteration": 2.677500009536743
    },
    {
      "auxiliary_loss_clip": 0.0120466,
      "auxiliary_loss_mlp": 0.01039967,
      "balance_loss_clip": 1.06096017,
      "balance_loss_mlp": 1.0307945,
      "epoch": 0.3750375759033247,
      "flos": 32160950167680.0,
      "grad_norm": 1.8726874446410053,
      "language_loss": 0.66021454,
      "learning_rate": 2.8759798454446314e-06,
      "loss": 0.68266082,
      "num_input_tokens_seen": 67027445,
      "step": 3119,
      "time_per_iteration": 2.6988725662231445
    },
    {
      "auxiliary_loss_clip": 0.01204397,
      "auxiliary_loss_mlp": 0.01044149,
      "balance_loss_clip": 1.06339407,
      "balance_loss_mlp": 1.03487492,
      "epoch": 0.3751578187939638,
      "flos": 23367791923200.0,
      "grad_norm": 2.2104380538945816,
      "language_loss": 0.81558716,
      "learning_rate": 2.8752794991803173e-06,
      "loss": 0.83807266,
      "num_input_tokens_seen": 67045130,
      "step": 3120,
      "time_per_iteration": 2.6136016845703125
    },
    {
      "auxiliary_loss_clip": 0.01199774,
      "auxiliary_loss_mlp": 0.01042003,
      "balance_loss_clip": 1.05994499,
      "balance_loss_mlp": 1.03293681,
      "epoch": 0.37527806168460287,
      "flos": 14605731878400.0,
      "grad_norm": 2.210764889237605,
      "language_loss": 0.75536126,
      "learning_rate": 2.8745790201367976e-06,
      "loss": 0.77777898,
      "num_input_tokens_seen": 67060885,
      "step": 3121,
      "time_per_iteration": 2.6364009380340576
    },
    {
      "auxiliary_loss_clip": 0.01192767,
      "auxiliary_loss_mlp": 0.01030017,
      "balance_loss_clip": 1.0576638,
      "balance_loss_mlp": 1.01964617,
      "epoch": 0.375398304575242,
      "flos": 26390823431040.0,
      "grad_norm": 2.314713887156926,
      "language_loss": 0.84433621,
      "learning_rate": 2.8738784084203373e-06,
      "loss": 0.86656404,
      "num_input_tokens_seen": 67080960,
      "step": 3122,
      "time_per_iteration": 2.6988749504089355
    },
    {
      "auxiliary_loss_clip": 0.01166449,
      "auxiliary_loss_mlp": 0.01040486,
      "balance_loss_clip": 1.05554628,
      "balance_loss_mlp": 1.03149772,
      "epoch": 0.3755185474658811,
      "flos": 22236605838720.0,
      "grad_norm": 1.6909758235393695,
      "language_loss": 0.78738981,
      "learning_rate": 2.873177664137216e-06,
      "loss": 0.80945915,
      "num_input_tokens_seen": 67101890,
      "step": 3123,
      "time_per_iteration": 2.6684482097625732
    },
    {
      "auxiliary_loss_clip": 0.01161605,
      "auxiliary_loss_mlp": 0.01033655,
      "balance_loss_clip": 1.05420136,
      "balance_loss_mlp": 1.02368879,
      "epoch": 0.37563879035652015,
      "flos": 30812935633920.0,
      "grad_norm": 1.9802667180119817,
      "language_loss": 0.69320917,
      "learning_rate": 2.8724767873937384e-06,
      "loss": 0.7151618,
      "num_input_tokens_seen": 67126010,
      "step": 3124,
      "time_per_iteration": 2.785250663757324
    },
    {
      "auxiliary_loss_clip": 0.01186457,
      "auxiliary_loss_mlp": 0.01037272,
      "balance_loss_clip": 1.05566883,
      "balance_loss_mlp": 1.02760971,
      "epoch": 0.37575903324715926,
      "flos": 20773533064320.0,
      "grad_norm": 2.2414763598880088,
      "language_loss": 0.87743247,
      "learning_rate": 2.871775778296225e-06,
      "loss": 0.89966977,
      "num_input_tokens_seen": 67143100,
      "step": 3125,
      "time_per_iteration": 2.618565797805786
    },
    {
      "auxiliary_loss_clip": 0.01188729,
      "auxiliary_loss_mlp": 0.01030646,
      "balance_loss_clip": 1.05757022,
      "balance_loss_mlp": 1.02060878,
      "epoch": 0.37587927613779837,
      "flos": 18697681244160.0,
      "grad_norm": 2.294124820469962,
      "language_loss": 0.78642714,
      "learning_rate": 2.8710746369510196e-06,
      "loss": 0.80862093,
      "num_input_tokens_seen": 67161085,
      "step": 3126,
      "time_per_iteration": 3.545625686645508
    },
    {
      "auxiliary_loss_clip": 0.01205263,
      "auxiliary_loss_mlp": 0.01038719,
      "balance_loss_clip": 1.06189644,
      "balance_loss_mlp": 1.02886081,
      "epoch": 0.3759995190284374,
      "flos": 13624796384640.0,
      "grad_norm": 7.160003803207909,
      "language_loss": 0.84000456,
      "learning_rate": 2.8703733634644846e-06,
      "loss": 0.8624444,
      "num_input_tokens_seen": 67175840,
      "step": 3127,
      "time_per_iteration": 2.534714698791504
    },
    {
      "auxiliary_loss_clip": 0.01163676,
      "auxiliary_loss_mlp": 0.01034872,
      "balance_loss_clip": 1.05391419,
      "balance_loss_mlp": 1.02506149,
      "epoch": 0.37611976191907653,
      "flos": 20484847457280.0,
      "grad_norm": 1.8168426333807866,
      "language_loss": 0.79394686,
      "learning_rate": 2.869671957943002e-06,
      "loss": 0.81593239,
      "num_input_tokens_seen": 67194995,
      "step": 3128,
      "time_per_iteration": 3.673386335372925
    },
    {
      "auxiliary_loss_clip": 0.01163341,
      "auxiliary_loss_mlp": 0.01033712,
      "balance_loss_clip": 1.05589056,
      "balance_loss_mlp": 1.02461028,
      "epoch": 0.37624000480971564,
      "flos": 21141797253120.0,
      "grad_norm": 2.9702040425152676,
      "language_loss": 0.74748749,
      "learning_rate": 2.8689704204929747e-06,
      "loss": 0.76945794,
      "num_input_tokens_seen": 67214175,
      "step": 3129,
      "time_per_iteration": 2.7566945552825928
    },
    {
      "auxiliary_loss_clip": 0.01201888,
      "auxiliary_loss_mlp": 0.01030722,
      "balance_loss_clip": 1.06100333,
      "balance_loss_mlp": 1.02151358,
      "epoch": 0.3763602477003547,
      "flos": 22564470205440.0,
      "grad_norm": 1.9099852259010512,
      "language_loss": 0.81251371,
      "learning_rate": 2.8682687512208253e-06,
      "loss": 0.83483982,
      "num_input_tokens_seen": 67233185,
      "step": 3130,
      "time_per_iteration": 3.5063610076904297
    },
    {
      "auxiliary_loss_clip": 0.01188444,
      "auxiliary_loss_mlp": 0.01040229,
      "balance_loss_clip": 1.05947053,
      "balance_loss_mlp": 1.02957153,
      "epoch": 0.3764804905909938,
      "flos": 27526857851520.0,
      "grad_norm": 2.3420221183150223,
      "language_loss": 0.8091352,
      "learning_rate": 2.8675669502329972e-06,
      "loss": 0.83142191,
      "num_input_tokens_seen": 67254715,
      "step": 3131,
      "time_per_iteration": 2.729152202606201
    },
    {
      "auxiliary_loss_clip": 0.01154634,
      "auxiliary_loss_mlp": 0.01035403,
      "balance_loss_clip": 1.05101037,
      "balance_loss_mlp": 1.02615261,
      "epoch": 0.3766007334816329,
      "flos": 22528092706560.0,
      "grad_norm": 2.667716001429682,
      "language_loss": 0.86384761,
      "learning_rate": 2.866865017635952e-06,
      "loss": 0.88574797,
      "num_input_tokens_seen": 67272535,
      "step": 3132,
      "time_per_iteration": 3.6901166439056396
    },
    {
      "auxiliary_loss_clip": 0.0120317,
      "auxiliary_loss_mlp": 0.01039494,
      "balance_loss_clip": 1.06038582,
      "balance_loss_mlp": 1.02947485,
      "epoch": 0.376720976372272,
      "flos": 25957166532480.0,
      "grad_norm": 1.7565982656693253,
      "language_loss": 0.79368579,
      "learning_rate": 2.866162953536174e-06,
      "loss": 0.81611246,
      "num_input_tokens_seen": 67293505,
      "step": 3133,
      "time_per_iteration": 2.6483314037323
    },
    {
      "auxiliary_loss_clip": 0.01200624,
      "auxiliary_loss_mlp": 0.01032897,
      "balance_loss_clip": 1.06061316,
      "balance_loss_mlp": 1.02360427,
      "epoch": 0.3768412192629111,
      "flos": 18041162411520.0,
      "grad_norm": 1.5700459976586891,
      "language_loss": 0.74850512,
      "learning_rate": 2.8654607580401634e-06,
      "loss": 0.77084029,
      "num_input_tokens_seen": 67313240,
      "step": 3134,
      "time_per_iteration": 2.655658721923828
    },
    {
      "auxiliary_loss_clip": 0.01078438,
      "auxiliary_loss_mlp": 0.01000303,
      "balance_loss_clip": 1.01801276,
      "balance_loss_mlp": 0.99871761,
      "epoch": 0.3769614621535502,
      "flos": 62989472304000.0,
      "grad_norm": 0.8870559320450523,
      "language_loss": 0.65021229,
      "learning_rate": 2.8647584312544446e-06,
      "loss": 0.67099965,
      "num_input_tokens_seen": 67378445,
      "step": 3135,
      "time_per_iteration": 3.243844985961914
    },
    {
      "auxiliary_loss_clip": 0.01202311,
      "auxiliary_loss_mlp": 0.01034289,
      "balance_loss_clip": 1.06003988,
      "balance_loss_mlp": 1.02479434,
      "epoch": 0.37708170504418925,
      "flos": 23661685002240.0,
      "grad_norm": 1.5540084191877734,
      "language_loss": 0.85304248,
      "learning_rate": 2.864055973285559e-06,
      "loss": 0.87540853,
      "num_input_tokens_seen": 67400445,
      "step": 3136,
      "time_per_iteration": 2.6710116863250732
    },
    {
      "auxiliary_loss_clip": 0.01188979,
      "auxiliary_loss_mlp": 0.01034806,
      "balance_loss_clip": 1.06064677,
      "balance_loss_mlp": 1.02600241,
      "epoch": 0.37720194793482836,
      "flos": 24423170353920.0,
      "grad_norm": 4.817925074609005,
      "language_loss": 0.86687487,
      "learning_rate": 2.8633533842400698e-06,
      "loss": 0.88911271,
      "num_input_tokens_seen": 67420645,
      "step": 3137,
      "time_per_iteration": 2.6831281185150146
    },
    {
      "auxiliary_loss_clip": 0.01192201,
      "auxiliary_loss_mlp": 0.00763883,
      "balance_loss_clip": 1.06049895,
      "balance_loss_mlp": 1.0009706,
      "epoch": 0.3773221908254674,
      "flos": 20996502739200.0,
      "grad_norm": 4.645481925009787,
      "language_loss": 0.78010505,
      "learning_rate": 2.862650664224558e-06,
      "loss": 0.79966593,
      "num_input_tokens_seen": 67439495,
      "step": 3138,
      "time_per_iteration": 2.688173770904541
    },
    {
      "auxiliary_loss_clip": 0.01187686,
      "auxiliary_loss_mlp": 0.01033357,
      "balance_loss_clip": 1.0606842,
      "balance_loss_mlp": 1.02453578,
      "epoch": 0.37744243371610653,
      "flos": 37631724958080.0,
      "grad_norm": 2.182867272588006,
      "language_loss": 0.69994193,
      "learning_rate": 2.861947813345627e-06,
      "loss": 0.72215241,
      "num_input_tokens_seen": 67462195,
      "step": 3139,
      "time_per_iteration": 2.8053224086761475
    },
    {
      "auxiliary_loss_clip": 0.01183421,
      "auxiliary_loss_mlp": 0.01036038,
      "balance_loss_clip": 1.05811417,
      "balance_loss_mlp": 1.02679348,
      "epoch": 0.37756267660674564,
      "flos": 26140526484480.0,
      "grad_norm": 2.2035550977141125,
      "language_loss": 0.72776198,
      "learning_rate": 2.8612448317098974e-06,
      "loss": 0.74995661,
      "num_input_tokens_seen": 67482530,
      "step": 3140,
      "time_per_iteration": 2.7194628715515137
    },
    {
      "auxiliary_loss_clip": 0.01172165,
      "auxiliary_loss_mlp": 0.01032937,
      "balance_loss_clip": 1.05306673,
      "balance_loss_mlp": 1.02236891,
      "epoch": 0.3776829194973847,
      "flos": 19427888828160.0,
      "grad_norm": 2.331807650025227,
      "language_loss": 0.83350295,
      "learning_rate": 2.8605417194240114e-06,
      "loss": 0.85555398,
      "num_input_tokens_seen": 67500890,
      "step": 3141,
      "time_per_iteration": 2.6469290256500244
    },
    {
      "auxiliary_loss_clip": 0.01198832,
      "auxiliary_loss_mlp": 0.01032946,
      "balance_loss_clip": 1.05872428,
      "balance_loss_mlp": 1.02476215,
      "epoch": 0.3778031623880238,
      "flos": 17382309194880.0,
      "grad_norm": 8.214743232939279,
      "language_loss": 0.79287279,
      "learning_rate": 2.8598384765946315e-06,
      "loss": 0.81519049,
      "num_input_tokens_seen": 67519545,
      "step": 3142,
      "time_per_iteration": 2.573814868927002
    },
    {
      "auxiliary_loss_clip": 0.01166987,
      "auxiliary_loss_mlp": 0.01031365,
      "balance_loss_clip": 1.05185008,
      "balance_loss_mlp": 1.02207899,
      "epoch": 0.3779234052786629,
      "flos": 27125843437440.0,
      "grad_norm": 2.405993344034332,
      "language_loss": 0.71486461,
      "learning_rate": 2.8591351033284377e-06,
      "loss": 0.73684812,
      "num_input_tokens_seen": 67539275,
      "step": 3143,
      "time_per_iteration": 2.6965925693511963
    },
    {
      "auxiliary_loss_clip": 0.01184223,
      "auxiliary_loss_mlp": 0.0103602,
      "balance_loss_clip": 1.05686748,
      "balance_loss_mlp": 1.02637899,
      "epoch": 0.37804364816930197,
      "flos": 19682639061120.0,
      "grad_norm": 2.831987504010638,
      "language_loss": 0.83807153,
      "learning_rate": 2.8584315997321325e-06,
      "loss": 0.86027396,
      "num_input_tokens_seen": 67558280,
      "step": 3144,
      "time_per_iteration": 2.731778144836426
    },
    {
      "auxiliary_loss_clip": 0.01186037,
      "auxiliary_loss_mlp": 0.01030503,
      "balance_loss_clip": 1.05513978,
      "balance_loss_mlp": 1.02125263,
      "epoch": 0.3781638910599411,
      "flos": 22702905221760.0,
      "grad_norm": 6.504170653366459,
      "language_loss": 0.78472114,
      "learning_rate": 2.8577279659124356e-06,
      "loss": 0.80688649,
      "num_input_tokens_seen": 67575955,
      "step": 3145,
      "time_per_iteration": 2.67236328125
    },
    {
      "auxiliary_loss_clip": 0.01155685,
      "auxiliary_loss_mlp": 0.01031722,
      "balance_loss_clip": 1.05002832,
      "balance_loss_mlp": 1.02305579,
      "epoch": 0.3782841339505802,
      "flos": 14647604158080.0,
      "grad_norm": 2.783079516295005,
      "language_loss": 0.83972371,
      "learning_rate": 2.857024201976089e-06,
      "loss": 0.86159784,
      "num_input_tokens_seen": 67593515,
      "step": 3146,
      "time_per_iteration": 2.6527748107910156
    },
    {
      "auxiliary_loss_clip": 0.01161299,
      "auxiliary_loss_mlp": 0.00763912,
      "balance_loss_clip": 1.05738592,
      "balance_loss_mlp": 1.00088596,
      "epoch": 0.37840437684121925,
      "flos": 32818223185920.0,
      "grad_norm": 2.618665912540364,
      "language_loss": 0.73420131,
      "learning_rate": 2.8563203080298516e-06,
      "loss": 0.75345337,
      "num_input_tokens_seen": 67614290,
      "step": 3147,
      "time_per_iteration": 2.808774948120117
    },
    {
      "auxiliary_loss_clip": 0.01186537,
      "auxiliary_loss_mlp": 0.01033213,
      "balance_loss_clip": 1.05898404,
      "balance_loss_mlp": 1.02423072,
      "epoch": 0.37852461973185836,
      "flos": 18369206346240.0,
      "grad_norm": 2.7379113145434304,
      "language_loss": 0.89856303,
      "learning_rate": 2.855616284180505e-06,
      "loss": 0.92076051,
      "num_input_tokens_seen": 67631340,
      "step": 3148,
      "time_per_iteration": 2.627307653427124
    },
    {
      "auxiliary_loss_clip": 0.01079037,
      "auxiliary_loss_mlp": 0.01003144,
      "balance_loss_clip": 1.01609504,
      "balance_loss_mlp": 1.00142717,
      "epoch": 0.37864486262249747,
      "flos": 59500680117120.0,
      "grad_norm": 0.8816985882780638,
      "language_loss": 0.66059738,
      "learning_rate": 2.8549121305348477e-06,
      "loss": 0.68141919,
      "num_input_tokens_seen": 67691125,
      "step": 3149,
      "time_per_iteration": 3.17810320854187
    },
    {
      "auxiliary_loss_clip": 0.01198084,
      "auxiliary_loss_mlp": 0.01032116,
      "balance_loss_clip": 1.05986905,
      "balance_loss_mlp": 1.02325904,
      "epoch": 0.3787651055131365,
      "flos": 23363015414400.0,
      "grad_norm": 2.720648091944283,
      "language_loss": 0.83549786,
      "learning_rate": 2.8542078471997006e-06,
      "loss": 0.85779989,
      "num_input_tokens_seen": 67708740,
      "step": 3150,
      "time_per_iteration": 2.6176328659057617
    },
    {
      "auxiliary_loss_clip": 0.0119806,
      "auxiliary_loss_mlp": 0.0103309,
      "balance_loss_clip": 1.05776978,
      "balance_loss_mlp": 1.02442992,
      "epoch": 0.37888534840377563,
      "flos": 24601394661120.0,
      "grad_norm": 2.029683100965993,
      "language_loss": 0.7618196,
      "learning_rate": 2.8535034342819013e-06,
      "loss": 0.78413105,
      "num_input_tokens_seen": 67726150,
      "step": 3151,
      "time_per_iteration": 2.656938076019287
    },
    {
      "auxiliary_loss_clip": 0.01197455,
      "auxiliary_loss_mlp": 0.01038581,
      "balance_loss_clip": 1.05758071,
      "balance_loss_mlp": 1.02915776,
      "epoch": 0.37900559129441475,
      "flos": 23986891762560.0,
      "grad_norm": 1.6762177146933859,
      "language_loss": 0.72740138,
      "learning_rate": 2.85279889188831e-06,
      "loss": 0.7497617,
      "num_input_tokens_seen": 67746525,
      "step": 3152,
      "time_per_iteration": 3.5667436122894287
    },
    {
      "auxiliary_loss_clip": 0.01201873,
      "auxiliary_loss_mlp": 0.0103234,
      "balance_loss_clip": 1.05991697,
      "balance_loss_mlp": 1.02217782,
      "epoch": 0.3791258341850538,
      "flos": 24644667571200.0,
      "grad_norm": 3.8372193514594097,
      "language_loss": 0.81549227,
      "learning_rate": 2.852094220125805e-06,
      "loss": 0.83783442,
      "num_input_tokens_seen": 67766035,
      "step": 3153,
      "time_per_iteration": 2.5982306003570557
    },
    {
      "auxiliary_loss_clip": 0.01202932,
      "auxiliary_loss_mlp": 0.01035979,
      "balance_loss_clip": 1.06085455,
      "balance_loss_mlp": 1.02663898,
      "epoch": 0.3792460770756929,
      "flos": 17420841509760.0,
      "grad_norm": 3.0054893904861713,
      "language_loss": 0.71358466,
      "learning_rate": 2.8513894191012846e-06,
      "loss": 0.73597378,
      "num_input_tokens_seen": 67785015,
      "step": 3154,
      "time_per_iteration": 3.600705623626709
    },
    {
      "auxiliary_loss_clip": 0.0120021,
      "auxiliary_loss_mlp": 0.01034213,
      "balance_loss_clip": 1.05858755,
      "balance_loss_mlp": 1.02549267,
      "epoch": 0.37936631996633197,
      "flos": 24206557386240.0,
      "grad_norm": 1.6029036227150917,
      "language_loss": 0.78955805,
      "learning_rate": 2.8506844889216664e-06,
      "loss": 0.81190234,
      "num_input_tokens_seen": 67804400,
      "step": 3155,
      "time_per_iteration": 2.6349713802337646
    },
    {
      "auxiliary_loss_clip": 0.01059734,
      "auxiliary_loss_mlp": 0.01000211,
      "balance_loss_clip": 1.01240301,
      "balance_loss_mlp": 0.99863714,
      "epoch": 0.3794865628569711,
      "flos": 70297114752000.0,
      "grad_norm": 0.8615485943631794,
      "language_loss": 0.62808895,
      "learning_rate": 2.849979429693887e-06,
      "loss": 0.64868844,
      "num_input_tokens_seen": 67865385,
      "step": 3156,
      "time_per_iteration": 4.3233513832092285
    },
    {
      "auxiliary_loss_clip": 0.01163049,
      "auxiliary_loss_mlp": 0.01031658,
      "balance_loss_clip": 1.05308604,
      "balance_loss_mlp": 1.02179384,
      "epoch": 0.3796068057476102,
      "flos": 15779364860160.0,
      "grad_norm": 2.089975713133767,
      "language_loss": 0.74051905,
      "learning_rate": 2.8492742415249042e-06,
      "loss": 0.76246619,
      "num_input_tokens_seen": 67883030,
      "step": 3157,
      "time_per_iteration": 2.7861194610595703
    },
    {
      "auxiliary_loss_clip": 0.01157122,
      "auxiliary_loss_mlp": 0.01035558,
      "balance_loss_clip": 1.05131531,
      "balance_loss_mlp": 1.02674806,
      "epoch": 0.37972704863824924,
      "flos": 25191694771200.0,
      "grad_norm": 1.9149146706464368,
      "language_loss": 0.76793551,
      "learning_rate": 2.848568924521694e-06,
      "loss": 0.78986233,
      "num_input_tokens_seen": 67903810,
      "step": 3158,
      "time_per_iteration": 3.660961151123047
    },
    {
      "auxiliary_loss_clip": 0.01167902,
      "auxiliary_loss_mlp": 0.01032607,
      "balance_loss_clip": 1.05502725,
      "balance_loss_mlp": 1.02324903,
      "epoch": 0.37984729152888835,
      "flos": 26210372480640.0,
      "grad_norm": 1.867995831103098,
      "language_loss": 0.73744994,
      "learning_rate": 2.8478634787912526e-06,
      "loss": 0.75945508,
      "num_input_tokens_seen": 67921865,
      "step": 3159,
      "time_per_iteration": 2.754634380340576
    },
    {
      "auxiliary_loss_clip": 0.01201322,
      "auxiliary_loss_mlp": 0.01035835,
      "balance_loss_clip": 1.06040835,
      "balance_loss_mlp": 1.02679276,
      "epoch": 0.37996753441952746,
      "flos": 25629302165760.0,
      "grad_norm": 2.337748201936623,
      "language_loss": 0.76433098,
      "learning_rate": 2.847157904440596e-06,
      "loss": 0.78670257,
      "num_input_tokens_seen": 67941595,
      "step": 3160,
      "time_per_iteration": 2.7355408668518066
    },
    {
      "auxiliary_loss_clip": 0.01184644,
      "auxiliary_loss_mlp": 0.0103772,
      "balance_loss_clip": 1.05578113,
      "balance_loss_mlp": 1.02837992,
      "epoch": 0.3800877773101665,
      "flos": 20118414862080.0,
      "grad_norm": 1.6393055825131395,
      "language_loss": 0.73846644,
      "learning_rate": 2.846452201576759e-06,
      "loss": 0.76069009,
      "num_input_tokens_seen": 67960970,
      "step": 3161,
      "time_per_iteration": 2.62608003616333
    },
    {
      "auxiliary_loss_clip": 0.01076888,
      "auxiliary_loss_mlp": 0.0100313,
      "balance_loss_clip": 1.01261616,
      "balance_loss_mlp": 1.00156808,
      "epoch": 0.38020802020080563,
      "flos": 63053608037760.0,
      "grad_norm": 0.8462236465403967,
      "language_loss": 0.62767303,
      "learning_rate": 2.845746370306795e-06,
      "loss": 0.6484732,
      "num_input_tokens_seen": 68026160,
      "step": 3162,
      "time_per_iteration": 3.3194100856781006
    },
    {
      "auxiliary_loss_clip": 0.01186768,
      "auxiliary_loss_mlp": 0.0102958,
      "balance_loss_clip": 1.05619884,
      "balance_loss_mlp": 1.02103317,
      "epoch": 0.38032826309144474,
      "flos": 21288420570240.0,
      "grad_norm": 1.9951354941246588,
      "language_loss": 0.78571379,
      "learning_rate": 2.84504041073778e-06,
      "loss": 0.80787724,
      "num_input_tokens_seen": 68044575,
      "step": 3163,
      "time_per_iteration": 2.702408790588379
    },
    {
      "auxiliary_loss_clip": 0.01169049,
      "auxiliary_loss_mlp": 0.01031474,
      "balance_loss_clip": 1.05646384,
      "balance_loss_mlp": 1.02242589,
      "epoch": 0.3804485059820838,
      "flos": 18954119416320.0,
      "grad_norm": 2.1331935142437413,
      "language_loss": 0.79137152,
      "learning_rate": 2.844334322976806e-06,
      "loss": 0.81337678,
      "num_input_tokens_seen": 68064790,
      "step": 3164,
      "time_per_iteration": 2.731304883956909
    },
    {
      "auxiliary_loss_clip": 0.01192216,
      "auxiliary_loss_mlp": 0.01041829,
      "balance_loss_clip": 1.05910051,
      "balance_loss_mlp": 1.03235233,
      "epoch": 0.3805687488727229,
      "flos": 21833759831040.0,
      "grad_norm": 3.8066952790547295,
      "language_loss": 0.83582634,
      "learning_rate": 2.8436281071309866e-06,
      "loss": 0.85816681,
      "num_input_tokens_seen": 68083330,
      "step": 3165,
      "time_per_iteration": 2.71447491645813
    },
    {
      "auxiliary_loss_clip": 0.01044769,
      "auxiliary_loss_mlp": 0.01004078,
      "balance_loss_clip": 1.01156211,
      "balance_loss_mlp": 1.0024451,
      "epoch": 0.380688991763362,
      "flos": 58546209968640.0,
      "grad_norm": 0.723571818879933,
      "language_loss": 0.52947962,
      "learning_rate": 2.842921763307455e-06,
      "loss": 0.54996812,
      "num_input_tokens_seen": 68146140,
      "step": 3166,
      "time_per_iteration": 3.2843551635742188
    },
    {
      "auxiliary_loss_clip": 0.01179758,
      "auxiliary_loss_mlp": 0.0076376,
      "balance_loss_clip": 1.05637491,
      "balance_loss_mlp": 1.0010252,
      "epoch": 0.38080923465400107,
      "flos": 23799509487360.0,
      "grad_norm": 1.9196703308870668,
      "language_loss": 0.83151984,
      "learning_rate": 2.842215291613361e-06,
      "loss": 0.85095501,
      "num_input_tokens_seen": 68164520,
      "step": 3167,
      "time_per_iteration": 2.717228889465332
    },
    {
      "auxiliary_loss_clip": 0.01053546,
      "auxiliary_loss_mlp": 0.01004477,
      "balance_loss_clip": 1.01261687,
      "balance_loss_mlp": 1.00271285,
      "epoch": 0.3809294775446402,
      "flos": 54969866380800.0,
      "grad_norm": 0.7741631744950733,
      "language_loss": 0.59155935,
      "learning_rate": 2.8415086921558774e-06,
      "loss": 0.61213958,
      "num_input_tokens_seen": 68227945,
      "step": 3168,
      "time_per_iteration": 3.2604801654815674
    },
    {
      "auxiliary_loss_clip": 0.01195474,
      "auxiliary_loss_mlp": 0.01030871,
      "balance_loss_clip": 1.05570972,
      "balance_loss_mlp": 1.02208567,
      "epoch": 0.38104972043527924,
      "flos": 24643697904000.0,
      "grad_norm": 1.7966845226821713,
      "language_loss": 0.79391837,
      "learning_rate": 2.840801965042194e-06,
      "loss": 0.81618178,
      "num_input_tokens_seen": 68247405,
      "step": 3169,
      "time_per_iteration": 2.7520084381103516
    },
    {
      "auxiliary_loss_clip": 0.01184217,
      "auxiliary_loss_mlp": 0.01036232,
      "balance_loss_clip": 1.0569582,
      "balance_loss_mlp": 1.02636719,
      "epoch": 0.38116996332591835,
      "flos": 22856783086080.0,
      "grad_norm": 3.229702293645529,
      "language_loss": 0.8428486,
      "learning_rate": 2.840095110379521e-06,
      "loss": 0.86505306,
      "num_input_tokens_seen": 68266925,
      "step": 3170,
      "time_per_iteration": 2.7147247791290283
    },
    {
      "auxiliary_loss_clip": 0.01073917,
      "auxiliary_loss_mlp": 0.01000485,
      "balance_loss_clip": 1.01226723,
      "balance_loss_mlp": 0.99898911,
      "epoch": 0.38129020621655746,
      "flos": 60836160804480.0,
      "grad_norm": 0.7344565831248305,
      "language_loss": 0.53816187,
      "learning_rate": 2.8393881282750884e-06,
      "loss": 0.5589059,
      "num_input_tokens_seen": 68329755,
      "step": 3171,
      "time_per_iteration": 3.168503999710083
    },
    {
      "auxiliary_loss_clip": 0.01200245,
      "auxiliary_loss_mlp": 0.01037498,
      "balance_loss_clip": 1.05843461,
      "balance_loss_mlp": 1.02794981,
      "epoch": 0.3814104491071965,
      "flos": 21648101408640.0,
      "grad_norm": 2.806240982386814,
      "language_loss": 0.78577191,
      "learning_rate": 2.838681018836144e-06,
      "loss": 0.80814934,
      "num_input_tokens_seen": 68347075,
      "step": 3172,
      "time_per_iteration": 2.675889015197754
    },
    {
      "auxiliary_loss_clip": 0.01198704,
      "auxiliary_loss_mlp": 0.01033624,
      "balance_loss_clip": 1.05905187,
      "balance_loss_mlp": 1.02407539,
      "epoch": 0.3815306919978356,
      "flos": 19099090707840.0,
      "grad_norm": 2.2344775538808856,
      "language_loss": 0.77960068,
      "learning_rate": 2.837973782169955e-06,
      "loss": 0.80192399,
      "num_input_tokens_seen": 68365450,
      "step": 3173,
      "time_per_iteration": 2.6524717807769775
    },
    {
      "auxiliary_loss_clip": 0.01085233,
      "auxiliary_loss_mlp": 0.01004084,
      "balance_loss_clip": 1.02563608,
      "balance_loss_mlp": 1.00279641,
      "epoch": 0.38165093488847474,
      "flos": 67067918156160.0,
      "grad_norm": 0.809159641226125,
      "language_loss": 0.59090126,
      "learning_rate": 2.8372664183838096e-06,
      "loss": 0.61179441,
      "num_input_tokens_seen": 68428470,
      "step": 3174,
      "time_per_iteration": 3.298797607421875
    },
    {
      "auxiliary_loss_clip": 0.01197909,
      "auxiliary_loss_mlp": 0.01034323,
      "balance_loss_clip": 1.05913949,
      "balance_loss_mlp": 1.02453017,
      "epoch": 0.3817711777791138,
      "flos": 22341105480960.0,
      "grad_norm": 2.994228917909951,
      "language_loss": 0.68786353,
      "learning_rate": 2.836558927585015e-06,
      "loss": 0.71018589,
      "num_input_tokens_seen": 68445440,
      "step": 3175,
      "time_per_iteration": 2.6466777324676514
    },
    {
      "auxiliary_loss_clip": 0.01185405,
      "auxiliary_loss_mlp": 0.01033822,
      "balance_loss_clip": 1.05881512,
      "balance_loss_mlp": 1.02366483,
      "epoch": 0.3818914206697529,
      "flos": 22820621068800.0,
      "grad_norm": 2.45356404301641,
      "language_loss": 0.82636082,
      "learning_rate": 2.8358513098808957e-06,
      "loss": 0.84855306,
      "num_input_tokens_seen": 68465755,
      "step": 3176,
      "time_per_iteration": 2.7922024726867676
    },
    {
      "auxiliary_loss_clip": 0.01184851,
      "auxiliary_loss_mlp": 0.01042177,
      "balance_loss_clip": 1.05514157,
      "balance_loss_mlp": 1.03311086,
      "epoch": 0.382011663560392,
      "flos": 24386074583040.0,
      "grad_norm": 2.344365580391421,
      "language_loss": 0.76633608,
      "learning_rate": 2.835143565378798e-06,
      "loss": 0.78860641,
      "num_input_tokens_seen": 68486220,
      "step": 3177,
      "time_per_iteration": 2.69431209564209
    },
    {
      "auxiliary_loss_clip": 0.01198628,
      "auxiliary_loss_mlp": 0.01039749,
      "balance_loss_clip": 1.05737448,
      "balance_loss_mlp": 1.02947927,
      "epoch": 0.38213190645103107,
      "flos": 21981568296960.0,
      "grad_norm": 2.4258616718261696,
      "language_loss": 0.78527528,
      "learning_rate": 2.8344356941860847e-06,
      "loss": 0.80765903,
      "num_input_tokens_seen": 68505850,
      "step": 3178,
      "time_per_iteration": 3.630765199661255
    },
    {
      "auxiliary_loss_clip": 0.01173952,
      "auxiliary_loss_mlp": 0.01038831,
      "balance_loss_clip": 1.05422664,
      "balance_loss_mlp": 1.03043318,
      "epoch": 0.3822521493416702,
      "flos": 35516945773440.0,
      "grad_norm": 9.749325428266587,
      "language_loss": 0.66678238,
      "learning_rate": 2.8337276964101403e-06,
      "loss": 0.68891019,
      "num_input_tokens_seen": 68526290,
      "step": 3179,
      "time_per_iteration": 2.785587787628174
    },
    {
      "auxiliary_loss_clip": 0.01197867,
      "auxiliary_loss_mlp": 0.01032796,
      "balance_loss_clip": 1.05734265,
      "balance_loss_mlp": 1.02359319,
      "epoch": 0.3823723922323093,
      "flos": 21069904181760.0,
      "grad_norm": 1.836648931816595,
      "language_loss": 0.76273894,
      "learning_rate": 2.833019572158367e-06,
      "loss": 0.78504562,
      "num_input_tokens_seen": 68544725,
      "step": 3180,
      "time_per_iteration": 3.6234004497528076
    },
    {
      "auxiliary_loss_clip": 0.01184075,
      "auxiliary_loss_mlp": 0.01045262,
      "balance_loss_clip": 1.05304933,
      "balance_loss_mlp": 1.03624415,
      "epoch": 0.38249263512294834,
      "flos": 19789149864960.0,
      "grad_norm": 2.192671328103947,
      "language_loss": 0.80191779,
      "learning_rate": 2.8323113215381872e-06,
      "loss": 0.82421124,
      "num_input_tokens_seen": 68563070,
      "step": 3181,
      "time_per_iteration": 2.689228057861328
    },
    {
      "auxiliary_loss_clip": 0.01203412,
      "auxiliary_loss_mlp": 0.01044214,
      "balance_loss_clip": 1.05846238,
      "balance_loss_mlp": 1.03451657,
      "epoch": 0.38261287801358745,
      "flos": 21433930565760.0,
      "grad_norm": 2.441806818741227,
      "language_loss": 0.76207554,
      "learning_rate": 2.831602944657042e-06,
      "loss": 0.78455186,
      "num_input_tokens_seen": 68581150,
      "step": 3182,
      "time_per_iteration": 3.548389434814453
    },
    {
      "auxiliary_loss_clip": 0.01183154,
      "auxiliary_loss_mlp": 0.01028412,
      "balance_loss_clip": 1.05666471,
      "balance_loss_mlp": 1.01960289,
      "epoch": 0.38273312090422656,
      "flos": 21981568296960.0,
      "grad_norm": 2.5491656348742335,
      "language_loss": 0.7421065,
      "learning_rate": 2.830894441622391e-06,
      "loss": 0.76422215,
      "num_input_tokens_seen": 68597800,
      "step": 3183,
      "time_per_iteration": 2.6015758514404297
    },
    {
      "auxiliary_loss_clip": 0.01184865,
      "auxiliary_loss_mlp": 0.01034312,
      "balance_loss_clip": 1.05570018,
      "balance_loss_mlp": 1.02482843,
      "epoch": 0.3828533637948656,
      "flos": 24790895838720.0,
      "grad_norm": 2.4207194027868746,
      "language_loss": 0.79796875,
      "learning_rate": 2.8301858125417134e-06,
      "loss": 0.82016051,
      "num_input_tokens_seen": 68617640,
      "step": 3184,
      "time_per_iteration": 3.6438722610473633
    },
    {
      "auxiliary_loss_clip": 0.01184426,
      "auxiliary_loss_mlp": 0.01035218,
      "balance_loss_clip": 1.05565763,
      "balance_loss_mlp": 1.02658772,
      "epoch": 0.38297360668550473,
      "flos": 22455445449600.0,
      "grad_norm": 1.7932971870275292,
      "language_loss": 0.73990703,
      "learning_rate": 2.8294770575225082e-06,
      "loss": 0.76210344,
      "num_input_tokens_seen": 68637770,
      "step": 3185,
      "time_per_iteration": 2.6831228733062744
    },
    {
      "auxiliary_loss_clip": 0.01199481,
      "auxiliary_loss_mlp": 0.01033098,
      "balance_loss_clip": 1.05932224,
      "balance_loss_mlp": 1.02347815,
      "epoch": 0.3830938495761438,
      "flos": 24896903852160.0,
      "grad_norm": 2.43094858056925,
      "language_loss": 0.83829194,
      "learning_rate": 2.828768176672293e-06,
      "loss": 0.86061776,
      "num_input_tokens_seen": 68656885,
      "step": 3186,
      "time_per_iteration": 2.6545159816741943
    },
    {
      "auxiliary_loss_clip": 0.01139391,
      "auxiliary_loss_mlp": 0.01031451,
      "balance_loss_clip": 1.04937959,
      "balance_loss_mlp": 1.02129447,
      "epoch": 0.3832140924667829,
      "flos": 33036236784000.0,
      "grad_norm": 1.8556445353318378,
      "language_loss": 0.71654916,
      "learning_rate": 2.8280591700986044e-06,
      "loss": 0.73825759,
      "num_input_tokens_seen": 68678750,
      "step": 3187,
      "time_per_iteration": 2.8540003299713135
    },
    {
      "auxiliary_loss_clip": 0.01166748,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.05495334,
      "balance_loss_mlp": 1.01817,
      "epoch": 0.383334335357422,
      "flos": 31903721896320.0,
      "grad_norm": 2.3157135640228077,
      "language_loss": 0.74946094,
      "learning_rate": 2.827350037908999e-06,
      "loss": 0.77139866,
      "num_input_tokens_seen": 68698190,
      "step": 3188,
      "time_per_iteration": 2.705667734146118
    },
    {
      "auxiliary_loss_clip": 0.0118166,
      "auxiliary_loss_mlp": 0.01032537,
      "balance_loss_clip": 1.05602872,
      "balance_loss_mlp": 1.02276826,
      "epoch": 0.38345457824806106,
      "flos": 19791915212160.0,
      "grad_norm": 3.0715834216323468,
      "language_loss": 0.7929244,
      "learning_rate": 2.8266407802110496e-06,
      "loss": 0.8150664,
      "num_input_tokens_seen": 68716445,
      "step": 3189,
      "time_per_iteration": 2.637688398361206
    },
    {
      "auxiliary_loss_clip": 0.01154469,
      "auxiliary_loss_mlp": 0.01028127,
      "balance_loss_clip": 1.05365252,
      "balance_loss_mlp": 1.018525,
      "epoch": 0.3835748211387002,
      "flos": 22419391173120.0,
      "grad_norm": 2.1981895937695195,
      "language_loss": 0.76801342,
      "learning_rate": 2.8259313971123515e-06,
      "loss": 0.78983939,
      "num_input_tokens_seen": 68737565,
      "step": 3190,
      "time_per_iteration": 2.8596415519714355
    },
    {
      "auxiliary_loss_clip": 0.01162614,
      "auxiliary_loss_mlp": 0.01028404,
      "balance_loss_clip": 1.05405116,
      "balance_loss_mlp": 1.02009463,
      "epoch": 0.3836950640293393,
      "flos": 25118436983040.0,
      "grad_norm": 1.6278233613715742,
      "language_loss": 0.78837949,
      "learning_rate": 2.8252218887205166e-06,
      "loss": 0.81028968,
      "num_input_tokens_seen": 68758255,
      "step": 3191,
      "time_per_iteration": 2.75119948387146
    },
    {
      "auxiliary_loss_clip": 0.01199949,
      "auxiliary_loss_mlp": 0.01036919,
      "balance_loss_clip": 1.06032753,
      "balance_loss_mlp": 1.02772856,
      "epoch": 0.38381530691997834,
      "flos": 21799213925760.0,
      "grad_norm": 1.895510842258153,
      "language_loss": 0.80800295,
      "learning_rate": 2.824512255143178e-06,
      "loss": 0.83037168,
      "num_input_tokens_seen": 68777490,
      "step": 3192,
      "time_per_iteration": 2.636979818344116
    },
    {
      "auxiliary_loss_clip": 0.01168216,
      "auxiliary_loss_mlp": 0.01029779,
      "balance_loss_clip": 1.05679345,
      "balance_loss_mlp": 1.02066565,
      "epoch": 0.38393554981061745,
      "flos": 21252689516160.0,
      "grad_norm": 2.462125349046634,
      "language_loss": 0.79412484,
      "learning_rate": 2.8238024964879855e-06,
      "loss": 0.81610477,
      "num_input_tokens_seen": 68798385,
      "step": 3193,
      "time_per_iteration": 2.6719443798065186
    },
    {
      "auxiliary_loss_clip": 0.01204717,
      "auxiliary_loss_mlp": 0.0103622,
      "balance_loss_clip": 1.0616405,
      "balance_loss_mlp": 1.0263257,
      "epoch": 0.38405579270125656,
      "flos": 17019360218880.0,
      "grad_norm": 2.3334739537763216,
      "language_loss": 0.77455223,
      "learning_rate": 2.8230926128626095e-06,
      "loss": 0.79696155,
      "num_input_tokens_seen": 68816880,
      "step": 3194,
      "time_per_iteration": 2.542367458343506
    },
    {
      "auxiliary_loss_clip": 0.0113566,
      "auxiliary_loss_mlp": 0.01033939,
      "balance_loss_clip": 1.04918885,
      "balance_loss_mlp": 1.02419949,
      "epoch": 0.3841760355918956,
      "flos": 21835375943040.0,
      "grad_norm": 1.965856889309525,
      "language_loss": 0.7949062,
      "learning_rate": 2.822382604374738e-06,
      "loss": 0.81660217,
      "num_input_tokens_seen": 68835805,
      "step": 3195,
      "time_per_iteration": 2.8002965450286865
    },
    {
      "auxiliary_loss_clip": 0.01171293,
      "auxiliary_loss_mlp": 0.01031083,
      "balance_loss_clip": 1.06021857,
      "balance_loss_mlp": 1.02120042,
      "epoch": 0.3842962784825347,
      "flos": 25915114684800.0,
      "grad_norm": 2.956356620956062,
      "language_loss": 0.66345268,
      "learning_rate": 2.8216724711320793e-06,
      "loss": 0.68547642,
      "num_input_tokens_seen": 68854930,
      "step": 3196,
      "time_per_iteration": 2.787611484527588
    },
    {
      "auxiliary_loss_clip": 0.01181735,
      "auxiliary_loss_mlp": 0.01027283,
      "balance_loss_clip": 1.05583537,
      "balance_loss_mlp": 1.01893222,
      "epoch": 0.38441652137317384,
      "flos": 25337492075520.0,
      "grad_norm": 1.6128429177647485,
      "language_loss": 0.79254436,
      "learning_rate": 2.820962213242361e-06,
      "loss": 0.81463456,
      "num_input_tokens_seen": 68874260,
      "step": 3197,
      "time_per_iteration": 2.7243709564208984
    },
    {
      "auxiliary_loss_clip": 0.01194731,
      "auxiliary_loss_mlp": 0.01035065,
      "balance_loss_clip": 1.05724502,
      "balance_loss_mlp": 1.0264101,
      "epoch": 0.3845367642638129,
      "flos": 18113486446080.0,
      "grad_norm": 2.997003957249618,
      "language_loss": 0.84257603,
      "learning_rate": 2.8202518308133264e-06,
      "loss": 0.86487389,
      "num_input_tokens_seen": 68891535,
      "step": 3198,
      "time_per_iteration": 2.7174105644226074
    },
    {
      "auxiliary_loss_clip": 0.01172759,
      "auxiliary_loss_mlp": 0.01035317,
      "balance_loss_clip": 1.05509675,
      "balance_loss_mlp": 1.02505898,
      "epoch": 0.384657007154452,
      "flos": 25228395492480.0,
      "grad_norm": 4.336637239783318,
      "language_loss": 0.74100077,
      "learning_rate": 2.8195413239527426e-06,
      "loss": 0.76308155,
      "num_input_tokens_seen": 68911275,
      "step": 3199,
      "time_per_iteration": 2.7810490131378174
    },
    {
      "auxiliary_loss_clip": 0.01166524,
      "auxiliary_loss_mlp": 0.01032928,
      "balance_loss_clip": 1.0526818,
      "balance_loss_mlp": 1.02418947,
      "epoch": 0.38477725004509106,
      "flos": 19865855358720.0,
      "grad_norm": 3.4396633755506336,
      "language_loss": 0.80691439,
      "learning_rate": 2.8188306927683906e-06,
      "loss": 0.82890892,
      "num_input_tokens_seen": 68930745,
      "step": 3200,
      "time_per_iteration": 2.818037748336792
    },
    {
      "auxiliary_loss_clip": 0.01200884,
      "auxiliary_loss_mlp": 0.01024668,
      "balance_loss_clip": 1.05912316,
      "balance_loss_mlp": 1.01529264,
      "epoch": 0.38489749293573017,
      "flos": 18259391491200.0,
      "grad_norm": 2.2357206438017525,
      "language_loss": 0.74843484,
      "learning_rate": 2.818119937368074e-06,
      "loss": 0.77069032,
      "num_input_tokens_seen": 68949380,
      "step": 3201,
      "time_per_iteration": 2.6334621906280518
    },
    {
      "auxiliary_loss_clip": 0.011798,
      "auxiliary_loss_mlp": 0.0103587,
      "balance_loss_clip": 1.05594742,
      "balance_loss_mlp": 1.02603531,
      "epoch": 0.3850177358263693,
      "flos": 24389163152640.0,
      "grad_norm": 2.439249368430084,
      "language_loss": 0.65650213,
      "learning_rate": 2.817409057859613e-06,
      "loss": 0.67865884,
      "num_input_tokens_seen": 68968370,
      "step": 3202,
      "time_per_iteration": 2.7769715785980225
    },
    {
      "auxiliary_loss_clip": 0.01192313,
      "auxiliary_loss_mlp": 0.01039979,
      "balance_loss_clip": 1.060009,
      "balance_loss_mlp": 1.03040016,
      "epoch": 0.38513797871700833,
      "flos": 17671533505920.0,
      "grad_norm": 2.004669884760494,
      "language_loss": 0.79255295,
      "learning_rate": 2.8166980543508482e-06,
      "loss": 0.81487584,
      "num_input_tokens_seen": 68984260,
      "step": 3203,
      "time_per_iteration": 2.646860361099243
    },
    {
      "auxiliary_loss_clip": 0.01186564,
      "auxiliary_loss_mlp": 0.01038146,
      "balance_loss_clip": 1.05782866,
      "balance_loss_mlp": 1.02838314,
      "epoch": 0.38525822160764744,
      "flos": 25739583897600.0,
      "grad_norm": 2.2942671165337267,
      "language_loss": 0.79998279,
      "learning_rate": 2.815986926949638e-06,
      "loss": 0.82222986,
      "num_input_tokens_seen": 69002760,
      "step": 3204,
      "time_per_iteration": 3.6438961029052734
    },
    {
      "auxiliary_loss_clip": 0.01185085,
      "auxiliary_loss_mlp": 0.01033306,
      "balance_loss_clip": 1.0597533,
      "balance_loss_mlp": 1.02439475,
      "epoch": 0.38537846449828655,
      "flos": 20193647898240.0,
      "grad_norm": 2.8795841652212957,
      "language_loss": 0.806481,
      "learning_rate": 2.8152756757638597e-06,
      "loss": 0.8286649,
      "num_input_tokens_seen": 69021260,
      "step": 3205,
      "time_per_iteration": 2.771404504776001
    },
    {
      "auxiliary_loss_clip": 0.01202535,
      "auxiliary_loss_mlp": 0.01036226,
      "balance_loss_clip": 1.06065094,
      "balance_loss_mlp": 1.02680826,
      "epoch": 0.3854987073889256,
      "flos": 23039352938880.0,
      "grad_norm": 2.5582151614758537,
      "language_loss": 0.84626406,
      "learning_rate": 2.8145643009014093e-06,
      "loss": 0.86865163,
      "num_input_tokens_seen": 69039755,
      "step": 3206,
      "time_per_iteration": 3.5635738372802734
    },
    {
      "auxiliary_loss_clip": 0.01199537,
      "auxiliary_loss_mlp": 0.01034992,
      "balance_loss_clip": 1.06092882,
      "balance_loss_mlp": 1.02625954,
      "epoch": 0.3856189502795647,
      "flos": 20190631155840.0,
      "grad_norm": 2.077120365027484,
      "language_loss": 0.79270607,
      "learning_rate": 2.813852802470202e-06,
      "loss": 0.81505132,
      "num_input_tokens_seen": 69057650,
      "step": 3207,
      "time_per_iteration": 2.6604511737823486
    },
    {
      "auxiliary_loss_clip": 0.01176221,
      "auxiliary_loss_mlp": 0.01036499,
      "balance_loss_clip": 1.05595446,
      "balance_loss_mlp": 1.02664685,
      "epoch": 0.38573919317020383,
      "flos": 25702631781120.0,
      "grad_norm": 2.119793519366054,
      "language_loss": 0.72648454,
      "learning_rate": 2.8131411805781717e-06,
      "loss": 0.74861169,
      "num_input_tokens_seen": 69077775,
      "step": 3208,
      "time_per_iteration": 3.700183391571045
    },
    {
      "auxiliary_loss_clip": 0.01128179,
      "auxiliary_loss_mlp": 0.01035328,
      "balance_loss_clip": 1.05745661,
      "balance_loss_mlp": 1.0251056,
      "epoch": 0.3858594360608429,
      "flos": 29821405628160.0,
      "grad_norm": 2.7745298880005516,
      "language_loss": 0.64583206,
      "learning_rate": 2.8124294353332707e-06,
      "loss": 0.66746712,
      "num_input_tokens_seen": 69096450,
      "step": 3209,
      "time_per_iteration": 2.9137911796569824
    },
    {
      "auxiliary_loss_clip": 0.01199394,
      "auxiliary_loss_mlp": 0.0076365,
      "balance_loss_clip": 1.06053495,
      "balance_loss_mlp": 1.00114536,
      "epoch": 0.385979678951482,
      "flos": 24790428961920.0,
      "grad_norm": 3.55409947272924,
      "language_loss": 0.77667475,
      "learning_rate": 2.8117175668434713e-06,
      "loss": 0.79630512,
      "num_input_tokens_seen": 69116110,
      "step": 3210,
      "time_per_iteration": 2.7448806762695312
    },
    {
      "auxiliary_loss_clip": 0.01167205,
      "auxiliary_loss_mlp": 0.01038167,
      "balance_loss_clip": 1.0548048,
      "balance_loss_mlp": 1.0288744,
      "epoch": 0.3860999218421211,
      "flos": 21287881866240.0,
      "grad_norm": 2.9620246576873606,
      "language_loss": 0.70277339,
      "learning_rate": 2.811005575216762e-06,
      "loss": 0.72482705,
      "num_input_tokens_seen": 69134825,
      "step": 3211,
      "time_per_iteration": 3.6517248153686523
    },
    {
      "auxiliary_loss_clip": 0.01200944,
      "auxiliary_loss_mlp": 0.0103777,
      "balance_loss_clip": 1.06029856,
      "balance_loss_mlp": 1.02782226,
      "epoch": 0.38622016473276016,
      "flos": 24536720223360.0,
      "grad_norm": 1.5478560907575678,
      "language_loss": 0.78927511,
      "learning_rate": 2.8102934605611513e-06,
      "loss": 0.81166226,
      "num_input_tokens_seen": 69156460,
      "step": 3212,
      "time_per_iteration": 2.685499668121338
    },
    {
      "auxiliary_loss_clip": 0.01197741,
      "auxiliary_loss_mlp": 0.01033794,
      "balance_loss_clip": 1.06426215,
      "balance_loss_mlp": 1.02460909,
      "epoch": 0.3863404076233993,
      "flos": 20558212986240.0,
      "grad_norm": 2.763559191387131,
      "language_loss": 0.67795283,
      "learning_rate": 2.8095812229846665e-06,
      "loss": 0.70026815,
      "num_input_tokens_seen": 69176420,
      "step": 3213,
      "time_per_iteration": 2.6975529193878174
    },
    {
      "auxiliary_loss_clip": 0.0119432,
      "auxiliary_loss_mlp": 0.0104323,
      "balance_loss_clip": 1.05880153,
      "balance_loss_mlp": 1.03326988,
      "epoch": 0.3864606505140384,
      "flos": 22346277039360.0,
      "grad_norm": 4.1842781333176235,
      "language_loss": 0.6844666,
      "learning_rate": 2.808868862595355e-06,
      "loss": 0.70684206,
      "num_input_tokens_seen": 69196665,
      "step": 3214,
      "time_per_iteration": 2.687955379486084
    },
    {
      "auxiliary_loss_clip": 0.01192288,
      "auxiliary_loss_mlp": 0.0103813,
      "balance_loss_clip": 1.06277132,
      "balance_loss_mlp": 1.02878404,
      "epoch": 0.38658089340467744,
      "flos": 25703601448320.0,
      "grad_norm": 1.9801410190061766,
      "language_loss": 0.79520357,
      "learning_rate": 2.8081563795012795e-06,
      "loss": 0.81750774,
      "num_input_tokens_seen": 69216290,
      "step": 3215,
      "time_per_iteration": 2.720853328704834
    },
    {
      "auxiliary_loss_clip": 0.01202568,
      "auxiliary_loss_mlp": 0.01034287,
      "balance_loss_clip": 1.06076622,
      "balance_loss_mlp": 1.02449405,
      "epoch": 0.38670113629531655,
      "flos": 33802534558080.0,
      "grad_norm": 2.2288813543094506,
      "language_loss": 0.73775887,
      "learning_rate": 2.807443773810524e-06,
      "loss": 0.76012743,
      "num_input_tokens_seen": 69237550,
      "step": 3216,
      "time_per_iteration": 2.726287364959717
    },
    {
      "auxiliary_loss_clip": 0.01203752,
      "auxiliary_loss_mlp": 0.01039116,
      "balance_loss_clip": 1.06133032,
      "balance_loss_mlp": 1.0297699,
      "epoch": 0.3868213791859556,
      "flos": 23331522165120.0,
      "grad_norm": 4.151389208898569,
      "language_loss": 0.89805102,
      "learning_rate": 2.80673104563119e-06,
      "loss": 0.92047971,
      "num_input_tokens_seen": 69258175,
      "step": 3217,
      "time_per_iteration": 2.687739133834839
    },
    {
      "auxiliary_loss_clip": 0.01199262,
      "auxiliary_loss_mlp": 0.01031029,
      "balance_loss_clip": 1.06061006,
      "balance_loss_mlp": 1.02208209,
      "epoch": 0.3869416220765947,
      "flos": 18441530380800.0,
      "grad_norm": 2.699708198130961,
      "language_loss": 0.79487973,
      "learning_rate": 2.8060181950713976e-06,
      "loss": 0.8171826,
      "num_input_tokens_seen": 69274965,
      "step": 3218,
      "time_per_iteration": 2.5995032787323
    },
    {
      "auxiliary_loss_clip": 0.01194063,
      "auxiliary_loss_mlp": 0.01039355,
      "balance_loss_clip": 1.06189144,
      "balance_loss_mlp": 1.0297823,
      "epoch": 0.3870618649672338,
      "flos": 15632992938240.0,
      "grad_norm": 4.874014689653251,
      "language_loss": 0.81261742,
      "learning_rate": 2.805305222239286e-06,
      "loss": 0.83495164,
      "num_input_tokens_seen": 69292220,
      "step": 3219,
      "time_per_iteration": 2.684563159942627
    },
    {
      "auxiliary_loss_clip": 0.01199258,
      "auxiliary_loss_mlp": 0.01039932,
      "balance_loss_clip": 1.06067753,
      "balance_loss_mlp": 1.03053236,
      "epoch": 0.3871821078578729,
      "flos": 23513804709120.0,
      "grad_norm": 2.2359600715596772,
      "language_loss": 0.74129701,
      "learning_rate": 2.8045921272430118e-06,
      "loss": 0.76368892,
      "num_input_tokens_seen": 69311900,
      "step": 3220,
      "time_per_iteration": 2.6139183044433594
    },
    {
      "auxiliary_loss_clip": 0.01156306,
      "auxiliary_loss_mlp": 0.01030453,
      "balance_loss_clip": 1.05198979,
      "balance_loss_mlp": 1.02000451,
      "epoch": 0.387302350748512,
      "flos": 17778259791360.0,
      "grad_norm": 6.473343200394442,
      "language_loss": 0.77077198,
      "learning_rate": 2.803878910190753e-06,
      "loss": 0.79263961,
      "num_input_tokens_seen": 69328820,
      "step": 3221,
      "time_per_iteration": 2.7293622493743896
    },
    {
      "auxiliary_loss_clip": 0.01203477,
      "auxiliary_loss_mlp": 0.01035308,
      "balance_loss_clip": 1.06176829,
      "balance_loss_mlp": 1.02615333,
      "epoch": 0.3874225936391511,
      "flos": 11503409097600.0,
      "grad_norm": 2.5488200548549567,
      "language_loss": 0.82463837,
      "learning_rate": 2.8031655711907017e-06,
      "loss": 0.84702629,
      "num_input_tokens_seen": 69342525,
      "step": 3222,
      "time_per_iteration": 2.5439200401306152
    },
    {
      "auxiliary_loss_clip": 0.01203107,
      "auxiliary_loss_mlp": 0.01026561,
      "balance_loss_clip": 1.06127524,
      "balance_loss_mlp": 1.01709628,
      "epoch": 0.38754283652979016,
      "flos": 21945154884480.0,
      "grad_norm": 2.8095671626175136,
      "language_loss": 0.80875444,
      "learning_rate": 2.8024521103510723e-06,
      "loss": 0.83105111,
      "num_input_tokens_seen": 69359295,
      "step": 3223,
      "time_per_iteration": 2.619593858718872
    },
    {
      "auxiliary_loss_clip": 0.01202948,
      "auxiliary_loss_mlp": 0.01036442,
      "balance_loss_clip": 1.06094682,
      "balance_loss_mlp": 1.02672029,
      "epoch": 0.38766307942042927,
      "flos": 21175984022400.0,
      "grad_norm": 2.2165725565030807,
      "language_loss": 0.75265634,
      "learning_rate": 2.8017385277800952e-06,
      "loss": 0.77505028,
      "num_input_tokens_seen": 69377650,
      "step": 3224,
      "time_per_iteration": 2.6472818851470947
    },
    {
      "auxiliary_loss_clip": 0.01194548,
      "auxiliary_loss_mlp": 0.01043389,
      "balance_loss_clip": 1.05948472,
      "balance_loss_mlp": 1.03280926,
      "epoch": 0.3877833223110684,
      "flos": 27417294391680.0,
      "grad_norm": 2.857565674216766,
      "language_loss": 0.75443816,
      "learning_rate": 2.8010248235860213e-06,
      "loss": 0.77681756,
      "num_input_tokens_seen": 69397765,
      "step": 3225,
      "time_per_iteration": 2.721184015274048
    },
    {
      "auxiliary_loss_clip": 0.01075788,
      "auxiliary_loss_mlp": 0.01004367,
      "balance_loss_clip": 1.01648033,
      "balance_loss_mlp": 1.00291276,
      "epoch": 0.38790356520170743,
      "flos": 64500019879680.0,
      "grad_norm": 0.843603567181157,
      "language_loss": 0.6273365,
      "learning_rate": 2.8003109978771192e-06,
      "loss": 0.64813805,
      "num_input_tokens_seen": 69458930,
      "step": 3226,
      "time_per_iteration": 3.2606823444366455
    },
    {
      "auxiliary_loss_clip": 0.01201531,
      "auxiliary_loss_mlp": 0.01034919,
      "balance_loss_clip": 1.06057215,
      "balance_loss_mlp": 1.02636528,
      "epoch": 0.38802380809234654,
      "flos": 22345415112960.0,
      "grad_norm": 2.3294610992565934,
      "language_loss": 0.79245609,
      "learning_rate": 2.799597050761674e-06,
      "loss": 0.81482053,
      "num_input_tokens_seen": 69475135,
      "step": 3227,
      "time_per_iteration": 2.5681183338165283
    },
    {
      "auxiliary_loss_clip": 0.01192948,
      "auxiliary_loss_mlp": 0.01037829,
      "balance_loss_clip": 1.0626893,
      "balance_loss_mlp": 1.02903748,
      "epoch": 0.38814405098298566,
      "flos": 25261361199360.0,
      "grad_norm": 4.655803117792449,
      "language_loss": 0.79401267,
      "learning_rate": 2.7988829823479924e-06,
      "loss": 0.81632042,
      "num_input_tokens_seen": 69493525,
      "step": 3228,
      "time_per_iteration": 2.6834962368011475
    },
    {
      "auxiliary_loss_clip": 0.0119214,
      "auxiliary_loss_mlp": 0.01037713,
      "balance_loss_clip": 1.05910265,
      "balance_loss_mlp": 1.02883172,
      "epoch": 0.3882642938736247,
      "flos": 18841180078080.0,
      "grad_norm": 2.1071040134413983,
      "language_loss": 0.64194083,
      "learning_rate": 2.7981687927443976e-06,
      "loss": 0.66423929,
      "num_input_tokens_seen": 69510325,
      "step": 3229,
      "time_per_iteration": 2.5855226516723633
    },
    {
      "auxiliary_loss_clip": 0.0116488,
      "auxiliary_loss_mlp": 0.01034883,
      "balance_loss_clip": 1.05314243,
      "balance_loss_mlp": 1.02625847,
      "epoch": 0.3883845367642638,
      "flos": 21652806090240.0,
      "grad_norm": 2.1868914603662604,
      "language_loss": 0.8555131,
      "learning_rate": 2.797454482059231e-06,
      "loss": 0.87751079,
      "num_input_tokens_seen": 69530480,
      "step": 3230,
      "time_per_iteration": 3.6857690811157227
    },
    {
      "auxiliary_loss_clip": 0.01154823,
      "auxiliary_loss_mlp": 0.01034704,
      "balance_loss_clip": 1.05749702,
      "balance_loss_mlp": 1.02588868,
      "epoch": 0.3885047796549029,
      "flos": 20557530627840.0,
      "grad_norm": 1.7833195744752315,
      "language_loss": 0.84316754,
      "learning_rate": 2.7967400504008537e-06,
      "loss": 0.86506277,
      "num_input_tokens_seen": 69549780,
      "step": 3231,
      "time_per_iteration": 2.7211666107177734
    },
    {
      "auxiliary_loss_clip": 0.01079512,
      "auxiliary_loss_mlp": 0.01003437,
      "balance_loss_clip": 1.01496267,
      "balance_loss_mlp": 1.00201821,
      "epoch": 0.388625022545542,
      "flos": 64325491695360.0,
      "grad_norm": 0.9326098591328296,
      "language_loss": 0.57398546,
      "learning_rate": 2.7960254978776456e-06,
      "loss": 0.59481502,
      "num_input_tokens_seen": 69611870,
      "step": 3232,
      "time_per_iteration": 4.2038116455078125
    },
    {
      "auxiliary_loss_clip": 0.0120437,
      "auxiliary_loss_mlp": 0.01041552,
      "balance_loss_clip": 1.0626229,
      "balance_loss_mlp": 1.03217614,
      "epoch": 0.3887452654361811,
      "flos": 18113881495680.0,
      "grad_norm": 2.152867241519223,
      "language_loss": 0.81819642,
      "learning_rate": 2.7953108245980006e-06,
      "loss": 0.84065568,
      "num_input_tokens_seen": 69630385,
      "step": 3233,
      "time_per_iteration": 2.6204261779785156
    },
    {
      "auxiliary_loss_clip": 0.01184965,
      "auxiliary_loss_mlp": 0.0103512,
      "balance_loss_clip": 1.0588429,
      "balance_loss_mlp": 1.02574432,
      "epoch": 0.38886550832682015,
      "flos": 24975261371520.0,
      "grad_norm": 2.2115289144986985,
      "language_loss": 0.741,
      "learning_rate": 2.7945960306703365e-06,
      "loss": 0.76320088,
      "num_input_tokens_seen": 69653370,
      "step": 3234,
      "time_per_iteration": 3.596292734146118
    },
    {
      "auxiliary_loss_clip": 0.01173861,
      "auxiliary_loss_mlp": 0.01033557,
      "balance_loss_clip": 1.05513692,
      "balance_loss_mlp": 1.02433014,
      "epoch": 0.38898575121745926,
      "flos": 27199496275200.0,
      "grad_norm": 1.9626971577031407,
      "language_loss": 0.65927756,
      "learning_rate": 2.7938811162030865e-06,
      "loss": 0.68135172,
      "num_input_tokens_seen": 69673635,
      "step": 3235,
      "time_per_iteration": 2.7179818153381348
    },
    {
      "auxiliary_loss_clip": 0.0116917,
      "auxiliary_loss_mlp": 0.0103188,
      "balance_loss_clip": 1.05553985,
      "balance_loss_mlp": 1.02347612,
      "epoch": 0.3891059941080984,
      "flos": 28763728727040.0,
      "grad_norm": 1.745930778956733,
      "language_loss": 0.82271373,
      "learning_rate": 2.793166081304702e-06,
      "loss": 0.84472418,
      "num_input_tokens_seen": 69694130,
      "step": 3236,
      "time_per_iteration": 2.780388116836548
    },
    {
      "auxiliary_loss_clip": 0.01190809,
      "auxiliary_loss_mlp": 0.01034817,
      "balance_loss_clip": 1.05744338,
      "balance_loss_mlp": 1.02593601,
      "epoch": 0.38922623699873743,
      "flos": 22893447893760.0,
      "grad_norm": 2.1230746354522467,
      "language_loss": 0.82912302,
      "learning_rate": 2.7924509260836543e-06,
      "loss": 0.85137933,
      "num_input_tokens_seen": 69713255,
      "step": 3237,
      "time_per_iteration": 3.561662435531616
    },
    {
      "auxiliary_loss_clip": 0.01175605,
      "auxiliary_loss_mlp": 0.01036839,
      "balance_loss_clip": 1.05811203,
      "balance_loss_mlp": 1.02825034,
      "epoch": 0.38934647988937654,
      "flos": 19792418002560.0,
      "grad_norm": 1.6257096261659005,
      "language_loss": 0.68620813,
      "learning_rate": 2.791735650648431e-06,
      "loss": 0.7083326,
      "num_input_tokens_seen": 69732375,
      "step": 3238,
      "time_per_iteration": 2.6726627349853516
    },
    {
      "auxiliary_loss_clip": 0.01185691,
      "auxiliary_loss_mlp": 0.01030377,
      "balance_loss_clip": 1.05642414,
      "balance_loss_mlp": 1.02133465,
      "epoch": 0.38946672278001565,
      "flos": 19202081978880.0,
      "grad_norm": 3.7515747805400097,
      "language_loss": 0.7466346,
      "learning_rate": 2.791020255107538e-06,
      "loss": 0.76879525,
      "num_input_tokens_seen": 69749745,
      "step": 3239,
      "time_per_iteration": 2.637683629989624
    },
    {
      "auxiliary_loss_clip": 0.01172634,
      "auxiliary_loss_mlp": 0.01032098,
      "balance_loss_clip": 1.05584502,
      "balance_loss_mlp": 1.02367043,
      "epoch": 0.3895869656706547,
      "flos": 24936477661440.0,
      "grad_norm": 4.631643072223411,
      "language_loss": 0.80739659,
      "learning_rate": 2.7903047395695023e-06,
      "loss": 0.82944393,
      "num_input_tokens_seen": 69769645,
      "step": 3240,
      "time_per_iteration": 2.7303719520568848
    },
    {
      "auxiliary_loss_clip": 0.0120079,
      "auxiliary_loss_mlp": 0.01031331,
      "balance_loss_clip": 1.06080544,
      "balance_loss_mlp": 1.02224731,
      "epoch": 0.3897072085612938,
      "flos": 24133622820480.0,
      "grad_norm": 5.183882428374234,
      "language_loss": 0.8991552,
      "learning_rate": 2.789589104142865e-06,
      "loss": 0.92147636,
      "num_input_tokens_seen": 69787270,
      "step": 3241,
      "time_per_iteration": 2.6513476371765137
    },
    {
      "auxiliary_loss_clip": 0.01185663,
      "auxiliary_loss_mlp": 0.01037385,
      "balance_loss_clip": 1.05600357,
      "balance_loss_mlp": 1.02824211,
      "epoch": 0.3898274514519329,
      "flos": 17166342672000.0,
      "grad_norm": 1.8133661143457795,
      "language_loss": 0.76980865,
      "learning_rate": 2.7888733489361895e-06,
      "loss": 0.79203916,
      "num_input_tokens_seen": 69805685,
      "step": 3242,
      "time_per_iteration": 2.6401960849761963
    },
    {
      "auxiliary_loss_clip": 0.01087334,
      "auxiliary_loss_mlp": 0.01002504,
      "balance_loss_clip": 1.01427889,
      "balance_loss_mlp": 1.00106716,
      "epoch": 0.389947694342572,
      "flos": 66074807952000.0,
      "grad_norm": 0.7356592534116196,
      "language_loss": 0.5863362,
      "learning_rate": 2.788157474058054e-06,
      "loss": 0.60723454,
      "num_input_tokens_seen": 69867960,
      "step": 3243,
      "time_per_iteration": 3.2581639289855957
    },
    {
      "auxiliary_loss_clip": 0.01165879,
      "auxiliary_loss_mlp": 0.00762721,
      "balance_loss_clip": 1.05326772,
      "balance_loss_mlp": 1.00110006,
      "epoch": 0.3900679372332111,
      "flos": 25740912700800.0,
      "grad_norm": 1.81007190743988,
      "language_loss": 0.7035321,
      "learning_rate": 2.7874414796170555e-06,
      "loss": 0.72281802,
      "num_input_tokens_seen": 69889450,
      "step": 3244,
      "time_per_iteration": 2.755519390106201
    },
    {
      "auxiliary_loss_clip": 0.01203246,
      "auxiliary_loss_mlp": 0.01037419,
      "balance_loss_clip": 1.06174076,
      "balance_loss_mlp": 1.02801394,
      "epoch": 0.3901881801238502,
      "flos": 11801611808640.0,
      "grad_norm": 3.138110628427604,
      "language_loss": 0.83724821,
      "learning_rate": 2.7867253657218113e-06,
      "loss": 0.8596549,
      "num_input_tokens_seen": 69903340,
      "step": 3245,
      "time_per_iteration": 2.616609811782837
    },
    {
      "auxiliary_loss_clip": 0.01188051,
      "auxiliary_loss_mlp": 0.01032493,
      "balance_loss_clip": 1.05572605,
      "balance_loss_mlp": 1.02297974,
      "epoch": 0.39030842301448926,
      "flos": 27308951994240.0,
      "grad_norm": 2.0031463384973223,
      "language_loss": 0.73417246,
      "learning_rate": 2.7860091324809544e-06,
      "loss": 0.75637794,
      "num_input_tokens_seen": 69924400,
      "step": 3246,
      "time_per_iteration": 2.6831867694854736
    },
    {
      "auxiliary_loss_clip": 0.01182766,
      "auxiliary_loss_mlp": 0.01035546,
      "balance_loss_clip": 1.05926275,
      "balance_loss_mlp": 1.02710593,
      "epoch": 0.39042866590512837,
      "flos": 27163334257920.0,
      "grad_norm": 2.063910871395198,
      "language_loss": 0.81171471,
      "learning_rate": 2.7852927800031377e-06,
      "loss": 0.83389783,
      "num_input_tokens_seen": 69944565,
      "step": 3247,
      "time_per_iteration": 2.7509429454803467
    },
    {
      "auxiliary_loss_clip": 0.01185109,
      "auxiliary_loss_mlp": 0.01028506,
      "balance_loss_clip": 1.05907464,
      "balance_loss_mlp": 1.01978552,
      "epoch": 0.3905489087957674,
      "flos": 29716115886720.0,
      "grad_norm": 2.838231642025726,
      "language_loss": 0.83130574,
      "learning_rate": 2.7845763083970298e-06,
      "loss": 0.85344183,
      "num_input_tokens_seen": 69964965,
      "step": 3248,
      "time_per_iteration": 2.706890344619751
    },
    {
      "auxiliary_loss_clip": 0.0117566,
      "auxiliary_loss_mlp": 0.01029324,
      "balance_loss_clip": 1.05753219,
      "balance_loss_mlp": 1.0205915,
      "epoch": 0.39066915168640653,
      "flos": 24498618871680.0,
      "grad_norm": 2.8938931980451925,
      "language_loss": 0.82299191,
      "learning_rate": 2.7838597177713205e-06,
      "loss": 0.84504169,
      "num_input_tokens_seen": 69986055,
      "step": 3249,
      "time_per_iteration": 2.766413927078247
    },
    {
      "auxiliary_loss_clip": 0.01201287,
      "auxiliary_loss_mlp": 0.01037602,
      "balance_loss_clip": 1.06118536,
      "balance_loss_mlp": 1.02850628,
      "epoch": 0.39078939457704565,
      "flos": 20558572122240.0,
      "grad_norm": 2.2625384605976806,
      "language_loss": 0.73952186,
      "learning_rate": 2.7831430082347143e-06,
      "loss": 0.7619108,
      "num_input_tokens_seen": 70005260,
      "step": 3250,
      "time_per_iteration": 2.70223069190979
    },
    {
      "auxiliary_loss_clip": 0.0114917,
      "auxiliary_loss_mlp": 0.01031516,
      "balance_loss_clip": 1.05696821,
      "balance_loss_mlp": 1.02331984,
      "epoch": 0.3909096374676847,
      "flos": 22783417557120.0,
      "grad_norm": 2.3258590792584144,
      "language_loss": 0.82888168,
      "learning_rate": 2.7824261798959373e-06,
      "loss": 0.85068858,
      "num_input_tokens_seen": 70023440,
      "step": 3251,
      "time_per_iteration": 2.7474076747894287
    },
    {
      "auxiliary_loss_clip": 0.01176391,
      "auxiliary_loss_mlp": 0.01031996,
      "balance_loss_clip": 1.05622721,
      "balance_loss_mlp": 1.02293038,
      "epoch": 0.3910298803583238,
      "flos": 23003119094400.0,
      "grad_norm": 2.7762795255115527,
      "language_loss": 0.7999537,
      "learning_rate": 2.78170923286373e-06,
      "loss": 0.82203758,
      "num_input_tokens_seen": 70043040,
      "step": 3252,
      "time_per_iteration": 2.7095937728881836
    },
    {
      "auxiliary_loss_clip": 0.0117413,
      "auxiliary_loss_mlp": 0.01039993,
      "balance_loss_clip": 1.05494046,
      "balance_loss_mlp": 1.0304985,
      "epoch": 0.3911501232489629,
      "flos": 24316264500480.0,
      "grad_norm": 2.874358220821667,
      "language_loss": 0.83823359,
      "learning_rate": 2.780992167246854e-06,
      "loss": 0.86037481,
      "num_input_tokens_seen": 70060565,
      "step": 3253,
      "time_per_iteration": 2.62111234664917
    },
    {
      "auxiliary_loss_clip": 0.01071249,
      "auxiliary_loss_mlp": 0.01005087,
      "balance_loss_clip": 1.01407957,
      "balance_loss_mlp": 1.00372815,
      "epoch": 0.391270366139602,
      "flos": 60869054684160.0,
      "grad_norm": 0.9918961927517767,
      "language_loss": 0.72126871,
      "learning_rate": 2.7802749831540883e-06,
      "loss": 0.74203211,
      "num_input_tokens_seen": 70119465,
      "step": 3254,
      "time_per_iteration": 3.2277419567108154
    },
    {
      "auxiliary_loss_clip": 0.01183115,
      "auxiliary_loss_mlp": 0.01026808,
      "balance_loss_clip": 1.05959535,
      "balance_loss_mlp": 1.01896405,
      "epoch": 0.3913906090302411,
      "flos": 21543494025600.0,
      "grad_norm": 2.270193538796485,
      "language_loss": 0.82122958,
      "learning_rate": 2.7795576806942268e-06,
      "loss": 0.84332883,
      "num_input_tokens_seen": 70138270,
      "step": 3255,
      "time_per_iteration": 2.7338428497314453
    },
    {
      "auxiliary_loss_clip": 0.01075086,
      "auxiliary_loss_mlp": 0.01000477,
      "balance_loss_clip": 1.01243818,
      "balance_loss_mlp": 0.99916565,
      "epoch": 0.3915108519208802,
      "flos": 49839953702400.0,
      "grad_norm": 0.7868517990006785,
      "language_loss": 0.54834002,
      "learning_rate": 2.778840259976085e-06,
      "loss": 0.56909561,
      "num_input_tokens_seen": 70193500,
      "step": 3256,
      "time_per_iteration": 4.082160949707031
    },
    {
      "auxiliary_loss_clip": 0.01202525,
      "auxiliary_loss_mlp": 0.01034879,
      "balance_loss_clip": 1.06143451,
      "balance_loss_mlp": 1.02594411,
      "epoch": 0.39163109481151925,
      "flos": 16506447960960.0,
      "grad_norm": 2.4387963088724653,
      "language_loss": 0.77208352,
      "learning_rate": 2.778122721108495e-06,
      "loss": 0.79445755,
      "num_input_tokens_seen": 70211730,
      "step": 3257,
      "time_per_iteration": 2.6272835731506348
    },
    {
      "auxiliary_loss_clip": 0.01187479,
      "auxiliary_loss_mlp": 0.01031923,
      "balance_loss_clip": 1.05902863,
      "balance_loss_mlp": 1.02331018,
      "epoch": 0.39175133770215836,
      "flos": 26067484177920.0,
      "grad_norm": 2.9439156637562753,
      "language_loss": 0.88485122,
      "learning_rate": 2.7774050642003076e-06,
      "loss": 0.90704525,
      "num_input_tokens_seen": 70232540,
      "step": 3258,
      "time_per_iteration": 3.7109973430633545
    },
    {
      "auxiliary_loss_clip": 0.01178902,
      "auxiliary_loss_mlp": 0.01031998,
      "balance_loss_clip": 1.05860305,
      "balance_loss_mlp": 1.02225876,
      "epoch": 0.3918715805927975,
      "flos": 21872076664320.0,
      "grad_norm": 5.358443130723839,
      "language_loss": 0.93680441,
      "learning_rate": 2.7766872893603896e-06,
      "loss": 0.95891345,
      "num_input_tokens_seen": 70252515,
      "step": 3259,
      "time_per_iteration": 2.7581770420074463
    },
    {
      "auxiliary_loss_clip": 0.01197469,
      "auxiliary_loss_mlp": 0.01034586,
      "balance_loss_clip": 1.05987501,
      "balance_loss_mlp": 1.02571082,
      "epoch": 0.39199182348343653,
      "flos": 20376181837440.0,
      "grad_norm": 1.7884146082302685,
      "language_loss": 0.73426962,
      "learning_rate": 2.7759693966976275e-06,
      "loss": 0.75659019,
      "num_input_tokens_seen": 70271020,
      "step": 3260,
      "time_per_iteration": 3.551630735397339
    },
    {
      "auxiliary_loss_clip": 0.01203652,
      "auxiliary_loss_mlp": 0.01034418,
      "balance_loss_clip": 1.0611527,
      "balance_loss_mlp": 1.02507198,
      "epoch": 0.39211206637407564,
      "flos": 21683545153920.0,
      "grad_norm": 2.080299337466728,
      "language_loss": 0.850896,
      "learning_rate": 2.7752513863209242e-06,
      "loss": 0.87327671,
      "num_input_tokens_seen": 70289600,
      "step": 3261,
      "time_per_iteration": 2.718123435974121
    },
    {
      "auxiliary_loss_clip": 0.01187735,
      "auxiliary_loss_mlp": 0.01038835,
      "balance_loss_clip": 1.06008399,
      "balance_loss_mlp": 1.03021002,
      "epoch": 0.39223230926471475,
      "flos": 21066276908160.0,
      "grad_norm": 1.6676241076838418,
      "language_loss": 0.84394825,
      "learning_rate": 2.774533258339203e-06,
      "loss": 0.86621392,
      "num_input_tokens_seen": 70307060,
      "step": 3262,
      "time_per_iteration": 2.7573835849761963
    },
    {
      "auxiliary_loss_clip": 0.01186118,
      "auxiliary_loss_mlp": 0.01039593,
      "balance_loss_clip": 1.0586946,
      "balance_loss_mlp": 1.03069389,
      "epoch": 0.3923525521553538,
      "flos": 17603016312960.0,
      "grad_norm": 2.961287185927822,
      "language_loss": 0.80093658,
      "learning_rate": 2.7738150128614014e-06,
      "loss": 0.82319367,
      "num_input_tokens_seen": 70324465,
      "step": 3263,
      "time_per_iteration": 2.633169174194336
    },
    {
      "auxiliary_loss_clip": 0.01184268,
      "auxiliary_loss_mlp": 0.01034174,
      "balance_loss_clip": 1.05743623,
      "balance_loss_mlp": 1.02513206,
      "epoch": 0.3924727950459929,
      "flos": 20558284813440.0,
      "grad_norm": 2.6465098979091364,
      "language_loss": 0.89780396,
      "learning_rate": 2.7730966499964777e-06,
      "loss": 0.91998839,
      "num_input_tokens_seen": 70341415,
      "step": 3264,
      "time_per_iteration": 3.5577127933502197
    },
    {
      "auxiliary_loss_clip": 0.01179306,
      "auxiliary_loss_mlp": 0.01033689,
      "balance_loss_clip": 1.05296409,
      "balance_loss_mlp": 1.02453935,
      "epoch": 0.39259303793663197,
      "flos": 16216110328320.0,
      "grad_norm": 2.8336938788156094,
      "language_loss": 0.81348813,
      "learning_rate": 2.772378169853408e-06,
      "loss": 0.83561802,
      "num_input_tokens_seen": 70358985,
      "step": 3265,
      "time_per_iteration": 2.7028579711914062
    },
    {
      "auxiliary_loss_clip": 0.01162337,
      "auxiliary_loss_mlp": 0.01032422,
      "balance_loss_clip": 1.05352497,
      "balance_loss_mlp": 1.02368999,
      "epoch": 0.3927132808272711,
      "flos": 16797001075200.0,
      "grad_norm": 2.429185049744173,
      "language_loss": 0.74500448,
      "learning_rate": 2.771659572541183e-06,
      "loss": 0.76695216,
      "num_input_tokens_seen": 70376915,
      "step": 3266,
      "time_per_iteration": 2.6885335445404053
    },
    {
      "auxiliary_loss_clip": 0.0116994,
      "auxiliary_loss_mlp": 0.01038111,
      "balance_loss_clip": 1.05812788,
      "balance_loss_mlp": 1.02968311,
      "epoch": 0.3928335237179102,
      "flos": 20267228908800.0,
      "grad_norm": 7.174204093567023,
      "language_loss": 0.87101924,
      "learning_rate": 2.7709408581688143e-06,
      "loss": 0.89309973,
      "num_input_tokens_seen": 70396900,
      "step": 3267,
      "time_per_iteration": 2.7126431465148926
    },
    {
      "auxiliary_loss_clip": 0.01188943,
      "auxiliary_loss_mlp": 0.01034818,
      "balance_loss_clip": 1.06034684,
      "balance_loss_mlp": 1.02576983,
      "epoch": 0.39295376660854925,
      "flos": 24973250209920.0,
      "grad_norm": 13.442255318220765,
      "language_loss": 0.87934864,
      "learning_rate": 2.7702220268453307e-06,
      "loss": 0.90158623,
      "num_input_tokens_seen": 70417260,
      "step": 3268,
      "time_per_iteration": 2.6742522716522217
    },
    {
      "auxiliary_loss_clip": 0.011695,
      "auxiliary_loss_mlp": 0.01034889,
      "balance_loss_clip": 1.0580461,
      "balance_loss_mlp": 1.02590656,
      "epoch": 0.39307400949918836,
      "flos": 18697788984960.0,
      "grad_norm": 2.246690400916314,
      "language_loss": 0.85069233,
      "learning_rate": 2.7695030786797785e-06,
      "loss": 0.87273622,
      "num_input_tokens_seen": 70433155,
      "step": 3269,
      "time_per_iteration": 2.725339412689209
    },
    {
      "auxiliary_loss_clip": 0.01199195,
      "auxiliary_loss_mlp": 0.01035371,
      "balance_loss_clip": 1.05989861,
      "balance_loss_mlp": 1.02650738,
      "epoch": 0.39319425238982747,
      "flos": 22415476590720.0,
      "grad_norm": 2.70736518691767,
      "language_loss": 0.74785388,
      "learning_rate": 2.7687840137812206e-06,
      "loss": 0.77019954,
      "num_input_tokens_seen": 70451240,
      "step": 3270,
      "time_per_iteration": 2.60722017288208
    },
    {
      "auxiliary_loss_clip": 0.01075103,
      "auxiliary_loss_mlp": 0.01000737,
      "balance_loss_clip": 1.01225591,
      "balance_loss_mlp": 0.99938959,
      "epoch": 0.3933144952804665,
      "flos": 66192954762240.0,
      "grad_norm": 0.7954827394208468,
      "language_loss": 0.61990631,
      "learning_rate": 2.7680648322587395e-06,
      "loss": 0.6406647,
      "num_input_tokens_seen": 70516115,
      "step": 3271,
      "time_per_iteration": 3.2633512020111084
    },
    {
      "auxiliary_loss_clip": 0.0119413,
      "auxiliary_loss_mlp": 0.01031217,
      "balance_loss_clip": 1.05614543,
      "balance_loss_mlp": 1.02249682,
      "epoch": 0.39343473817110564,
      "flos": 15487159720320.0,
      "grad_norm": 2.0437185099071318,
      "language_loss": 0.81032866,
      "learning_rate": 2.7673455342214334e-06,
      "loss": 0.83258212,
      "num_input_tokens_seen": 70533105,
      "step": 3272,
      "time_per_iteration": 2.538654327392578
    },
    {
      "auxiliary_loss_clip": 0.01171142,
      "auxiliary_loss_mlp": 0.01039123,
      "balance_loss_clip": 1.05438089,
      "balance_loss_mlp": 1.03023577,
      "epoch": 0.39355498106174475,
      "flos": 21324905809920.0,
      "grad_norm": 2.2531477174646977,
      "language_loss": 0.76340014,
      "learning_rate": 2.7666261197784198e-06,
      "loss": 0.78550279,
      "num_input_tokens_seen": 70551920,
      "step": 3273,
      "time_per_iteration": 2.726350784301758
    },
    {
      "auxiliary_loss_clip": 0.01185782,
      "auxiliary_loss_mlp": 0.01027626,
      "balance_loss_clip": 1.05720258,
      "balance_loss_mlp": 1.01935315,
      "epoch": 0.3936752239523838,
      "flos": 13296357400320.0,
      "grad_norm": 2.101425942721551,
      "language_loss": 0.76454508,
      "learning_rate": 2.7659065890388336e-06,
      "loss": 0.78667915,
      "num_input_tokens_seen": 70567920,
      "step": 3274,
      "time_per_iteration": 2.583340644836426
    },
    {
      "auxiliary_loss_clip": 0.01182388,
      "auxiliary_loss_mlp": 0.01035929,
      "balance_loss_clip": 1.05558276,
      "balance_loss_mlp": 1.02671432,
      "epoch": 0.3937954668430229,
      "flos": 16800161472000.0,
      "grad_norm": 2.07992402586797,
      "language_loss": 0.84994888,
      "learning_rate": 2.7651869421118266e-06,
      "loss": 0.87213212,
      "num_input_tokens_seen": 70584530,
      "step": 3275,
      "time_per_iteration": 2.7514469623565674
    },
    {
      "auxiliary_loss_clip": 0.01166743,
      "auxiliary_loss_mlp": 0.0103121,
      "balance_loss_clip": 1.05561399,
      "balance_loss_mlp": 1.0223887,
      "epoch": 0.393915709733662,
      "flos": 21064229832960.0,
      "grad_norm": 1.7269958702337336,
      "language_loss": 0.831967,
      "learning_rate": 2.76446717910657e-06,
      "loss": 0.85394651,
      "num_input_tokens_seen": 70605235,
      "step": 3276,
      "time_per_iteration": 2.7553460597991943
    },
    {
      "auxiliary_loss_clip": 0.01185792,
      "auxiliary_loss_mlp": 0.01034956,
      "balance_loss_clip": 1.05725312,
      "balance_loss_mlp": 1.02600956,
      "epoch": 0.3940359526243011,
      "flos": 17165265264000.0,
      "grad_norm": 2.871098190334614,
      "language_loss": 0.77119696,
      "learning_rate": 2.763747300132249e-06,
      "loss": 0.79340446,
      "num_input_tokens_seen": 70622675,
      "step": 3277,
      "time_per_iteration": 2.6093056201934814
    },
    {
      "auxiliary_loss_clip": 0.01188286,
      "auxiliary_loss_mlp": 0.0103389,
      "balance_loss_clip": 1.05878282,
      "balance_loss_mlp": 1.0255816,
      "epoch": 0.3941561955149402,
      "flos": 20995856294400.0,
      "grad_norm": 2.071089356132871,
      "language_loss": 0.86478841,
      "learning_rate": 2.7630273052980704e-06,
      "loss": 0.88701022,
      "num_input_tokens_seen": 70643265,
      "step": 3278,
      "time_per_iteration": 2.659806489944458
    },
    {
      "auxiliary_loss_clip": 0.01159029,
      "auxiliary_loss_mlp": 0.01031538,
      "balance_loss_clip": 1.05084062,
      "balance_loss_mlp": 1.02284169,
      "epoch": 0.39427643840557924,
      "flos": 18843406721280.0,
      "grad_norm": 13.898117010791813,
      "language_loss": 0.67622495,
      "learning_rate": 2.7623071947132554e-06,
      "loss": 0.69813067,
      "num_input_tokens_seen": 70660295,
      "step": 3279,
      "time_per_iteration": 2.6814725399017334
    },
    {
      "auxiliary_loss_clip": 0.01190739,
      "auxiliary_loss_mlp": 0.0103249,
      "balance_loss_clip": 1.05839086,
      "balance_loss_mlp": 1.02311397,
      "epoch": 0.39439668129621835,
      "flos": 23258659426560.0,
      "grad_norm": 1.8996631300701508,
      "language_loss": 0.78710407,
      "learning_rate": 2.7615869684870458e-06,
      "loss": 0.8093363,
      "num_input_tokens_seen": 70679605,
      "step": 3280,
      "time_per_iteration": 2.720355749130249
    },
    {
      "auxiliary_loss_clip": 0.01197755,
      "auxiliary_loss_mlp": 0.01033457,
      "balance_loss_clip": 1.05933082,
      "balance_loss_mlp": 1.02505827,
      "epoch": 0.39451692418685746,
      "flos": 26652289507200.0,
      "grad_norm": 1.6390343860785528,
      "language_loss": 0.84808779,
      "learning_rate": 2.7608666267286986e-06,
      "loss": 0.87039995,
      "num_input_tokens_seen": 70699835,
      "step": 3281,
      "time_per_iteration": 2.6525020599365234
    },
    {
      "auxiliary_loss_clip": 0.01187638,
      "auxiliary_loss_mlp": 0.01031544,
      "balance_loss_clip": 1.05697691,
      "balance_loss_mlp": 1.02211487,
      "epoch": 0.3946371670774965,
      "flos": 18258709132800.0,
      "grad_norm": 3.533048730127141,
      "language_loss": 0.87060124,
      "learning_rate": 2.760146169547489e-06,
      "loss": 0.89279306,
      "num_input_tokens_seen": 70716600,
      "step": 3282,
      "time_per_iteration": 2.6744515895843506
    },
    {
      "auxiliary_loss_clip": 0.01169235,
      "auxiliary_loss_mlp": 0.01041404,
      "balance_loss_clip": 1.05910265,
      "balance_loss_mlp": 1.03277636,
      "epoch": 0.39475740996813563,
      "flos": 24206126423040.0,
      "grad_norm": 1.712536860717565,
      "language_loss": 0.76339841,
      "learning_rate": 2.75942559705271e-06,
      "loss": 0.7855047,
      "num_input_tokens_seen": 70736335,
      "step": 3283,
      "time_per_iteration": 3.6000876426696777
    },
    {
      "auxiliary_loss_clip": 0.01166729,
      "auxiliary_loss_mlp": 0.0103376,
      "balance_loss_clip": 1.05739808,
      "balance_loss_mlp": 1.02455664,
      "epoch": 0.39487765285877474,
      "flos": 19317858491520.0,
      "grad_norm": 1.8653397673919332,
      "language_loss": 0.89164412,
      "learning_rate": 2.7587049093536713e-06,
      "loss": 0.91364908,
      "num_input_tokens_seen": 70752665,
      "step": 3284,
      "time_per_iteration": 3.617626428604126
    },
    {
      "auxiliary_loss_clip": 0.01171206,
      "auxiliary_loss_mlp": 0.01030794,
      "balance_loss_clip": 1.05611348,
      "balance_loss_mlp": 1.02240205,
      "epoch": 0.3949978957494138,
      "flos": 17311744926720.0,
      "grad_norm": 1.9477847246372164,
      "language_loss": 0.8090055,
      "learning_rate": 2.757984106559701e-06,
      "loss": 0.83102554,
      "num_input_tokens_seen": 70771650,
      "step": 3285,
      "time_per_iteration": 2.725210666656494
    },
    {
      "auxiliary_loss_clip": 0.01183288,
      "auxiliary_loss_mlp": 0.01033237,
      "balance_loss_clip": 1.05814815,
      "balance_loss_mlp": 1.02492249,
      "epoch": 0.3951181386400529,
      "flos": 36317861280000.0,
      "grad_norm": 2.7049591427149564,
      "language_loss": 0.71739125,
      "learning_rate": 2.7572631887801446e-06,
      "loss": 0.73955655,
      "num_input_tokens_seen": 70793275,
      "step": 3286,
      "time_per_iteration": 3.806269407272339
    },
    {
      "auxiliary_loss_clip": 0.01187534,
      "auxiliary_loss_mlp": 0.01030954,
      "balance_loss_clip": 1.05878401,
      "balance_loss_mlp": 1.02199554,
      "epoch": 0.395238381530692,
      "flos": 23110348170240.0,
      "grad_norm": 1.8277961880973528,
      "language_loss": 0.76759923,
      "learning_rate": 2.7565421561243654e-06,
      "loss": 0.78978413,
      "num_input_tokens_seen": 70811440,
      "step": 3287,
      "time_per_iteration": 2.680216073989868
    },
    {
      "auxiliary_loss_clip": 0.01154379,
      "auxiliary_loss_mlp": 0.01030865,
      "balance_loss_clip": 1.0527972,
      "balance_loss_mlp": 1.02265787,
      "epoch": 0.3953586244213311,
      "flos": 24347614095360.0,
      "grad_norm": 2.4262818654489053,
      "language_loss": 0.82593209,
      "learning_rate": 2.7558210087017413e-06,
      "loss": 0.84778452,
      "num_input_tokens_seen": 70831375,
      "step": 3288,
      "time_per_iteration": 2.836850166320801
    },
    {
      "auxiliary_loss_clip": 0.01199569,
      "auxiliary_loss_mlp": 0.01035042,
      "balance_loss_clip": 1.06008875,
      "balance_loss_mlp": 1.0259881,
      "epoch": 0.3954788673119702,
      "flos": 23440080044160.0,
      "grad_norm": 3.0516020598065343,
      "language_loss": 0.73216832,
      "learning_rate": 2.7550997466216724e-06,
      "loss": 0.75451446,
      "num_input_tokens_seen": 70849170,
      "step": 3289,
      "time_per_iteration": 2.6296722888946533
    },
    {
      "auxiliary_loss_clip": 0.0119851,
      "auxiliary_loss_mlp": 0.01038513,
      "balance_loss_clip": 1.06099248,
      "balance_loss_mlp": 1.02954841,
      "epoch": 0.3955991102026093,
      "flos": 17494063384320.0,
      "grad_norm": 1.9446688942267867,
      "language_loss": 0.81110013,
      "learning_rate": 2.7543783699935714e-06,
      "loss": 0.83347034,
      "num_input_tokens_seen": 70867200,
      "step": 3290,
      "time_per_iteration": 3.5575568675994873
    },
    {
      "auxiliary_loss_clip": 0.01173454,
      "auxiliary_loss_mlp": 0.01035335,
      "balance_loss_clip": 1.05613434,
      "balance_loss_mlp": 1.02688336,
      "epoch": 0.39571935309324835,
      "flos": 18221326053120.0,
      "grad_norm": 3.008042187978964,
      "language_loss": 0.85447586,
      "learning_rate": 2.753656878926872e-06,
      "loss": 0.87656367,
      "num_input_tokens_seen": 70883080,
      "step": 3291,
      "time_per_iteration": 2.5980968475341797
    },
    {
      "auxiliary_loss_clip": 0.01188426,
      "auxiliary_loss_mlp": 0.0103552,
      "balance_loss_clip": 1.05748653,
      "balance_loss_mlp": 1.02676761,
      "epoch": 0.39583959598388746,
      "flos": 17748813617280.0,
      "grad_norm": 1.7802833495665242,
      "language_loss": 0.74968743,
      "learning_rate": 2.752935273531023e-06,
      "loss": 0.77192688,
      "num_input_tokens_seen": 70901230,
      "step": 3292,
      "time_per_iteration": 2.680912971496582
    },
    {
      "auxiliary_loss_clip": 0.0115307,
      "auxiliary_loss_mlp": 0.01038071,
      "balance_loss_clip": 1.05395317,
      "balance_loss_mlp": 1.02825999,
      "epoch": 0.39595983887452657,
      "flos": 19352368483200.0,
      "grad_norm": 2.30552932585984,
      "language_loss": 0.78928143,
      "learning_rate": 2.752213553915492e-06,
      "loss": 0.81119275,
      "num_input_tokens_seen": 70919585,
      "step": 3293,
      "time_per_iteration": 2.7269551753997803
    },
    {
      "auxiliary_loss_clip": 0.01073722,
      "auxiliary_loss_mlp": 0.01004927,
      "balance_loss_clip": 1.01368594,
      "balance_loss_mlp": 1.00363374,
      "epoch": 0.3960800817651656,
      "flos": 60682282940160.0,
      "grad_norm": 0.8270048709488252,
      "language_loss": 0.66046458,
      "learning_rate": 2.751491720189762e-06,
      "loss": 0.68125111,
      "num_input_tokens_seen": 70977695,
      "step": 3294,
      "time_per_iteration": 3.153404712677002
    },
    {
      "auxiliary_loss_clip": 0.01177948,
      "auxiliary_loss_mlp": 0.01031577,
      "balance_loss_clip": 1.05422664,
      "balance_loss_mlp": 1.02300584,
      "epoch": 0.39620032465580474,
      "flos": 16836718538880.0,
      "grad_norm": 2.2171239357025403,
      "language_loss": 0.91949272,
      "learning_rate": 2.7507697724633364e-06,
      "loss": 0.94158792,
      "num_input_tokens_seen": 70994455,
      "step": 3295,
      "time_per_iteration": 2.6448161602020264
    },
    {
      "auxiliary_loss_clip": 0.01050873,
      "auxiliary_loss_mlp": 0.01003937,
      "balance_loss_clip": 1.01276648,
      "balance_loss_mlp": 1.00267935,
      "epoch": 0.3963205675464438,
      "flos": 69071445941760.0,
      "grad_norm": 0.7767116001275123,
      "language_loss": 0.54624534,
      "learning_rate": 2.7500477108457327e-06,
      "loss": 0.56679344,
      "num_input_tokens_seen": 71046465,
      "step": 3296,
      "time_per_iteration": 3.1142220497131348
    },
    {
      "auxiliary_loss_clip": 0.01197259,
      "auxiliary_loss_mlp": 0.01030365,
      "balance_loss_clip": 1.05831134,
      "balance_loss_mlp": 1.02180576,
      "epoch": 0.3964408104370829,
      "flos": 25667439431040.0,
      "grad_norm": 3.3301327095587974,
      "language_loss": 0.80924189,
      "learning_rate": 2.7493255354464877e-06,
      "loss": 0.83151811,
      "num_input_tokens_seen": 71064275,
      "step": 3297,
      "time_per_iteration": 2.715538263320923
    },
    {
      "auxiliary_loss_clip": 0.01194833,
      "auxiliary_loss_mlp": 0.01037206,
      "balance_loss_clip": 1.0586735,
      "balance_loss_mlp": 1.02787757,
      "epoch": 0.396561053327722,
      "flos": 24277480790400.0,
      "grad_norm": 1.878923005805417,
      "language_loss": 0.7625016,
      "learning_rate": 2.748603246375156e-06,
      "loss": 0.78482199,
      "num_input_tokens_seen": 71082290,
      "step": 3298,
      "time_per_iteration": 2.6890506744384766
    },
    {
      "auxiliary_loss_clip": 0.01197992,
      "auxiliary_loss_mlp": 0.0103239,
      "balance_loss_clip": 1.0602901,
      "balance_loss_mlp": 1.02341402,
      "epoch": 0.39668129621836107,
      "flos": 20522302364160.0,
      "grad_norm": 2.5882381917567727,
      "language_loss": 0.70049679,
      "learning_rate": 2.7478808437413055e-06,
      "loss": 0.72280061,
      "num_input_tokens_seen": 71101700,
      "step": 3299,
      "time_per_iteration": 2.6607115268707275
    },
    {
      "auxiliary_loss_clip": 0.01201733,
      "auxiliary_loss_mlp": 0.01037276,
      "balance_loss_clip": 1.06151462,
      "balance_loss_mlp": 1.02866912,
      "epoch": 0.3968015391090002,
      "flos": 27052585649280.0,
      "grad_norm": 2.2270739620634705,
      "language_loss": 0.663715,
      "learning_rate": 2.7471583276545263e-06,
      "loss": 0.68610501,
      "num_input_tokens_seen": 71122360,
      "step": 3300,
      "time_per_iteration": 2.619554042816162
    },
    {
      "auxiliary_loss_clip": 0.01178621,
      "auxiliary_loss_mlp": 0.01038762,
      "balance_loss_clip": 1.05507314,
      "balance_loss_mlp": 1.03022087,
      "epoch": 0.3969217819996393,
      "flos": 12531819392640.0,
      "grad_norm": 3.9247658839813644,
      "language_loss": 0.71151036,
      "learning_rate": 2.7464356982244224e-06,
      "loss": 0.73368418,
      "num_input_tokens_seen": 71140360,
      "step": 3301,
      "time_per_iteration": 2.6610684394836426
    },
    {
      "auxiliary_loss_clip": 0.01082369,
      "auxiliary_loss_mlp": 0.01001378,
      "balance_loss_clip": 1.01112127,
      "balance_loss_mlp": 1.00012648,
      "epoch": 0.39704202489027834,
      "flos": 66241399230720.0,
      "grad_norm": 0.7774546578961268,
      "language_loss": 0.61710387,
      "learning_rate": 2.745712955560617e-06,
      "loss": 0.6379413,
      "num_input_tokens_seen": 71196565,
      "step": 3302,
      "time_per_iteration": 3.1411290168762207
    },
    {
      "auxiliary_loss_clip": 0.01175092,
      "auxiliary_loss_mlp": 0.01030879,
      "balance_loss_clip": 1.05679572,
      "balance_loss_mlp": 1.02155685,
      "epoch": 0.39716226778091746,
      "flos": 16982982720000.0,
      "grad_norm": 2.27776991477389,
      "language_loss": 0.7775265,
      "learning_rate": 2.7449900997727496e-06,
      "loss": 0.79958618,
      "num_input_tokens_seen": 71214675,
      "step": 3303,
      "time_per_iteration": 2.6806998252868652
    },
    {
      "auxiliary_loss_clip": 0.01157686,
      "auxiliary_loss_mlp": 0.01035115,
      "balance_loss_clip": 1.05446649,
      "balance_loss_mlp": 1.02675271,
      "epoch": 0.39728251067155657,
      "flos": 23477139901440.0,
      "grad_norm": 2.5118269184503728,
      "language_loss": 0.84095067,
      "learning_rate": 2.744267130970476e-06,
      "loss": 0.86287874,
      "num_input_tokens_seen": 71234400,
      "step": 3304,
      "time_per_iteration": 2.784898042678833
    },
    {
      "auxiliary_loss_clip": 0.0118507,
      "auxiliary_loss_mlp": 0.01036143,
      "balance_loss_clip": 1.0601511,
      "balance_loss_mlp": 1.02713084,
      "epoch": 0.3974027535621956,
      "flos": 20704441253760.0,
      "grad_norm": 1.9240888236165705,
      "language_loss": 0.77002239,
      "learning_rate": 2.7435440492634697e-06,
      "loss": 0.79223454,
      "num_input_tokens_seen": 71253725,
      "step": 3305,
      "time_per_iteration": 2.7262470722198486
    },
    {
      "auxiliary_loss_clip": 0.01202701,
      "auxiliary_loss_mlp": 0.01033421,
      "balance_loss_clip": 1.05861139,
      "balance_loss_mlp": 1.0234971,
      "epoch": 0.39752299645283473,
      "flos": 21543278544000.0,
      "grad_norm": 2.2223627336657543,
      "language_loss": 0.67164361,
      "learning_rate": 2.7428208547614228e-06,
      "loss": 0.69400477,
      "num_input_tokens_seen": 71273220,
      "step": 3306,
      "time_per_iteration": 2.6781914234161377
    },
    {
      "auxiliary_loss_clip": 0.01118986,
      "auxiliary_loss_mlp": 0.0103415,
      "balance_loss_clip": 1.04630804,
      "balance_loss_mlp": 1.02483368,
      "epoch": 0.39764323934347384,
      "flos": 19208295031680.0,
      "grad_norm": 2.2742335335712167,
      "language_loss": 0.77478862,
      "learning_rate": 2.742097547574043e-06,
      "loss": 0.79631996,
      "num_input_tokens_seen": 71291445,
      "step": 3307,
      "time_per_iteration": 2.7294411659240723
    },
    {
      "auxiliary_loss_clip": 0.01199455,
      "auxiliary_loss_mlp": 0.01031721,
      "balance_loss_clip": 1.05902863,
      "balance_loss_mlp": 1.02265537,
      "epoch": 0.3977634822341129,
      "flos": 20850202644480.0,
      "grad_norm": 2.700825605084323,
      "language_loss": 0.77962935,
      "learning_rate": 2.7413741278110544e-06,
      "loss": 0.80194104,
      "num_input_tokens_seen": 71310135,
      "step": 3308,
      "time_per_iteration": 3.4027352333068848
    },
    {
      "auxiliary_loss_clip": 0.01184965,
      "auxiliary_loss_mlp": 0.01039255,
      "balance_loss_clip": 1.05963469,
      "balance_loss_mlp": 1.0305829,
      "epoch": 0.397883725124752,
      "flos": 39786042038400.0,
      "grad_norm": 2.971957699986104,
      "language_loss": 0.69475973,
      "learning_rate": 2.7406505955822016e-06,
      "loss": 0.71700191,
      "num_input_tokens_seen": 71331160,
      "step": 3309,
      "time_per_iteration": 2.6978938579559326
    },
    {
      "auxiliary_loss_clip": 0.01182211,
      "auxiliary_loss_mlp": 0.01029835,
      "balance_loss_clip": 1.0575676,
      "balance_loss_mlp": 1.02094793,
      "epoch": 0.39800396801539106,
      "flos": 17379507934080.0,
      "grad_norm": 2.9474000330591505,
      "language_loss": 0.66479981,
      "learning_rate": 2.7399269509972415e-06,
      "loss": 0.68692029,
      "num_input_tokens_seen": 71345315,
      "step": 3310,
      "time_per_iteration": 3.246547222137451
    },
    {
      "auxiliary_loss_clip": 0.01201634,
      "auxiliary_loss_mlp": 0.01031873,
      "balance_loss_clip": 1.05919373,
      "balance_loss_mlp": 1.02252698,
      "epoch": 0.3981242109060302,
      "flos": 19202764337280.0,
      "grad_norm": 2.5924481472523735,
      "language_loss": 0.84915215,
      "learning_rate": 2.7392031941659514e-06,
      "loss": 0.87148726,
      "num_input_tokens_seen": 71363160,
      "step": 3311,
      "time_per_iteration": 2.562527656555176
    },
    {
      "auxiliary_loss_clip": 0.01184804,
      "auxiliary_loss_mlp": 0.01032238,
      "balance_loss_clip": 1.06043005,
      "balance_loss_mlp": 1.02328539,
      "epoch": 0.3982444537966693,
      "flos": 24565124903040.0,
      "grad_norm": 2.5994683779697945,
      "language_loss": 0.86354876,
      "learning_rate": 2.7384793251981244e-06,
      "loss": 0.88571918,
      "num_input_tokens_seen": 71382145,
      "step": 3312,
      "time_per_iteration": 3.5531699657440186
    },
    {
      "auxiliary_loss_clip": 0.01199415,
      "auxiliary_loss_mlp": 0.01027565,
      "balance_loss_clip": 1.05901623,
      "balance_loss_mlp": 1.01827884,
      "epoch": 0.39836469668730834,
      "flos": 26213856099840.0,
      "grad_norm": 2.384973639723125,
      "language_loss": 0.81019437,
      "learning_rate": 2.737755344203571e-06,
      "loss": 0.8324641,
      "num_input_tokens_seen": 71402095,
      "step": 3313,
      "time_per_iteration": 2.6867072582244873
    },
    {
      "auxiliary_loss_clip": 0.0116741,
      "auxiliary_loss_mlp": 0.01030889,
      "balance_loss_clip": 1.05900407,
      "balance_loss_mlp": 1.02220488,
      "epoch": 0.39848493957794745,
      "flos": 27636134002560.0,
      "grad_norm": 1.70460054954484,
      "language_loss": 0.8014462,
      "learning_rate": 2.7370312512921186e-06,
      "loss": 0.82342911,
      "num_input_tokens_seen": 71423875,
      "step": 3314,
      "time_per_iteration": 2.744166374206543
    },
    {
      "auxiliary_loss_clip": 0.0117125,
      "auxiliary_loss_mlp": 0.0076303,
      "balance_loss_clip": 1.05226374,
      "balance_loss_mlp": 1.00107503,
      "epoch": 0.39860518246858656,
      "flos": 12239326944000.0,
      "grad_norm": 2.6511622414850446,
      "language_loss": 0.77288651,
      "learning_rate": 2.736307046573611e-06,
      "loss": 0.79222935,
      "num_input_tokens_seen": 71439745,
      "step": 3315,
      "time_per_iteration": 2.6518359184265137
    },
    {
      "auxiliary_loss_clip": 0.01196711,
      "auxiliary_loss_mlp": 0.01026915,
      "balance_loss_clip": 1.05936134,
      "balance_loss_mlp": 1.01839161,
      "epoch": 0.3987254253592256,
      "flos": 22379135005440.0,
      "grad_norm": 1.721712692745014,
      "language_loss": 0.81630099,
      "learning_rate": 2.73558273015791e-06,
      "loss": 0.83853722,
      "num_input_tokens_seen": 71459575,
      "step": 3316,
      "time_per_iteration": 3.5329220294952393
    },
    {
      "auxiliary_loss_clip": 0.01183853,
      "auxiliary_loss_mlp": 0.0104455,
      "balance_loss_clip": 1.05978107,
      "balance_loss_mlp": 1.0357821,
      "epoch": 0.3988456682498647,
      "flos": 23514020190720.0,
      "grad_norm": 4.344968290309083,
      "language_loss": 0.70985085,
      "learning_rate": 2.734858302154894e-06,
      "loss": 0.73213494,
      "num_input_tokens_seen": 71481075,
      "step": 3317,
      "time_per_iteration": 2.849367141723633
    },
    {
      "auxiliary_loss_clip": 0.01172978,
      "auxiliary_loss_mlp": 0.01029781,
      "balance_loss_clip": 1.0566349,
      "balance_loss_mlp": 1.02095985,
      "epoch": 0.39896591114050384,
      "flos": 19208761908480.0,
      "grad_norm": 2.2256844699253238,
      "language_loss": 0.76793176,
      "learning_rate": 2.734133762674457e-06,
      "loss": 0.78995931,
      "num_input_tokens_seen": 71500665,
      "step": 3318,
      "time_per_iteration": 2.669743061065674
    },
    {
      "auxiliary_loss_clip": 0.01186771,
      "auxiliary_loss_mlp": 0.01031515,
      "balance_loss_clip": 1.05755472,
      "balance_loss_mlp": 1.02238965,
      "epoch": 0.3990861540311429,
      "flos": 28401031146240.0,
      "grad_norm": 2.6708083893878993,
      "language_loss": 0.71136856,
      "learning_rate": 2.7334091118265124e-06,
      "loss": 0.7335515,
      "num_input_tokens_seen": 71522560,
      "step": 3319,
      "time_per_iteration": 2.6769638061523438
    },
    {
      "auxiliary_loss_clip": 0.01052131,
      "auxiliary_loss_mlp": 0.01002388,
      "balance_loss_clip": 1.01208425,
      "balance_loss_mlp": 1.00113034,
      "epoch": 0.399206396921782,
      "flos": 61758563086080.0,
      "grad_norm": 0.6844115121601526,
      "language_loss": 0.57805771,
      "learning_rate": 2.732684349720989e-06,
      "loss": 0.59860283,
      "num_input_tokens_seen": 71590520,
      "step": 3320,
      "time_per_iteration": 3.3306350708007812
    },
    {
      "auxiliary_loss_clip": 0.01185332,
      "auxiliary_loss_mlp": 0.01032798,
      "balance_loss_clip": 1.05921209,
      "balance_loss_mlp": 1.02349377,
      "epoch": 0.3993266398124211,
      "flos": 28074567409920.0,
      "grad_norm": 1.6276951139004774,
      "language_loss": 0.75585097,
      "learning_rate": 2.7319594764678318e-06,
      "loss": 0.77803224,
      "num_input_tokens_seen": 71612620,
      "step": 3321,
      "time_per_iteration": 2.717238664627075
    },
    {
      "auxiliary_loss_clip": 0.01183985,
      "auxiliary_loss_mlp": 0.01036673,
      "balance_loss_clip": 1.05965614,
      "balance_loss_mlp": 1.0271008,
      "epoch": 0.39944688270306017,
      "flos": 23225083188480.0,
      "grad_norm": 1.9856751780128379,
      "language_loss": 0.83788466,
      "learning_rate": 2.7312344921770044e-06,
      "loss": 0.86009121,
      "num_input_tokens_seen": 71634320,
      "step": 3322,
      "time_per_iteration": 2.6872332096099854
    },
    {
      "auxiliary_loss_clip": 0.01197403,
      "auxiliary_loss_mlp": 0.0103093,
      "balance_loss_clip": 1.05914283,
      "balance_loss_mlp": 1.02255607,
      "epoch": 0.3995671255936993,
      "flos": 19390433921280.0,
      "grad_norm": 1.977391699397491,
      "language_loss": 0.78383207,
      "learning_rate": 2.7305093969584857e-06,
      "loss": 0.80611539,
      "num_input_tokens_seen": 71653145,
      "step": 3323,
      "time_per_iteration": 2.654217004776001
    },
    {
      "auxiliary_loss_clip": 0.01171606,
      "auxiliary_loss_mlp": 0.0103421,
      "balance_loss_clip": 1.05672479,
      "balance_loss_mlp": 1.02569199,
      "epoch": 0.3996873684843384,
      "flos": 23842638743040.0,
      "grad_norm": 2.186531389942788,
      "language_loss": 0.80217773,
      "learning_rate": 2.729784190922272e-06,
      "loss": 0.82423586,
      "num_input_tokens_seen": 71674580,
      "step": 3324,
      "time_per_iteration": 2.7782280445098877
    },
    {
      "auxiliary_loss_clip": 0.01065981,
      "auxiliary_loss_mlp": 0.01001547,
      "balance_loss_clip": 1.01539588,
      "balance_loss_mlp": 1.00031948,
      "epoch": 0.39980761137497745,
      "flos": 66576877280640.0,
      "grad_norm": 0.9319800743067211,
      "language_loss": 0.57078028,
      "learning_rate": 2.729058874178378e-06,
      "loss": 0.59145558,
      "num_input_tokens_seen": 71745260,
      "step": 3325,
      "time_per_iteration": 3.389360189437866
    },
    {
      "auxiliary_loss_clip": 0.01198917,
      "auxiliary_loss_mlp": 0.01031484,
      "balance_loss_clip": 1.06015408,
      "balance_loss_mlp": 1.02276397,
      "epoch": 0.39992785426561656,
      "flos": 28549162834560.0,
      "grad_norm": 1.9789388873054203,
      "language_loss": 0.68863767,
      "learning_rate": 2.7283334468368315e-06,
      "loss": 0.71094167,
      "num_input_tokens_seen": 71766540,
      "step": 3326,
      "time_per_iteration": 2.6707353591918945
    },
    {
      "auxiliary_loss_clip": 0.01198915,
      "auxiliary_loss_mlp": 0.01029696,
      "balance_loss_clip": 1.05885053,
      "balance_loss_mlp": 1.02021337,
      "epoch": 0.4000480971562556,
      "flos": 15049408671360.0,
      "grad_norm": 3.6688821893378027,
      "language_loss": 0.73503435,
      "learning_rate": 2.72760790900768e-06,
      "loss": 0.7573204,
      "num_input_tokens_seen": 71783125,
      "step": 3327,
      "time_per_iteration": 2.6164281368255615
    },
    {
      "auxiliary_loss_clip": 0.01176202,
      "auxiliary_loss_mlp": 0.01038225,
      "balance_loss_clip": 1.06031466,
      "balance_loss_mlp": 1.02971339,
      "epoch": 0.4001683400468947,
      "flos": 23915609222400.0,
      "grad_norm": 2.181826268510732,
      "language_loss": 0.79218042,
      "learning_rate": 2.7268822608009875e-06,
      "loss": 0.81432468,
      "num_input_tokens_seen": 71802500,
      "step": 3328,
      "time_per_iteration": 2.691131353378296
    },
    {
      "auxiliary_loss_clip": 0.0120088,
      "auxiliary_loss_mlp": 0.01038504,
      "balance_loss_clip": 1.05998313,
      "balance_loss_mlp": 1.02890205,
      "epoch": 0.40028858293753383,
      "flos": 24352677912960.0,
      "grad_norm": 3.8754468156132704,
      "language_loss": 0.78601432,
      "learning_rate": 2.726156502326834e-06,
      "loss": 0.80840814,
      "num_input_tokens_seen": 71823800,
      "step": 3329,
      "time_per_iteration": 2.762647867202759
    },
    {
      "auxiliary_loss_clip": 0.01085649,
      "auxiliary_loss_mlp": 0.01002736,
      "balance_loss_clip": 1.01390958,
      "balance_loss_mlp": 1.0015496,
      "epoch": 0.4004088258281729,
      "flos": 66787025800320.0,
      "grad_norm": 0.6964195826772585,
      "language_loss": 0.6023016,
      "learning_rate": 2.725430633695316e-06,
      "loss": 0.62318546,
      "num_input_tokens_seen": 71886880,
      "step": 3330,
      "time_per_iteration": 3.2484676837921143
    },
    {
      "auxiliary_loss_clip": 0.01070713,
      "auxiliary_loss_mlp": 0.01001512,
      "balance_loss_clip": 1.01305199,
      "balance_loss_mlp": 1.00038528,
      "epoch": 0.400529068718812,
      "flos": 58598386473600.0,
      "grad_norm": 0.8830661736753832,
      "language_loss": 0.57874918,
      "learning_rate": 2.7247046550165485e-06,
      "loss": 0.59947145,
      "num_input_tokens_seen": 71939005,
      "step": 3331,
      "time_per_iteration": 3.182528495788574
    },
    {
      "auxiliary_loss_clip": 0.01201277,
      "auxiliary_loss_mlp": 0.01029544,
      "balance_loss_clip": 1.06259036,
      "balance_loss_mlp": 1.02004337,
      "epoch": 0.4006493116094511,
      "flos": 25377460934400.0,
      "grad_norm": 1.918621537395705,
      "language_loss": 0.76117897,
      "learning_rate": 2.7239785664006606e-06,
      "loss": 0.7834872,
      "num_input_tokens_seen": 71962545,
      "step": 3332,
      "time_per_iteration": 2.7229936122894287
    },
    {
      "auxiliary_loss_clip": 0.01060788,
      "auxiliary_loss_mlp": 0.00753256,
      "balance_loss_clip": 1.01175499,
      "balance_loss_mlp": 1.00023639,
      "epoch": 0.40076955450009016,
      "flos": 60280729822080.0,
      "grad_norm": 0.770073295455211,
      "language_loss": 0.6178652,
      "learning_rate": 2.7232523679578002e-06,
      "loss": 0.63600564,
      "num_input_tokens_seen": 72025625,
      "step": 3333,
      "time_per_iteration": 3.3212389945983887
    },
    {
      "auxiliary_loss_clip": 0.01200218,
      "auxiliary_loss_mlp": 0.01029126,
      "balance_loss_clip": 1.06122756,
      "balance_loss_mlp": 1.02076983,
      "epoch": 0.4008897973907293,
      "flos": 16617268396800.0,
      "grad_norm": 2.1175284628690774,
      "language_loss": 0.7947399,
      "learning_rate": 2.7225260597981295e-06,
      "loss": 0.81703335,
      "num_input_tokens_seen": 72043330,
      "step": 3334,
      "time_per_iteration": 3.60719895362854
    },
    {
      "auxiliary_loss_clip": 0.01199056,
      "auxiliary_loss_mlp": 0.00763381,
      "balance_loss_clip": 1.059044,
      "balance_loss_mlp": 1.00108981,
      "epoch": 0.4010100402813684,
      "flos": 15377344865280.0,
      "grad_norm": 3.1361976182265217,
      "language_loss": 0.79194552,
      "learning_rate": 2.721799642031831e-06,
      "loss": 0.81156993,
      "num_input_tokens_seen": 72059500,
      "step": 3335,
      "time_per_iteration": 2.6703872680664062
    },
    {
      "auxiliary_loss_clip": 0.01201051,
      "auxiliary_loss_mlp": 0.01029561,
      "balance_loss_clip": 1.06150305,
      "balance_loss_mlp": 1.02031088,
      "epoch": 0.40113028317200744,
      "flos": 13298835438720.0,
      "grad_norm": 2.2387167442086997,
      "language_loss": 0.78147554,
      "learning_rate": 2.721073114769101e-06,
      "loss": 0.80378163,
      "num_input_tokens_seen": 72077175,
      "step": 3336,
      "time_per_iteration": 3.5511345863342285
    },
    {
      "auxiliary_loss_clip": 0.01180177,
      "auxiliary_loss_mlp": 0.01028459,
      "balance_loss_clip": 1.05619776,
      "balance_loss_mlp": 1.01942253,
      "epoch": 0.40125052606264655,
      "flos": 20668027841280.0,
      "grad_norm": 2.1804699935141634,
      "language_loss": 0.75543541,
      "learning_rate": 2.7203464781201523e-06,
      "loss": 0.77752173,
      "num_input_tokens_seen": 72096490,
      "step": 3337,
      "time_per_iteration": 2.6230545043945312
    },
    {
      "auxiliary_loss_clip": 0.01204037,
      "auxiliary_loss_mlp": 0.00762757,
      "balance_loss_clip": 1.06404328,
      "balance_loss_mlp": 1.00113475,
      "epoch": 0.40137076895328566,
      "flos": 24607679541120.0,
      "grad_norm": 2.501603977021052,
      "language_loss": 0.7841984,
      "learning_rate": 2.719619732195215e-06,
      "loss": 0.80386633,
      "num_input_tokens_seen": 72118130,
      "step": 3338,
      "time_per_iteration": 3.660447120666504
    },
    {
      "auxiliary_loss_clip": 0.01185583,
      "auxiliary_loss_mlp": 0.01033192,
      "balance_loss_clip": 1.05573726,
      "balance_loss_mlp": 1.0242033,
      "epoch": 0.4014910118439247,
      "flos": 24206593299840.0,
      "grad_norm": 3.8884463767145148,
      "language_loss": 0.72898626,
      "learning_rate": 2.7188928771045377e-06,
      "loss": 0.75117397,
      "num_input_tokens_seen": 72139450,
      "step": 3339,
      "time_per_iteration": 2.7120673656463623
    },
    {
      "auxiliary_loss_clip": 0.01164378,
      "auxiliary_loss_mlp": 0.01035092,
      "balance_loss_clip": 1.05181742,
      "balance_loss_mlp": 1.02619267,
      "epoch": 0.4016112547345638,
      "flos": 26725080418560.0,
      "grad_norm": 2.103909563517906,
      "language_loss": 0.80291098,
      "learning_rate": 2.7181659129583815e-06,
      "loss": 0.82490569,
      "num_input_tokens_seen": 72159040,
      "step": 3340,
      "time_per_iteration": 2.708678960800171
    },
    {
      "auxiliary_loss_clip": 0.011674,
      "auxiliary_loss_mlp": 0.0102653,
      "balance_loss_clip": 1.0559622,
      "balance_loss_mlp": 1.01753008,
      "epoch": 0.4017314976252029,
      "flos": 21288025520640.0,
      "grad_norm": 1.8586730770547213,
      "language_loss": 0.76064193,
      "learning_rate": 2.7174388398670276e-06,
      "loss": 0.78258121,
      "num_input_tokens_seen": 72178220,
      "step": 3341,
      "time_per_iteration": 3.6766116619110107
    },
    {
      "auxiliary_loss_clip": 0.01160597,
      "auxiliary_loss_mlp": 0.01035822,
      "balance_loss_clip": 1.05274498,
      "balance_loss_mlp": 1.02618361,
      "epoch": 0.401851740515842,
      "flos": 25484690010240.0,
      "grad_norm": 2.48270828614843,
      "language_loss": 0.92182267,
      "learning_rate": 2.716711657940773e-06,
      "loss": 0.94378686,
      "num_input_tokens_seen": 72199230,
      "step": 3342,
      "time_per_iteration": 2.683657646179199
    },
    {
      "auxiliary_loss_clip": 0.01062212,
      "auxiliary_loss_mlp": 0.01001404,
      "balance_loss_clip": 1.01147509,
      "balance_loss_mlp": 1.0000807,
      "epoch": 0.4019719834064811,
      "flos": 55395334978560.0,
      "grad_norm": 0.8110043056279437,
      "language_loss": 0.56519735,
      "learning_rate": 2.7159843672899284e-06,
      "loss": 0.58583355,
      "num_input_tokens_seen": 72263430,
      "step": 3343,
      "time_per_iteration": 3.380948066711426
    },
    {
      "auxiliary_loss_clip": 0.01146985,
      "auxiliary_loss_mlp": 0.01036809,
      "balance_loss_clip": 1.05509448,
      "balance_loss_mlp": 1.02751029,
      "epoch": 0.40209222629712016,
      "flos": 18180100218240.0,
      "grad_norm": 2.219506311300114,
      "language_loss": 0.81699455,
      "learning_rate": 2.715256968024825e-06,
      "loss": 0.8388325,
      "num_input_tokens_seen": 72280505,
      "step": 3344,
      "time_per_iteration": 2.7411420345306396
    },
    {
      "auxiliary_loss_clip": 0.01180039,
      "auxiliary_loss_mlp": 0.01034265,
      "balance_loss_clip": 1.05802441,
      "balance_loss_mlp": 1.02525902,
      "epoch": 0.40221246918775927,
      "flos": 25961009287680.0,
      "grad_norm": 3.3937472237601227,
      "language_loss": 0.82616436,
      "learning_rate": 2.7145294602558083e-06,
      "loss": 0.84830737,
      "num_input_tokens_seen": 72301215,
      "step": 3345,
      "time_per_iteration": 2.7389659881591797
    },
    {
      "auxiliary_loss_clip": 0.01182724,
      "auxiliary_loss_mlp": 0.01033697,
      "balance_loss_clip": 1.05727434,
      "balance_loss_mlp": 1.02456009,
      "epoch": 0.4023327120783984,
      "flos": 33838912056960.0,
      "grad_norm": 1.8416589539783192,
      "language_loss": 0.71089029,
      "learning_rate": 2.713801844093241e-06,
      "loss": 0.73305446,
      "num_input_tokens_seen": 72322365,
      "step": 3346,
      "time_per_iteration": 2.806234836578369
    },
    {
      "auxiliary_loss_clip": 0.01198624,
      "auxiliary_loss_mlp": 0.01038337,
      "balance_loss_clip": 1.05996156,
      "balance_loss_mlp": 1.02957559,
      "epoch": 0.40245295496903744,
      "flos": 26900252069760.0,
      "grad_norm": 13.75410132609205,
      "language_loss": 0.88563508,
      "learning_rate": 2.7130741196475014e-06,
      "loss": 0.9080047,
      "num_input_tokens_seen": 72340495,
      "step": 3347,
      "time_per_iteration": 2.6459476947784424
    },
    {
      "auxiliary_loss_clip": 0.01183928,
      "auxiliary_loss_mlp": 0.01032213,
      "balance_loss_clip": 1.05763912,
      "balance_loss_mlp": 1.02193713,
      "epoch": 0.40257319785967655,
      "flos": 36902738436480.0,
      "grad_norm": 3.6152231855775407,
      "language_loss": 0.79598486,
      "learning_rate": 2.7123462870289848e-06,
      "loss": 0.81814623,
      "num_input_tokens_seen": 72360545,
      "step": 3348,
      "time_per_iteration": 2.9529271125793457
    },
    {
      "auxiliary_loss_clip": 0.01200045,
      "auxiliary_loss_mlp": 0.01037951,
      "balance_loss_clip": 1.06012928,
      "balance_loss_mlp": 1.02834868,
      "epoch": 0.40269344075031566,
      "flos": 24353180703360.0,
      "grad_norm": 1.8067160698774318,
      "language_loss": 0.8133769,
      "learning_rate": 2.711618346348102e-06,
      "loss": 0.8357569,
      "num_input_tokens_seen": 72381070,
      "step": 3349,
      "time_per_iteration": 2.639892101287842
    },
    {
      "auxiliary_loss_clip": 0.01162417,
      "auxiliary_loss_mlp": 0.01035635,
      "balance_loss_clip": 1.05585957,
      "balance_loss_mlp": 1.02667689,
      "epoch": 0.4028136836409547,
      "flos": 14389657614720.0,
      "grad_norm": 1.7625437048339163,
      "language_loss": 0.64004242,
      "learning_rate": 2.7108902977152825e-06,
      "loss": 0.66202295,
      "num_input_tokens_seen": 72398970,
      "step": 3350,
      "time_per_iteration": 2.724085569381714
    },
    {
      "auxiliary_loss_clip": 0.01162377,
      "auxiliary_loss_mlp": 0.01032594,
      "balance_loss_clip": 1.05533659,
      "balance_loss_mlp": 1.02237737,
      "epoch": 0.4029339265315938,
      "flos": 26136037284480.0,
      "grad_norm": 2.081502368953358,
      "language_loss": 0.7478978,
      "learning_rate": 2.7101621412409704e-06,
      "loss": 0.76984751,
      "num_input_tokens_seen": 72418455,
      "step": 3351,
      "time_per_iteration": 2.7307286262512207
    },
    {
      "auxiliary_loss_clip": 0.01184302,
      "auxiliary_loss_mlp": 0.01031004,
      "balance_loss_clip": 1.05818892,
      "balance_loss_mlp": 1.02213478,
      "epoch": 0.40305416942223293,
      "flos": 23256325042560.0,
      "grad_norm": 1.8768529757778822,
      "language_loss": 0.85987353,
      "learning_rate": 2.7094338770356256e-06,
      "loss": 0.88202661,
      "num_input_tokens_seen": 72437540,
      "step": 3352,
      "time_per_iteration": 2.6722209453582764
    },
    {
      "auxiliary_loss_clip": 0.0119898,
      "auxiliary_loss_mlp": 0.01031627,
      "balance_loss_clip": 1.05859768,
      "balance_loss_mlp": 1.02198255,
      "epoch": 0.403174412312872,
      "flos": 27089645506560.0,
      "grad_norm": 3.534817618391811,
      "language_loss": 0.64566356,
      "learning_rate": 2.708705505209726e-06,
      "loss": 0.66796958,
      "num_input_tokens_seen": 72458315,
      "step": 3353,
      "time_per_iteration": 2.7150697708129883
    },
    {
      "auxiliary_loss_clip": 0.01189732,
      "auxiliary_loss_mlp": 0.00762892,
      "balance_loss_clip": 1.05944836,
      "balance_loss_mlp": 1.00102949,
      "epoch": 0.4032946552035111,
      "flos": 21756336065280.0,
      "grad_norm": 2.1771820263577584,
      "language_loss": 0.92056847,
      "learning_rate": 2.7079770258737646e-06,
      "loss": 0.94009471,
      "num_input_tokens_seen": 72476225,
      "step": 3354,
      "time_per_iteration": 2.6556971073150635
    },
    {
      "auxiliary_loss_clip": 0.01202014,
      "auxiliary_loss_mlp": 0.01032976,
      "balance_loss_clip": 1.0597887,
      "balance_loss_mlp": 1.02385616,
      "epoch": 0.4034148980941502,
      "flos": 17343956448000.0,
      "grad_norm": 2.349404503633191,
      "language_loss": 0.75273186,
      "learning_rate": 2.707248439138251e-06,
      "loss": 0.77508175,
      "num_input_tokens_seen": 72492460,
      "step": 3355,
      "time_per_iteration": 2.6148810386657715
    },
    {
      "auxiliary_loss_clip": 0.01182707,
      "auxiliary_loss_mlp": 0.01038401,
      "balance_loss_clip": 1.05958652,
      "balance_loss_mlp": 1.02910829,
      "epoch": 0.40353514098478926,
      "flos": 22017838055040.0,
      "grad_norm": 1.7858104997880015,
      "language_loss": 0.65646422,
      "learning_rate": 2.7065197451137114e-06,
      "loss": 0.67867529,
      "num_input_tokens_seen": 72513840,
      "step": 3356,
      "time_per_iteration": 2.686040163040161
    },
    {
      "auxiliary_loss_clip": 0.01183681,
      "auxiliary_loss_mlp": 0.01029574,
      "balance_loss_clip": 1.05785286,
      "balance_loss_mlp": 1.02106237,
      "epoch": 0.4036553838754284,
      "flos": 14246446089600.0,
      "grad_norm": 2.138746980770998,
      "language_loss": 0.67896545,
      "learning_rate": 2.7057909439106894e-06,
      "loss": 0.70109802,
      "num_input_tokens_seen": 72531695,
      "step": 3357,
      "time_per_iteration": 2.7009215354919434
    },
    {
      "auxiliary_loss_clip": 0.01188936,
      "auxiliary_loss_mlp": 0.01038516,
      "balance_loss_clip": 1.06126118,
      "balance_loss_mlp": 1.02992702,
      "epoch": 0.40377562676606743,
      "flos": 24790644443520.0,
      "grad_norm": 1.836725264397293,
      "language_loss": 0.78634632,
      "learning_rate": 2.7050620356397417e-06,
      "loss": 0.80862087,
      "num_input_tokens_seen": 72550645,
      "step": 3358,
      "time_per_iteration": 2.6575961112976074
    },
    {
      "auxiliary_loss_clip": 0.01198119,
      "auxiliary_loss_mlp": 0.01040351,
      "balance_loss_clip": 1.06162739,
      "balance_loss_mlp": 1.03176856,
      "epoch": 0.40389586965670654,
      "flos": 24061226958720.0,
      "grad_norm": 2.2744861861659342,
      "language_loss": 0.72581404,
      "learning_rate": 2.7043330204114437e-06,
      "loss": 0.74819875,
      "num_input_tokens_seen": 72569355,
      "step": 3359,
      "time_per_iteration": 2.725466728210449
    },
    {
      "auxiliary_loss_clip": 0.01191722,
      "auxiliary_loss_mlp": 0.01033149,
      "balance_loss_clip": 1.05556035,
      "balance_loss_mlp": 1.02433324,
      "epoch": 0.40401611254734565,
      "flos": 16399613934720.0,
      "grad_norm": 2.5446914133220533,
      "language_loss": 0.85708153,
      "learning_rate": 2.7036038983363862e-06,
      "loss": 0.87933028,
      "num_input_tokens_seen": 72585960,
      "step": 3360,
      "time_per_iteration": 3.5492165088653564
    },
    {
      "auxiliary_loss_clip": 0.01172328,
      "auxiliary_loss_mlp": 0.01029593,
      "balance_loss_clip": 1.05450046,
      "balance_loss_mlp": 1.02124786,
      "epoch": 0.4041363554379847,
      "flos": 23988220565760.0,
      "grad_norm": 54.205922948320854,
      "language_loss": 0.84328377,
      "learning_rate": 2.702874669525177e-06,
      "loss": 0.86530304,
      "num_input_tokens_seen": 72604440,
      "step": 3361,
      "time_per_iteration": 2.7672948837280273
    },
    {
      "auxiliary_loss_clip": 0.01188129,
      "auxiliary_loss_mlp": 0.01042039,
      "balance_loss_clip": 1.06095123,
      "balance_loss_mlp": 1.03257382,
      "epoch": 0.4042565983286238,
      "flos": 28401964899840.0,
      "grad_norm": 2.0796939547109825,
      "language_loss": 0.69912368,
      "learning_rate": 2.7021453340884394e-06,
      "loss": 0.72142541,
      "num_input_tokens_seen": 72622165,
      "step": 3362,
      "time_per_iteration": 3.6732499599456787
    },
    {
      "auxiliary_loss_clip": 0.01201433,
      "auxiliary_loss_mlp": 0.01040645,
      "balance_loss_clip": 1.06332827,
      "balance_loss_mlp": 1.03196096,
      "epoch": 0.40437684121926293,
      "flos": 17710963660800.0,
      "grad_norm": 2.7145321900718784,
      "language_loss": 0.73816371,
      "learning_rate": 2.7014158921368125e-06,
      "loss": 0.76058453,
      "num_input_tokens_seen": 72640490,
      "step": 3363,
      "time_per_iteration": 2.5582752227783203
    },
    {
      "auxiliary_loss_clip": 0.0119853,
      "auxiliary_loss_mlp": 0.01031909,
      "balance_loss_clip": 1.05949605,
      "balance_loss_mlp": 1.02315271,
      "epoch": 0.404497084109902,
      "flos": 24018959629440.0,
      "grad_norm": 2.276328100876864,
      "language_loss": 0.85202342,
      "learning_rate": 2.700686343780953e-06,
      "loss": 0.87432778,
      "num_input_tokens_seen": 72660360,
      "step": 3364,
      "time_per_iteration": 3.6581485271453857
    },
    {
      "auxiliary_loss_clip": 0.01201024,
      "auxiliary_loss_mlp": 0.01037304,
      "balance_loss_clip": 1.06123495,
      "balance_loss_mlp": 1.02867317,
      "epoch": 0.4046173270005411,
      "flos": 22929861306240.0,
      "grad_norm": 1.767935097680523,
      "language_loss": 0.88588655,
      "learning_rate": 2.699956689131532e-06,
      "loss": 0.90826982,
      "num_input_tokens_seen": 72680345,
      "step": 3365,
      "time_per_iteration": 2.6688740253448486
    },
    {
      "auxiliary_loss_clip": 0.01181479,
      "auxiliary_loss_mlp": 0.01031919,
      "balance_loss_clip": 1.05724931,
      "balance_loss_mlp": 1.02305031,
      "epoch": 0.4047375698911802,
      "flos": 20668135582080.0,
      "grad_norm": 2.516413559120577,
      "language_loss": 0.85245812,
      "learning_rate": 2.699226928299238e-06,
      "loss": 0.87459207,
      "num_input_tokens_seen": 72698365,
      "step": 3366,
      "time_per_iteration": 2.6402952671051025
    },
    {
      "auxiliary_loss_clip": 0.01198107,
      "auxiliary_loss_mlp": 0.01029758,
      "balance_loss_clip": 1.05957842,
      "balance_loss_mlp": 1.02035809,
      "epoch": 0.40485781278181926,
      "flos": 28912865996160.0,
      "grad_norm": 2.4403624388952845,
      "language_loss": 0.79280859,
      "learning_rate": 2.698497061394774e-06,
      "loss": 0.8150872,
      "num_input_tokens_seen": 72716850,
      "step": 3367,
      "time_per_iteration": 2.7049694061279297
    },
    {
      "auxiliary_loss_clip": 0.0120395,
      "auxiliary_loss_mlp": 0.0103447,
      "balance_loss_clip": 1.06253052,
      "balance_loss_mlp": 1.02525496,
      "epoch": 0.40497805567245837,
      "flos": 23148377694720.0,
      "grad_norm": 1.5890261261378071,
      "language_loss": 0.8096118,
      "learning_rate": 2.6977670885288627e-06,
      "loss": 0.83199596,
      "num_input_tokens_seen": 72738250,
      "step": 3368,
      "time_per_iteration": 3.539931535720825
    },
    {
      "auxiliary_loss_clip": 0.011999,
      "auxiliary_loss_mlp": 0.01031906,
      "balance_loss_clip": 1.05924582,
      "balance_loss_mlp": 1.02235782,
      "epoch": 0.4050982985630975,
      "flos": 16289404030080.0,
      "grad_norm": 2.2936430380448645,
      "language_loss": 0.75570685,
      "learning_rate": 2.6970370098122378e-06,
      "loss": 0.77802491,
      "num_input_tokens_seen": 72755235,
      "step": 3369,
      "time_per_iteration": 2.6351277828216553
    },
    {
      "auxiliary_loss_clip": 0.01182816,
      "auxiliary_loss_mlp": 0.0103719,
      "balance_loss_clip": 1.05700135,
      "balance_loss_mlp": 1.02815437,
      "epoch": 0.40521854145373654,
      "flos": 34459484353920.0,
      "grad_norm": 1.5973134747100541,
      "language_loss": 0.86562216,
      "learning_rate": 2.6963068253556535e-06,
      "loss": 0.88782215,
      "num_input_tokens_seen": 72776620,
      "step": 3370,
      "time_per_iteration": 2.812718391418457
    },
    {
      "auxiliary_loss_clip": 0.01199949,
      "auxiliary_loss_mlp": 0.01031306,
      "balance_loss_clip": 1.05826664,
      "balance_loss_mlp": 1.02160883,
      "epoch": 0.40533878434437565,
      "flos": 25331099454720.0,
      "grad_norm": 6.246420671015426,
      "language_loss": 0.85750008,
      "learning_rate": 2.6955765352698763e-06,
      "loss": 0.8798126,
      "num_input_tokens_seen": 72796765,
      "step": 3371,
      "time_per_iteration": 2.7085378170013428
    },
    {
      "auxiliary_loss_clip": 0.01202378,
      "auxiliary_loss_mlp": 0.00763688,
      "balance_loss_clip": 1.05861986,
      "balance_loss_mlp": 1.00092125,
      "epoch": 0.40545902723501476,
      "flos": 15012061505280.0,
      "grad_norm": 2.6822073802538786,
      "language_loss": 0.73791206,
      "learning_rate": 2.6948461396656923e-06,
      "loss": 0.75757265,
      "num_input_tokens_seen": 72814175,
      "step": 3372,
      "time_per_iteration": 2.6452572345733643
    },
    {
      "auxiliary_loss_clip": 0.01188779,
      "auxiliary_loss_mlp": 0.01027611,
      "balance_loss_clip": 1.05980492,
      "balance_loss_mlp": 1.01799726,
      "epoch": 0.4055792701256538,
      "flos": 25521103422720.0,
      "grad_norm": 2.954266094807206,
      "language_loss": 0.74906379,
      "learning_rate": 2.6941156386539013e-06,
      "loss": 0.77122772,
      "num_input_tokens_seen": 72834125,
      "step": 3373,
      "time_per_iteration": 2.729100465774536
    },
    {
      "auxiliary_loss_clip": 0.01174212,
      "auxiliary_loss_mlp": 0.01038235,
      "balance_loss_clip": 1.05904508,
      "balance_loss_mlp": 1.02874601,
      "epoch": 0.4056995130162929,
      "flos": 19574583972480.0,
      "grad_norm": 2.1893859754974963,
      "language_loss": 0.81038713,
      "learning_rate": 2.6933850323453203e-06,
      "loss": 0.83251154,
      "num_input_tokens_seen": 72852570,
      "step": 3374,
      "time_per_iteration": 2.6937925815582275
    },
    {
      "auxiliary_loss_clip": 0.01200494,
      "auxiliary_loss_mlp": 0.01030533,
      "balance_loss_clip": 1.06180882,
      "balance_loss_mlp": 1.02190828,
      "epoch": 0.405819755906932,
      "flos": 15413794191360.0,
      "grad_norm": 1.867571254495762,
      "language_loss": 0.74932837,
      "learning_rate": 2.6926543208507806e-06,
      "loss": 0.77163863,
      "num_input_tokens_seen": 72871250,
      "step": 3375,
      "time_per_iteration": 2.639474868774414
    },
    {
      "auxiliary_loss_clip": 0.01200029,
      "auxiliary_loss_mlp": 0.01037335,
      "balance_loss_clip": 1.06011128,
      "balance_loss_mlp": 1.02872252,
      "epoch": 0.4059399987975711,
      "flos": 21433930565760.0,
      "grad_norm": 2.1662133523992986,
      "language_loss": 0.80841732,
      "learning_rate": 2.6919235042811316e-06,
      "loss": 0.830791,
      "num_input_tokens_seen": 72890035,
      "step": 3376,
      "time_per_iteration": 2.6274492740631104
    },
    {
      "auxiliary_loss_clip": 0.01201045,
      "auxiliary_loss_mlp": 0.0103112,
      "balance_loss_clip": 1.06030917,
      "balance_loss_mlp": 1.02126741,
      "epoch": 0.4060602416882102,
      "flos": 25556942217600.0,
      "grad_norm": 2.1905721158653164,
      "language_loss": 0.76531225,
      "learning_rate": 2.691192582747237e-06,
      "loss": 0.7876339,
      "num_input_tokens_seen": 72909665,
      "step": 3377,
      "time_per_iteration": 2.589175224304199
    },
    {
      "auxiliary_loss_clip": 0.01172935,
      "auxiliary_loss_mlp": 0.01036629,
      "balance_loss_clip": 1.05259728,
      "balance_loss_mlp": 1.02723587,
      "epoch": 0.40618048457884925,
      "flos": 23766759262080.0,
      "grad_norm": 3.4382142681168353,
      "language_loss": 0.74436098,
      "learning_rate": 2.6904615563599765e-06,
      "loss": 0.7664566,
      "num_input_tokens_seen": 72929465,
      "step": 3378,
      "time_per_iteration": 2.737421751022339
    },
    {
      "auxiliary_loss_clip": 0.0119573,
      "auxiliary_loss_mlp": 0.01030793,
      "balance_loss_clip": 1.05695891,
      "balance_loss_mlp": 1.02187014,
      "epoch": 0.40630072746948837,
      "flos": 17639681120640.0,
      "grad_norm": 1.878329114845346,
      "language_loss": 0.83559918,
      "learning_rate": 2.6897304252302477e-06,
      "loss": 0.85786438,
      "num_input_tokens_seen": 72946785,
      "step": 3379,
      "time_per_iteration": 2.582448959350586
    },
    {
      "auxiliary_loss_clip": 0.01081006,
      "auxiliary_loss_mlp": 0.01005062,
      "balance_loss_clip": 1.01054096,
      "balance_loss_mlp": 1.00375676,
      "epoch": 0.4064209703601275,
      "flos": 60836053063680.0,
      "grad_norm": 0.7883996699406043,
      "language_loss": 0.5474816,
      "learning_rate": 2.688999189468962e-06,
      "loss": 0.56834227,
      "num_input_tokens_seen": 73003215,
      "step": 3380,
      "time_per_iteration": 3.07786226272583
    },
    {
      "auxiliary_loss_clip": 0.01186587,
      "auxiliary_loss_mlp": 0.01032786,
      "balance_loss_clip": 1.0567292,
      "balance_loss_mlp": 1.02377987,
      "epoch": 0.40654121325076653,
      "flos": 24024346669440.0,
      "grad_norm": 2.5362969613597586,
      "language_loss": 0.76747429,
      "learning_rate": 2.6882678491870464e-06,
      "loss": 0.78966796,
      "num_input_tokens_seen": 73023650,
      "step": 3381,
      "time_per_iteration": 2.6696934700012207
    },
    {
      "auxiliary_loss_clip": 0.0118789,
      "auxiliary_loss_mlp": 0.01032768,
      "balance_loss_clip": 1.05937171,
      "balance_loss_mlp": 1.02286148,
      "epoch": 0.40666145614140564,
      "flos": 27344252085120.0,
      "grad_norm": 2.1010314727121955,
      "language_loss": 0.713153,
      "learning_rate": 2.6875364044954453e-06,
      "loss": 0.73535961,
      "num_input_tokens_seen": 73043880,
      "step": 3382,
      "time_per_iteration": 2.6786446571350098
    },
    {
      "auxiliary_loss_clip": 0.01176915,
      "auxiliary_loss_mlp": 0.01034623,
      "balance_loss_clip": 1.05693674,
      "balance_loss_mlp": 1.02621317,
      "epoch": 0.40678169903204475,
      "flos": 26176724415360.0,
      "grad_norm": 3.1090552460891288,
      "language_loss": 0.82562864,
      "learning_rate": 2.6868048555051185e-06,
      "loss": 0.84774405,
      "num_input_tokens_seen": 73065410,
      "step": 3383,
      "time_per_iteration": 2.7434616088867188
    },
    {
      "auxiliary_loss_clip": 0.01189654,
      "auxiliary_loss_mlp": 0.00763363,
      "balance_loss_clip": 1.05819869,
      "balance_loss_mlp": 1.00087142,
      "epoch": 0.4069019419226838,
      "flos": 28622420622720.0,
      "grad_norm": 2.6541750638281942,
      "language_loss": 0.86201668,
      "learning_rate": 2.686073202327041e-06,
      "loss": 0.88154685,
      "num_input_tokens_seen": 73084410,
      "step": 3384,
      "time_per_iteration": 2.7417898178100586
    },
    {
      "auxiliary_loss_clip": 0.01196932,
      "auxiliary_loss_mlp": 0.01030679,
      "balance_loss_clip": 1.05965781,
      "balance_loss_mlp": 1.02241206,
      "epoch": 0.4070221848133229,
      "flos": 25229006023680.0,
      "grad_norm": 1.7639388208116098,
      "language_loss": 0.7406016,
      "learning_rate": 2.6853414450722043e-06,
      "loss": 0.7628777,
      "num_input_tokens_seen": 73104075,
      "step": 3385,
      "time_per_iteration": 2.745769500732422
    },
    {
      "auxiliary_loss_clip": 0.01179176,
      "auxiliary_loss_mlp": 0.01035665,
      "balance_loss_clip": 1.05734396,
      "balance_loss_mlp": 1.02701616,
      "epoch": 0.40714242770396203,
      "flos": 18405224709120.0,
      "grad_norm": 1.7599980360141647,
      "language_loss": 0.85700125,
      "learning_rate": 2.684609583851616e-06,
      "loss": 0.87914968,
      "num_input_tokens_seen": 73122250,
      "step": 3386,
      "time_per_iteration": 2.6691019535064697
    },
    {
      "auxiliary_loss_clip": 0.01180829,
      "auxiliary_loss_mlp": 0.01026816,
      "balance_loss_clip": 1.0567677,
      "balance_loss_mlp": 1.01813149,
      "epoch": 0.4072626705946011,
      "flos": 30228920403840.0,
      "grad_norm": 1.7280161969307568,
      "language_loss": 0.80747128,
      "learning_rate": 2.683877618776297e-06,
      "loss": 0.82954764,
      "num_input_tokens_seen": 73144505,
      "step": 3387,
      "time_per_iteration": 3.7219185829162598
    },
    {
      "auxiliary_loss_clip": 0.01177159,
      "auxiliary_loss_mlp": 0.00763693,
      "balance_loss_clip": 1.05534434,
      "balance_loss_mlp": 1.00086308,
      "epoch": 0.4073829134852402,
      "flos": 21834549930240.0,
      "grad_norm": 6.324110375550757,
      "language_loss": 0.74422508,
      "learning_rate": 2.6831455499572876e-06,
      "loss": 0.76363361,
      "num_input_tokens_seen": 73162440,
      "step": 3388,
      "time_per_iteration": 3.6827101707458496
    },
    {
      "auxiliary_loss_clip": 0.01196914,
      "auxiliary_loss_mlp": 0.01030815,
      "balance_loss_clip": 1.05659378,
      "balance_loss_mlp": 1.02142715,
      "epoch": 0.40750315637587925,
      "flos": 25260211964160.0,
      "grad_norm": 2.455688905601451,
      "language_loss": 0.77951431,
      "learning_rate": 2.682413377505641e-06,
      "loss": 0.80179161,
      "num_input_tokens_seen": 73181245,
      "step": 3389,
      "time_per_iteration": 2.6281914710998535
    },
    {
      "auxiliary_loss_clip": 0.01181974,
      "auxiliary_loss_mlp": 0.01026949,
      "balance_loss_clip": 1.05576229,
      "balance_loss_mlp": 1.01828301,
      "epoch": 0.40762339926651836,
      "flos": 19712767593600.0,
      "grad_norm": 2.0007609995491973,
      "language_loss": 0.76734316,
      "learning_rate": 2.6816811015324284e-06,
      "loss": 0.78943247,
      "num_input_tokens_seen": 73199295,
      "step": 3390,
      "time_per_iteration": 2.6705574989318848
    },
    {
      "auxiliary_loss_clip": 0.01063995,
      "auxiliary_loss_mlp": 0.01003895,
      "balance_loss_clip": 1.01053095,
      "balance_loss_mlp": 1.00248241,
      "epoch": 0.40774364215715747,
      "flos": 71449307314560.0,
      "grad_norm": 0.7266829368043991,
      "language_loss": 0.56593132,
      "learning_rate": 2.6809487221487343e-06,
      "loss": 0.5866102,
      "num_input_tokens_seen": 73258780,
      "step": 3391,
      "time_per_iteration": 4.161775350570679
    },
    {
      "auxiliary_loss_clip": 0.01196109,
      "auxiliary_loss_mlp": 0.01035051,
      "balance_loss_clip": 1.05666876,
      "balance_loss_mlp": 1.02600312,
      "epoch": 0.4078638850477965,
      "flos": 15084134144640.0,
      "grad_norm": 3.757024809635586,
      "language_loss": 0.82177591,
      "learning_rate": 2.6802162394656605e-06,
      "loss": 0.84408748,
      "num_input_tokens_seen": 73275490,
      "step": 3392,
      "time_per_iteration": 2.6456120014190674
    },
    {
      "auxiliary_loss_clip": 0.01195303,
      "auxiliary_loss_mlp": 0.01030168,
      "balance_loss_clip": 1.05701017,
      "balance_loss_mlp": 1.02090549,
      "epoch": 0.40798412793843564,
      "flos": 23842890138240.0,
      "grad_norm": 2.004636123422568,
      "language_loss": 0.72232819,
      "learning_rate": 2.679483653594324e-06,
      "loss": 0.74458289,
      "num_input_tokens_seen": 73297260,
      "step": 3393,
      "time_per_iteration": 2.6887247562408447
    },
    {
      "auxiliary_loss_clip": 0.01160229,
      "auxiliary_loss_mlp": 0.01034568,
      "balance_loss_clip": 1.05004811,
      "balance_loss_mlp": 1.02551436,
      "epoch": 0.40810437082907475,
      "flos": 21065774117760.0,
      "grad_norm": 2.6585203285241694,
      "language_loss": 0.76704091,
      "learning_rate": 2.678750964645857e-06,
      "loss": 0.78898889,
      "num_input_tokens_seen": 73316340,
      "step": 3394,
      "time_per_iteration": 3.6572325229644775
    },
    {
      "auxiliary_loss_clip": 0.01198388,
      "auxiliary_loss_mlp": 0.01036688,
      "balance_loss_clip": 1.05888748,
      "balance_loss_mlp": 1.02696061,
      "epoch": 0.4082246137197138,
      "flos": 11321377948800.0,
      "grad_norm": 2.9375797356926503,
      "language_loss": 0.83439773,
      "learning_rate": 2.6780181727314094e-06,
      "loss": 0.85674846,
      "num_input_tokens_seen": 73331245,
      "step": 3395,
      "time_per_iteration": 2.6109235286712646
    },
    {
      "auxiliary_loss_clip": 0.01186238,
      "auxiliary_loss_mlp": 0.0103501,
      "balance_loss_clip": 1.05807137,
      "balance_loss_mlp": 1.02605736,
      "epoch": 0.4083448566103529,
      "flos": 19062569554560.0,
      "grad_norm": 1.8421774439516905,
      "language_loss": 0.78110683,
      "learning_rate": 2.6772852779621435e-06,
      "loss": 0.80331933,
      "num_input_tokens_seen": 73349105,
      "step": 3396,
      "time_per_iteration": 2.668492078781128
    },
    {
      "auxiliary_loss_clip": 0.01192292,
      "auxiliary_loss_mlp": 0.01028556,
      "balance_loss_clip": 1.05764341,
      "balance_loss_mlp": 1.02030075,
      "epoch": 0.408465099500992,
      "flos": 23550254035200.0,
      "grad_norm": 2.474774369048925,
      "language_loss": 0.86626422,
      "learning_rate": 2.676552280449239e-06,
      "loss": 0.88847268,
      "num_input_tokens_seen": 73368990,
      "step": 3397,
      "time_per_iteration": 2.6638669967651367
    },
    {
      "auxiliary_loss_clip": 0.01197476,
      "auxiliary_loss_mlp": 0.01037329,
      "balance_loss_clip": 1.05828166,
      "balance_loss_mlp": 1.02728009,
      "epoch": 0.4085853423916311,
      "flos": 12750012558720.0,
      "grad_norm": 2.526380905380863,
      "language_loss": 0.75871801,
      "learning_rate": 2.6758191803038917e-06,
      "loss": 0.78106612,
      "num_input_tokens_seen": 73387485,
      "step": 3398,
      "time_per_iteration": 2.57731294631958
    },
    {
      "auxiliary_loss_clip": 0.01191243,
      "auxiliary_loss_mlp": 0.01037172,
      "balance_loss_clip": 1.05982089,
      "balance_loss_mlp": 1.02736151,
      "epoch": 0.4087055852822702,
      "flos": 24353072962560.0,
      "grad_norm": 1.767155900036683,
      "language_loss": 0.83148211,
      "learning_rate": 2.6750859776373125e-06,
      "loss": 0.8537662,
      "num_input_tokens_seen": 73406940,
      "step": 3399,
      "time_per_iteration": 2.7179501056671143
    },
    {
      "auxiliary_loss_clip": 0.01047813,
      "auxiliary_loss_mlp": 0.01003188,
      "balance_loss_clip": 1.01241243,
      "balance_loss_mlp": 1.00173378,
      "epoch": 0.4088258281729093,
      "flos": 66387950720640.0,
      "grad_norm": 0.7684337492084602,
      "language_loss": 0.60317165,
      "learning_rate": 2.674352672560727e-06,
      "loss": 0.62368166,
      "num_input_tokens_seen": 73468385,
      "step": 3400,
      "time_per_iteration": 3.29522705078125
    },
    {
      "auxiliary_loss_clip": 0.0117087,
      "auxiliary_loss_mlp": 0.01030092,
      "balance_loss_clip": 1.05258393,
      "balance_loss_mlp": 1.02080584,
      "epoch": 0.40894607106354836,
      "flos": 20449260057600.0,
      "grad_norm": 4.609941947268175,
      "language_loss": 0.7689687,
      "learning_rate": 2.673619265185377e-06,
      "loss": 0.79097837,
      "num_input_tokens_seen": 73488225,
      "step": 3401,
      "time_per_iteration": 2.761308193206787
    },
    {
      "auxiliary_loss_clip": 0.01185312,
      "auxiliary_loss_mlp": 0.01035504,
      "balance_loss_clip": 1.05997133,
      "balance_loss_mlp": 1.02672446,
      "epoch": 0.40906631395418747,
      "flos": 27053627143680.0,
      "grad_norm": 1.745331465261908,
      "language_loss": 0.78423333,
      "learning_rate": 2.672885755622521e-06,
      "loss": 0.80644155,
      "num_input_tokens_seen": 73510640,
      "step": 3402,
      "time_per_iteration": 2.729522228240967
    },
    {
      "auxiliary_loss_clip": 0.01193294,
      "auxiliary_loss_mlp": 0.01037232,
      "balance_loss_clip": 1.05574453,
      "balance_loss_mlp": 1.02894139,
      "epoch": 0.4091865568448266,
      "flos": 25484151306240.0,
      "grad_norm": 2.6404591273347005,
      "language_loss": 0.71229213,
      "learning_rate": 2.67215214398343e-06,
      "loss": 0.73459744,
      "num_input_tokens_seen": 73530655,
      "step": 3403,
      "time_per_iteration": 2.647007942199707
    },
    {
      "auxiliary_loss_clip": 0.01181677,
      "auxiliary_loss_mlp": 0.01031069,
      "balance_loss_clip": 1.05633783,
      "balance_loss_mlp": 1.02189016,
      "epoch": 0.40930679973546563,
      "flos": 28657864368000.0,
      "grad_norm": 2.8145400071553777,
      "language_loss": 0.78143871,
      "learning_rate": 2.671418430379393e-06,
      "loss": 0.80356622,
      "num_input_tokens_seen": 73549340,
      "step": 3404,
      "time_per_iteration": 2.722517728805542
    },
    {
      "auxiliary_loss_clip": 0.01157532,
      "auxiliary_loss_mlp": 0.01033969,
      "balance_loss_clip": 1.0512048,
      "balance_loss_mlp": 1.0249629,
      "epoch": 0.40942704262610474,
      "flos": 20886292834560.0,
      "grad_norm": 1.7996750827140426,
      "language_loss": 0.83746719,
      "learning_rate": 2.670684614921715e-06,
      "loss": 0.85938227,
      "num_input_tokens_seen": 73568315,
      "step": 3405,
      "time_per_iteration": 2.7220776081085205
    },
    {
      "auxiliary_loss_clip": 0.01173587,
      "auxiliary_loss_mlp": 0.01033379,
      "balance_loss_clip": 1.05988932,
      "balance_loss_mlp": 1.02408028,
      "epoch": 0.4095472855167438,
      "flos": 21618080616960.0,
      "grad_norm": 3.8219049922762327,
      "language_loss": 0.69400382,
      "learning_rate": 2.6699506977217128e-06,
      "loss": 0.71607351,
      "num_input_tokens_seen": 73588490,
      "step": 3406,
      "time_per_iteration": 2.834841251373291
    },
    {
      "auxiliary_loss_clip": 0.01196989,
      "auxiliary_loss_mlp": 0.01032564,
      "balance_loss_clip": 1.05937505,
      "balance_loss_mlp": 1.02358162,
      "epoch": 0.4096675284073829,
      "flos": 27926112499200.0,
      "grad_norm": 2.3515780001180153,
      "language_loss": 0.70126641,
      "learning_rate": 2.6692166788907233e-06,
      "loss": 0.72356188,
      "num_input_tokens_seen": 73608685,
      "step": 3407,
      "time_per_iteration": 2.719503164291382
    },
    {
      "auxiliary_loss_clip": 0.01199493,
      "auxiliary_loss_mlp": 0.01035784,
      "balance_loss_clip": 1.06003356,
      "balance_loss_mlp": 1.02637815,
      "epoch": 0.409787771298022,
      "flos": 19206607092480.0,
      "grad_norm": 2.593583742600869,
      "language_loss": 0.76811922,
      "learning_rate": 2.6684825585400957e-06,
      "loss": 0.79047191,
      "num_input_tokens_seen": 73627630,
      "step": 3408,
      "time_per_iteration": 2.6007895469665527
    },
    {
      "auxiliary_loss_clip": 0.01083932,
      "auxiliary_loss_mlp": 0.01004919,
      "balance_loss_clip": 1.01333022,
      "balance_loss_mlp": 1.0037148,
      "epoch": 0.4099080141886611,
      "flos": 59269234832640.0,
      "grad_norm": 0.8204252630959616,
      "language_loss": 0.65117216,
      "learning_rate": 2.6677483367811947e-06,
      "loss": 0.67206073,
      "num_input_tokens_seen": 73687670,
      "step": 3409,
      "time_per_iteration": 3.2937324047088623
    },
    {
      "auxiliary_loss_clip": 0.01180665,
      "auxiliary_loss_mlp": 0.01027323,
      "balance_loss_clip": 1.05642891,
      "balance_loss_mlp": 1.01837027,
      "epoch": 0.4100282570793002,
      "flos": 21906443001600.0,
      "grad_norm": 1.8403485141745353,
      "language_loss": 0.75805992,
      "learning_rate": 2.6670140137254028e-06,
      "loss": 0.78013974,
      "num_input_tokens_seen": 73707145,
      "step": 3410,
      "time_per_iteration": 2.6122026443481445
    },
    {
      "auxiliary_loss_clip": 0.01194036,
      "auxiliary_loss_mlp": 0.010438,
      "balance_loss_clip": 1.05643034,
      "balance_loss_mlp": 1.03460956,
      "epoch": 0.4101484999699393,
      "flos": 18551596631040.0,
      "grad_norm": 2.524550802895484,
      "language_loss": 0.89575976,
      "learning_rate": 2.666279589484115e-06,
      "loss": 0.91813815,
      "num_input_tokens_seen": 73725045,
      "step": 3411,
      "time_per_iteration": 2.6610946655273438
    },
    {
      "auxiliary_loss_clip": 0.01186414,
      "auxiliary_loss_mlp": 0.01034917,
      "balance_loss_clip": 1.05915296,
      "balance_loss_mlp": 1.02552938,
      "epoch": 0.41026874286057835,
      "flos": 19094529680640.0,
      "grad_norm": 3.031732361295017,
      "language_loss": 0.81208539,
      "learning_rate": 2.6655450641687435e-06,
      "loss": 0.83429873,
      "num_input_tokens_seen": 73742610,
      "step": 3412,
      "time_per_iteration": 2.6067190170288086
    },
    {
      "auxiliary_loss_clip": 0.01180316,
      "auxiliary_loss_mlp": 0.01032505,
      "balance_loss_clip": 1.05850613,
      "balance_loss_mlp": 1.02404666,
      "epoch": 0.41038898575121746,
      "flos": 31209568588800.0,
      "grad_norm": 1.7902553524296116,
      "language_loss": 0.69395232,
      "learning_rate": 2.664810437890715e-06,
      "loss": 0.71608055,
      "num_input_tokens_seen": 73764280,
      "step": 3413,
      "time_per_iteration": 3.642207622528076
    },
    {
      "auxiliary_loss_clip": 0.01148264,
      "auxiliary_loss_mlp": 0.01038373,
      "balance_loss_clip": 1.05522108,
      "balance_loss_mlp": 1.02939069,
      "epoch": 0.41050922864185657,
      "flos": 14355865895040.0,
      "grad_norm": 2.221982401314438,
      "language_loss": 0.8004688,
      "learning_rate": 2.6640757107614714e-06,
      "loss": 0.82233518,
      "num_input_tokens_seen": 73782375,
      "step": 3414,
      "time_per_iteration": 3.7002830505371094
    },
    {
      "auxiliary_loss_clip": 0.0119846,
      "auxiliary_loss_mlp": 0.01027916,
      "balance_loss_clip": 1.05996823,
      "balance_loss_mlp": 1.0187788,
      "epoch": 0.4106294715324956,
      "flos": 30956290813440.0,
      "grad_norm": 2.2755084659417855,
      "language_loss": 0.68798923,
      "learning_rate": 2.6633408828924697e-06,
      "loss": 0.71025294,
      "num_input_tokens_seen": 73801240,
      "step": 3415,
      "time_per_iteration": 2.704834461212158
    },
    {
      "auxiliary_loss_clip": 0.01185435,
      "auxiliary_loss_mlp": 0.01037411,
      "balance_loss_clip": 1.05727255,
      "balance_loss_mlp": 1.02804708,
      "epoch": 0.41074971442313474,
      "flos": 24457321209600.0,
      "grad_norm": 2.23302327282162,
      "language_loss": 0.70328724,
      "learning_rate": 2.662605954395185e-06,
      "loss": 0.72551572,
      "num_input_tokens_seen": 73821200,
      "step": 3416,
      "time_per_iteration": 2.775259256362915
    },
    {
      "auxiliary_loss_clip": 0.01197308,
      "auxiliary_loss_mlp": 0.01026645,
      "balance_loss_clip": 1.06013298,
      "balance_loss_mlp": 1.01842558,
      "epoch": 0.41086995731377385,
      "flos": 21542991235200.0,
      "grad_norm": 1.7444048299211736,
      "language_loss": 0.83681023,
      "learning_rate": 2.6618709253811027e-06,
      "loss": 0.8590498,
      "num_input_tokens_seen": 73840655,
      "step": 3417,
      "time_per_iteration": 3.7861480712890625
    },
    {
      "auxiliary_loss_clip": 0.0115547,
      "auxiliary_loss_mlp": 0.01036603,
      "balance_loss_clip": 1.05249679,
      "balance_loss_mlp": 1.02853274,
      "epoch": 0.4109902002044129,
      "flos": 20702753314560.0,
      "grad_norm": 4.463631623269803,
      "language_loss": 0.88079906,
      "learning_rate": 2.6611357959617277e-06,
      "loss": 0.90271974,
      "num_input_tokens_seen": 73860275,
      "step": 3418,
      "time_per_iteration": 2.701294422149658
    },
    {
      "auxiliary_loss_clip": 0.01178462,
      "auxiliary_loss_mlp": 0.01027538,
      "balance_loss_clip": 1.05614507,
      "balance_loss_mlp": 1.01871085,
      "epoch": 0.411110443095052,
      "flos": 18179992477440.0,
      "grad_norm": 1.8684881991517428,
      "language_loss": 0.90876299,
      "learning_rate": 2.660400566248578e-06,
      "loss": 0.93082297,
      "num_input_tokens_seen": 73878400,
      "step": 3419,
      "time_per_iteration": 2.6668992042541504
    },
    {
      "auxiliary_loss_clip": 0.01182638,
      "auxiliary_loss_mlp": 0.01033905,
      "balance_loss_clip": 1.05586147,
      "balance_loss_mlp": 1.02467787,
      "epoch": 0.41123068598569107,
      "flos": 14575244209920.0,
      "grad_norm": 2.781168944459553,
      "language_loss": 0.67039967,
      "learning_rate": 2.6596652363531876e-06,
      "loss": 0.69256508,
      "num_input_tokens_seen": 73894275,
      "step": 3420,
      "time_per_iteration": 2.5999996662139893
    },
    {
      "auxiliary_loss_clip": 0.01166307,
      "auxiliary_loss_mlp": 0.0103307,
      "balance_loss_clip": 1.05398774,
      "balance_loss_mlp": 1.02388477,
      "epoch": 0.4113509288763302,
      "flos": 21177995184000.0,
      "grad_norm": 1.6662914878107316,
      "language_loss": 0.78673792,
      "learning_rate": 2.6589298063871055e-06,
      "loss": 0.80873168,
      "num_input_tokens_seen": 73914450,
      "step": 3421,
      "time_per_iteration": 3.6264560222625732
    },
    {
      "auxiliary_loss_clip": 0.01194153,
      "auxiliary_loss_mlp": 0.01036619,
      "balance_loss_clip": 1.05843878,
      "balance_loss_mlp": 1.02774954,
      "epoch": 0.4114711717669693,
      "flos": 18442212739200.0,
      "grad_norm": 1.8974760647570041,
      "language_loss": 0.70036161,
      "learning_rate": 2.658194276461895e-06,
      "loss": 0.7226693,
      "num_input_tokens_seen": 73932375,
      "step": 3422,
      "time_per_iteration": 2.579345703125
    },
    {
      "auxiliary_loss_clip": 0.01183893,
      "auxiliary_loss_mlp": 0.01034257,
      "balance_loss_clip": 1.05656266,
      "balance_loss_mlp": 1.02500081,
      "epoch": 0.41159141465760835,
      "flos": 27233395735680.0,
      "grad_norm": 4.694886463658511,
      "language_loss": 0.6730361,
      "learning_rate": 2.6574586466891368e-06,
      "loss": 0.69521761,
      "num_input_tokens_seen": 73952850,
      "step": 3423,
      "time_per_iteration": 2.721099615097046
    },
    {
      "auxiliary_loss_clip": 0.01191914,
      "auxiliary_loss_mlp": 0.01030747,
      "balance_loss_clip": 1.05770874,
      "balance_loss_mlp": 1.02242625,
      "epoch": 0.41171165754824746,
      "flos": 20006876154240.0,
      "grad_norm": 1.9590233610204664,
      "language_loss": 0.64903599,
      "learning_rate": 2.6567229171804247e-06,
      "loss": 0.67126262,
      "num_input_tokens_seen": 73970735,
      "step": 3424,
      "time_per_iteration": 2.6295340061187744
    },
    {
      "auxiliary_loss_clip": 0.01178334,
      "auxiliary_loss_mlp": 0.0103041,
      "balance_loss_clip": 1.05261135,
      "balance_loss_mlp": 1.02121329,
      "epoch": 0.41183190043888657,
      "flos": 18004318035840.0,
      "grad_norm": 8.006451597753085,
      "language_loss": 0.87870544,
      "learning_rate": 2.655987088047368e-06,
      "loss": 0.90079296,
      "num_input_tokens_seen": 73989080,
      "step": 3425,
      "time_per_iteration": 2.6927595138549805
    },
    {
      "auxiliary_loss_clip": 0.01187051,
      "auxiliary_loss_mlp": 0.01032176,
      "balance_loss_clip": 1.05847526,
      "balance_loss_mlp": 1.02310443,
      "epoch": 0.4119521433295256,
      "flos": 27163370171520.0,
      "grad_norm": 1.8660285103383591,
      "language_loss": 0.78818864,
      "learning_rate": 2.6552511594015912e-06,
      "loss": 0.81038088,
      "num_input_tokens_seen": 74009470,
      "step": 3426,
      "time_per_iteration": 2.7031843662261963
    },
    {
      "auxiliary_loss_clip": 0.01168735,
      "auxiliary_loss_mlp": 0.01029349,
      "balance_loss_clip": 1.05568385,
      "balance_loss_mlp": 1.0201993,
      "epoch": 0.41207238622016473,
      "flos": 15122020014720.0,
      "grad_norm": 2.4917219776319315,
      "language_loss": 0.85862839,
      "learning_rate": 2.654515131354735e-06,
      "loss": 0.88060927,
      "num_input_tokens_seen": 74027735,
      "step": 3427,
      "time_per_iteration": 2.6213412284851074
    },
    {
      "auxiliary_loss_clip": 0.0118371,
      "auxiliary_loss_mlp": 0.01044711,
      "balance_loss_clip": 1.05780101,
      "balance_loss_mlp": 1.03546643,
      "epoch": 0.41219262911080384,
      "flos": 27052872958080.0,
      "grad_norm": 2.082402741065846,
      "language_loss": 0.85690045,
      "learning_rate": 2.653779004018453e-06,
      "loss": 0.87918472,
      "num_input_tokens_seen": 74048300,
      "step": 3428,
      "time_per_iteration": 2.747657299041748
    },
    {
      "auxiliary_loss_clip": 0.01156784,
      "auxiliary_loss_mlp": 0.01033492,
      "balance_loss_clip": 1.05080235,
      "balance_loss_mlp": 1.02443767,
      "epoch": 0.4123128720014429,
      "flos": 24686360282880.0,
      "grad_norm": 1.9639072933942026,
      "language_loss": 0.82765758,
      "learning_rate": 2.653042777504417e-06,
      "loss": 0.84956038,
      "num_input_tokens_seen": 74070890,
      "step": 3429,
      "time_per_iteration": 2.7356672286987305
    },
    {
      "auxiliary_loss_clip": 0.01185383,
      "auxiliary_loss_mlp": 0.01032127,
      "balance_loss_clip": 1.05635774,
      "balance_loss_mlp": 1.02195835,
      "epoch": 0.412433114892082,
      "flos": 26244774731520.0,
      "grad_norm": 3.1844535569151673,
      "language_loss": 0.80314267,
      "learning_rate": 2.6523064519243105e-06,
      "loss": 0.8253178,
      "num_input_tokens_seen": 74090460,
      "step": 3430,
      "time_per_iteration": 2.7467761039733887
    },
    {
      "auxiliary_loss_clip": 0.01185258,
      "auxiliary_loss_mlp": 0.01036164,
      "balance_loss_clip": 1.05807292,
      "balance_loss_mlp": 1.02700853,
      "epoch": 0.4125533577827211,
      "flos": 21361031913600.0,
      "grad_norm": 2.9254520204106034,
      "language_loss": 0.7902385,
      "learning_rate": 2.6515700273898333e-06,
      "loss": 0.81245273,
      "num_input_tokens_seen": 74108335,
      "step": 3431,
      "time_per_iteration": 2.652951955795288
    },
    {
      "auxiliary_loss_clip": 0.01196213,
      "auxiliary_loss_mlp": 0.01030904,
      "balance_loss_clip": 1.0578692,
      "balance_loss_mlp": 1.02200508,
      "epoch": 0.4126736006733602,
      "flos": 26067556005120.0,
      "grad_norm": 3.5618916738685016,
      "language_loss": 0.69194025,
      "learning_rate": 2.6508335040127018e-06,
      "loss": 0.71421134,
      "num_input_tokens_seen": 74128030,
      "step": 3432,
      "time_per_iteration": 2.803131341934204
    },
    {
      "auxiliary_loss_clip": 0.01198916,
      "auxiliary_loss_mlp": 0.01045207,
      "balance_loss_clip": 1.06047058,
      "balance_loss_mlp": 1.0357542,
      "epoch": 0.4127938435639993,
      "flos": 25666146541440.0,
      "grad_norm": 1.4479206117375987,
      "language_loss": 0.77128643,
      "learning_rate": 2.6500968819046446e-06,
      "loss": 0.79372764,
      "num_input_tokens_seen": 74148330,
      "step": 3433,
      "time_per_iteration": 2.644843101501465
    },
    {
      "auxiliary_loss_clip": 0.01146415,
      "auxiliary_loss_mlp": 0.0103037,
      "balance_loss_clip": 1.04978347,
      "balance_loss_mlp": 1.02137029,
      "epoch": 0.4129140864546384,
      "flos": 17995914253440.0,
      "grad_norm": 3.1748489150000565,
      "language_loss": 0.59410119,
      "learning_rate": 2.649360161177408e-06,
      "loss": 0.61586899,
      "num_input_tokens_seen": 74163390,
      "step": 3434,
      "time_per_iteration": 2.6807284355163574
    },
    {
      "auxiliary_loss_clip": 0.01116097,
      "auxiliary_loss_mlp": 0.01030218,
      "balance_loss_clip": 1.04639173,
      "balance_loss_mlp": 1.02052641,
      "epoch": 0.41303432934527745,
      "flos": 23732895715200.0,
      "grad_norm": 1.8535678206711756,
      "language_loss": 0.73612189,
      "learning_rate": 2.6486233419427504e-06,
      "loss": 0.75758505,
      "num_input_tokens_seen": 74183205,
      "step": 3435,
      "time_per_iteration": 2.791300058364868
    },
    {
      "auxiliary_loss_clip": 0.01191898,
      "auxiliary_loss_mlp": 0.01034323,
      "balance_loss_clip": 1.05897284,
      "balance_loss_mlp": 1.0249474,
      "epoch": 0.41315457223591656,
      "flos": 19755286318080.0,
      "grad_norm": 2.3223860357843495,
      "language_loss": 0.75306439,
      "learning_rate": 2.6478864243124484e-06,
      "loss": 0.77532661,
      "num_input_tokens_seen": 74202870,
      "step": 3436,
      "time_per_iteration": 2.7255282402038574
    },
    {
      "auxiliary_loss_clip": 0.01167344,
      "auxiliary_loss_mlp": 0.0102879,
      "balance_loss_clip": 1.05767727,
      "balance_loss_mlp": 1.01993895,
      "epoch": 0.4132748151265556,
      "flos": 20923316778240.0,
      "grad_norm": 2.914720239432902,
      "language_loss": 0.85184264,
      "learning_rate": 2.6471494083982903e-06,
      "loss": 0.87380397,
      "num_input_tokens_seen": 74222255,
      "step": 3437,
      "time_per_iteration": 2.795156717300415
    },
    {
      "auxiliary_loss_clip": 0.01199196,
      "auxiliary_loss_mlp": 0.01030391,
      "balance_loss_clip": 1.06032598,
      "balance_loss_mlp": 1.02052617,
      "epoch": 0.4133950580171947,
      "flos": 32232520016640.0,
      "grad_norm": 1.9561177156234375,
      "language_loss": 0.75075889,
      "learning_rate": 2.6464122943120818e-06,
      "loss": 0.77305484,
      "num_input_tokens_seen": 74242480,
      "step": 3438,
      "time_per_iteration": 2.7116167545318604
    },
    {
      "auxiliary_loss_clip": 0.0116502,
      "auxiliary_loss_mlp": 0.01033984,
      "balance_loss_clip": 1.05488133,
      "balance_loss_mlp": 1.02513909,
      "epoch": 0.41351530090783384,
      "flos": 23292487059840.0,
      "grad_norm": 4.1145097484316535,
      "language_loss": 0.82390463,
      "learning_rate": 2.645675082165642e-06,
      "loss": 0.84589469,
      "num_input_tokens_seen": 74258690,
      "step": 3439,
      "time_per_iteration": 3.689728260040283
    },
    {
      "auxiliary_loss_clip": 0.01199123,
      "auxiliary_loss_mlp": 0.01033544,
      "balance_loss_clip": 1.05859971,
      "balance_loss_mlp": 1.02467489,
      "epoch": 0.4136355437984729,
      "flos": 25593571111680.0,
      "grad_norm": 4.106919172072876,
      "language_loss": 0.75277746,
      "learning_rate": 2.644937772070806e-06,
      "loss": 0.77510417,
      "num_input_tokens_seen": 74277135,
      "step": 3440,
      "time_per_iteration": 3.672687292098999
    },
    {
      "auxiliary_loss_clip": 0.01190536,
      "auxiliary_loss_mlp": 0.0102639,
      "balance_loss_clip": 1.06009674,
      "balance_loss_mlp": 1.01809311,
      "epoch": 0.413755786689112,
      "flos": 19828615933440.0,
      "grad_norm": 30.268612784113913,
      "language_loss": 0.83803892,
      "learning_rate": 2.6442003641394225e-06,
      "loss": 0.86020815,
      "num_input_tokens_seen": 74294730,
      "step": 3441,
      "time_per_iteration": 2.6538584232330322
    },
    {
      "auxiliary_loss_clip": 0.01184135,
      "auxiliary_loss_mlp": 0.0103592,
      "balance_loss_clip": 1.05599034,
      "balance_loss_mlp": 1.02699089,
      "epoch": 0.4138760295797511,
      "flos": 26870446759680.0,
      "grad_norm": 4.286838738787525,
      "language_loss": 0.84343028,
      "learning_rate": 2.643462858483356e-06,
      "loss": 0.86563081,
      "num_input_tokens_seen": 74315015,
      "step": 3442,
      "time_per_iteration": 2.7757506370544434
    },
    {
      "auxiliary_loss_clip": 0.01167202,
      "auxiliary_loss_mlp": 0.01032658,
      "balance_loss_clip": 1.05685604,
      "balance_loss_mlp": 1.02355623,
      "epoch": 0.41399627247039017,
      "flos": 16399254798720.0,
      "grad_norm": 1.9429904439204053,
      "language_loss": 0.73065364,
      "learning_rate": 2.6427252552144856e-06,
      "loss": 0.75265223,
      "num_input_tokens_seen": 74333665,
      "step": 3443,
      "time_per_iteration": 3.6457386016845703
    },
    {
      "auxiliary_loss_clip": 0.01188439,
      "auxiliary_loss_mlp": 0.01037854,
      "balance_loss_clip": 1.05938137,
      "balance_loss_mlp": 1.02868092,
      "epoch": 0.4141165153610293,
      "flos": 22930220442240.0,
      "grad_norm": 7.03396347799374,
      "language_loss": 0.75374019,
      "learning_rate": 2.6419875544447044e-06,
      "loss": 0.77600306,
      "num_input_tokens_seen": 74355065,
      "step": 3444,
      "time_per_iteration": 2.6841259002685547
    },
    {
      "auxiliary_loss_clip": 0.01171768,
      "auxiliary_loss_mlp": 0.01035804,
      "balance_loss_clip": 1.05448341,
      "balance_loss_mlp": 1.02673197,
      "epoch": 0.4142367582516684,
      "flos": 25192556697600.0,
      "grad_norm": 2.00251315862945,
      "language_loss": 0.71693474,
      "learning_rate": 2.6412497562859218e-06,
      "loss": 0.73901051,
      "num_input_tokens_seen": 74376345,
      "step": 3445,
      "time_per_iteration": 2.71512770652771
    },
    {
      "auxiliary_loss_clip": 0.01161625,
      "auxiliary_loss_mlp": 0.01036783,
      "balance_loss_clip": 1.05705285,
      "balance_loss_mlp": 1.02750897,
      "epoch": 0.41435700114230745,
      "flos": 21690476478720.0,
      "grad_norm": 2.775370225574097,
      "language_loss": 0.76464295,
      "learning_rate": 2.6405118608500617e-06,
      "loss": 0.78662705,
      "num_input_tokens_seen": 74395170,
      "step": 3446,
      "time_per_iteration": 2.786836862564087
    },
    {
      "auxiliary_loss_clip": 0.01152627,
      "auxiliary_loss_mlp": 0.01028654,
      "balance_loss_clip": 1.05602956,
      "balance_loss_mlp": 1.02026725,
      "epoch": 0.41447724403294656,
      "flos": 25995160143360.0,
      "grad_norm": 4.083282099178032,
      "language_loss": 0.81537116,
      "learning_rate": 2.6397738682490613e-06,
      "loss": 0.83718389,
      "num_input_tokens_seen": 74416070,
      "step": 3447,
      "time_per_iteration": 3.768272876739502
    },
    {
      "auxiliary_loss_clip": 0.01151706,
      "auxiliary_loss_mlp": 0.01031657,
      "balance_loss_clip": 1.05009651,
      "balance_loss_mlp": 1.0230087,
      "epoch": 0.41459748692358567,
      "flos": 18259678800000.0,
      "grad_norm": 1.6897008712216368,
      "language_loss": 0.75457209,
      "learning_rate": 2.6390357785948734e-06,
      "loss": 0.77640575,
      "num_input_tokens_seen": 74433185,
      "step": 3448,
      "time_per_iteration": 2.720869302749634
    },
    {
      "auxiliary_loss_clip": 0.01201677,
      "auxiliary_loss_mlp": 0.01029696,
      "balance_loss_clip": 1.06217539,
      "balance_loss_mlp": 1.02048707,
      "epoch": 0.4147177298142247,
      "flos": 24168456034560.0,
      "grad_norm": 1.9917023809645837,
      "language_loss": 0.80411041,
      "learning_rate": 2.6382975919994667e-06,
      "loss": 0.82642412,
      "num_input_tokens_seen": 74453760,
      "step": 3449,
      "time_per_iteration": 2.7310333251953125
    },
    {
      "auxiliary_loss_clip": 0.01182355,
      "auxiliary_loss_mlp": 0.01031344,
      "balance_loss_clip": 1.06207061,
      "balance_loss_mlp": 1.02305257,
      "epoch": 0.41483797270486383,
      "flos": 20084659056000.0,
      "grad_norm": 1.717416910857655,
      "language_loss": 0.72950542,
      "learning_rate": 2.637559308574822e-06,
      "loss": 0.75164247,
      "num_input_tokens_seen": 74473505,
      "step": 3450,
      "time_per_iteration": 2.6363606452941895
    },
    {
      "auxiliary_loss_clip": 0.01197453,
      "auxiliary_loss_mlp": 0.00762899,
      "balance_loss_clip": 1.05967283,
      "balance_loss_mlp": 1.00087667,
      "epoch": 0.4149582155955029,
      "flos": 30081040110720.0,
      "grad_norm": 2.4869831990404316,
      "language_loss": 0.71876752,
      "learning_rate": 2.6368209284329376e-06,
      "loss": 0.73837101,
      "num_input_tokens_seen": 74494135,
      "step": 3451,
      "time_per_iteration": 2.7332890033721924
    },
    {
      "auxiliary_loss_clip": 0.01181968,
      "auxiliary_loss_mlp": 0.01030377,
      "balance_loss_clip": 1.05848408,
      "balance_loss_mlp": 1.02140617,
      "epoch": 0.415078458486142,
      "flos": 16764394504320.0,
      "grad_norm": 2.013769902759248,
      "language_loss": 0.75823879,
      "learning_rate": 2.636082451685825e-06,
      "loss": 0.78036225,
      "num_input_tokens_seen": 74512335,
      "step": 3452,
      "time_per_iteration": 2.6902475357055664
    },
    {
      "auxiliary_loss_clip": 0.0118282,
      "auxiliary_loss_mlp": 0.01038972,
      "balance_loss_clip": 1.0611223,
      "balance_loss_mlp": 1.02991819,
      "epoch": 0.4151987013767811,
      "flos": 26033692458240.0,
      "grad_norm": 1.8199993708516498,
      "language_loss": 0.86676252,
      "learning_rate": 2.6353438784455094e-06,
      "loss": 0.88898039,
      "num_input_tokens_seen": 74535620,
      "step": 3453,
      "time_per_iteration": 2.735300302505493
    },
    {
      "auxiliary_loss_clip": 0.01192647,
      "auxiliary_loss_mlp": 0.0103715,
      "balance_loss_clip": 1.06103623,
      "balance_loss_mlp": 1.02792323,
      "epoch": 0.41531894426742016,
      "flos": 24608002763520.0,
      "grad_norm": 2.626885661925387,
      "language_loss": 0.71643811,
      "learning_rate": 2.6346052088240326e-06,
      "loss": 0.73873609,
      "num_input_tokens_seen": 74555140,
      "step": 3454,
      "time_per_iteration": 2.7644975185394287
    },
    {
      "auxiliary_loss_clip": 0.01173243,
      "auxiliary_loss_mlp": 0.01031123,
      "balance_loss_clip": 1.05387211,
      "balance_loss_mlp": 1.02267683,
      "epoch": 0.4154391871580593,
      "flos": 14975791747200.0,
      "grad_norm": 2.06591584088403,
      "language_loss": 0.77363223,
      "learning_rate": 2.63386644293345e-06,
      "loss": 0.79567593,
      "num_input_tokens_seen": 74571485,
      "step": 3455,
      "time_per_iteration": 2.6400179862976074
    },
    {
      "auxiliary_loss_clip": 0.01182108,
      "auxiliary_loss_mlp": 0.01028111,
      "balance_loss_clip": 1.05886245,
      "balance_loss_mlp": 1.01920569,
      "epoch": 0.4155594300486984,
      "flos": 14647173194880.0,
      "grad_norm": 2.281526362400263,
      "language_loss": 0.82800907,
      "learning_rate": 2.633127580885833e-06,
      "loss": 0.85011125,
      "num_input_tokens_seen": 74585985,
      "step": 3456,
      "time_per_iteration": 2.687882900238037
    },
    {
      "auxiliary_loss_clip": 0.0117559,
      "auxiliary_loss_mlp": 0.01036346,
      "balance_loss_clip": 1.06029749,
      "balance_loss_mlp": 1.02778697,
      "epoch": 0.41567967293933744,
      "flos": 29497276275840.0,
      "grad_norm": 3.4073633795760885,
      "language_loss": 0.65559387,
      "learning_rate": 2.632388622793265e-06,
      "loss": 0.67771316,
      "num_input_tokens_seen": 74605140,
      "step": 3457,
      "time_per_iteration": 2.7559378147125244
    },
    {
      "auxiliary_loss_clip": 0.01165074,
      "auxiliary_loss_mlp": 0.01032437,
      "balance_loss_clip": 1.05753994,
      "balance_loss_mlp": 1.02406895,
      "epoch": 0.41579991582997655,
      "flos": 19238387650560.0,
      "grad_norm": 2.317406313856308,
      "language_loss": 0.68060839,
      "learning_rate": 2.6316495687678457e-06,
      "loss": 0.70258343,
      "num_input_tokens_seen": 74623790,
      "step": 3458,
      "time_per_iteration": 2.786050319671631
    },
    {
      "auxiliary_loss_clip": 0.01170686,
      "auxiliary_loss_mlp": 0.01029708,
      "balance_loss_clip": 1.05669665,
      "balance_loss_mlp": 1.02049887,
      "epoch": 0.41592015872061566,
      "flos": 24462061804800.0,
      "grad_norm": 2.9302178282516387,
      "language_loss": 0.76852518,
      "learning_rate": 2.6309104189216887e-06,
      "loss": 0.79052913,
      "num_input_tokens_seen": 74641355,
      "step": 3459,
      "time_per_iteration": 2.7490427494049072
    },
    {
      "auxiliary_loss_clip": 0.01160469,
      "auxiliary_loss_mlp": 0.01036412,
      "balance_loss_clip": 1.05159199,
      "balance_loss_mlp": 1.02701831,
      "epoch": 0.4160404016112547,
      "flos": 20775651966720.0,
      "grad_norm": 2.6983777501401076,
      "language_loss": 0.74941456,
      "learning_rate": 2.630171173366923e-06,
      "loss": 0.77138329,
      "num_input_tokens_seen": 74657155,
      "step": 3460,
      "time_per_iteration": 2.745750665664673
    },
    {
      "auxiliary_loss_clip": 0.011848,
      "auxiliary_loss_mlp": 0.01033139,
      "balance_loss_clip": 1.0600872,
      "balance_loss_mlp": 1.02419281,
      "epoch": 0.41616064450189383,
      "flos": 13916462820480.0,
      "grad_norm": 3.5806543930615202,
      "language_loss": 0.74645138,
      "learning_rate": 2.629431832215691e-06,
      "loss": 0.76863074,
      "num_input_tokens_seen": 74671960,
      "step": 3461,
      "time_per_iteration": 2.6863059997558594
    },
    {
      "auxiliary_loss_clip": 0.01195346,
      "auxiliary_loss_mlp": 0.01032064,
      "balance_loss_clip": 1.05698156,
      "balance_loss_mlp": 1.02296853,
      "epoch": 0.41628088739253294,
      "flos": 20010826650240.0,
      "grad_norm": 1.7111995744768056,
      "language_loss": 0.87715232,
      "learning_rate": 2.628692395580151e-06,
      "loss": 0.89942646,
      "num_input_tokens_seen": 74692050,
      "step": 3462,
      "time_per_iteration": 2.7318432331085205
    },
    {
      "auxiliary_loss_clip": 0.01172324,
      "auxiliary_loss_mlp": 0.01039011,
      "balance_loss_clip": 1.057868,
      "balance_loss_mlp": 1.02903295,
      "epoch": 0.416401130283172,
      "flos": 29168801377920.0,
      "grad_norm": 1.9237596089745335,
      "language_loss": 0.79617202,
      "learning_rate": 2.6279528635724747e-06,
      "loss": 0.81828541,
      "num_input_tokens_seen": 74712205,
      "step": 3463,
      "time_per_iteration": 2.797023296356201
    },
    {
      "auxiliary_loss_clip": 0.0119961,
      "auxiliary_loss_mlp": 0.01034581,
      "balance_loss_clip": 1.05909467,
      "balance_loss_mlp": 1.02528298,
      "epoch": 0.4165213731738111,
      "flos": 16246813478400.0,
      "grad_norm": 2.565945726182602,
      "language_loss": 0.79263699,
      "learning_rate": 2.627213236304848e-06,
      "loss": 0.81497896,
      "num_input_tokens_seen": 74729005,
      "step": 3464,
      "time_per_iteration": 2.6110405921936035
    },
    {
      "auxiliary_loss_clip": 0.0119595,
      "auxiliary_loss_mlp": 0.01033413,
      "balance_loss_clip": 1.05860841,
      "balance_loss_mlp": 1.02415609,
      "epoch": 0.4166416160644502,
      "flos": 33765438787200.0,
      "grad_norm": 2.143120349985999,
      "language_loss": 0.70796055,
      "learning_rate": 2.626473513889472e-06,
      "loss": 0.73025417,
      "num_input_tokens_seen": 74751385,
      "step": 3465,
      "time_per_iteration": 2.755011796951294
    },
    {
      "auxiliary_loss_clip": 0.01145587,
      "auxiliary_loss_mlp": 0.01030887,
      "balance_loss_clip": 1.05140781,
      "balance_loss_mlp": 1.02261353,
      "epoch": 0.41676185895508927,
      "flos": 20917498775040.0,
      "grad_norm": 1.9857299217705047,
      "language_loss": 0.8279711,
      "learning_rate": 2.625733696438562e-06,
      "loss": 0.84973586,
      "num_input_tokens_seen": 74768890,
      "step": 3466,
      "time_per_iteration": 4.554953336715698
    },
    {
      "auxiliary_loss_clip": 0.01188081,
      "auxiliary_loss_mlp": 0.01038376,
      "balance_loss_clip": 1.06066847,
      "balance_loss_mlp": 1.02907777,
      "epoch": 0.4168821018457284,
      "flos": 18406122549120.0,
      "grad_norm": 1.931783286564902,
      "language_loss": 0.75711334,
      "learning_rate": 2.6249937840643476e-06,
      "loss": 0.77937782,
      "num_input_tokens_seen": 74787195,
      "step": 3467,
      "time_per_iteration": 2.686951160430908
    },
    {
      "auxiliary_loss_clip": 0.01201458,
      "auxiliary_loss_mlp": 0.01036605,
      "balance_loss_clip": 1.06173658,
      "balance_loss_mlp": 1.02727056,
      "epoch": 0.41700234473636744,
      "flos": 18698399516160.0,
      "grad_norm": 1.6262070803670066,
      "language_loss": 0.66935611,
      "learning_rate": 2.6242537768790733e-06,
      "loss": 0.69173676,
      "num_input_tokens_seen": 74806350,
      "step": 3468,
      "time_per_iteration": 2.6532020568847656
    },
    {
      "auxiliary_loss_clip": 0.01185559,
      "auxiliary_loss_mlp": 0.01034941,
      "balance_loss_clip": 1.05829263,
      "balance_loss_mlp": 1.0254755,
      "epoch": 0.41712258762700655,
      "flos": 31033283616000.0,
      "grad_norm": 2.267052288539736,
      "language_loss": 0.69041479,
      "learning_rate": 2.6235136749949975e-06,
      "loss": 0.71261978,
      "num_input_tokens_seen": 74829800,
      "step": 3469,
      "time_per_iteration": 3.639336347579956
    },
    {
      "auxiliary_loss_clip": 0.01159067,
      "auxiliary_loss_mlp": 0.01032165,
      "balance_loss_clip": 1.05556631,
      "balance_loss_mlp": 1.02287853,
      "epoch": 0.41724283051764566,
      "flos": 35914763877120.0,
      "grad_norm": 12.974925562216939,
      "language_loss": 0.6155436,
      "learning_rate": 2.6227734785243924e-06,
      "loss": 0.63745594,
      "num_input_tokens_seen": 74849760,
      "step": 3470,
      "time_per_iteration": 2.8809127807617188
    },
    {
      "auxiliary_loss_clip": 0.01197675,
      "auxiliary_loss_mlp": 0.01028642,
      "balance_loss_clip": 1.05851233,
      "balance_loss_mlp": 1.0199455,
      "epoch": 0.4173630734082847,
      "flos": 25333649320320.0,
      "grad_norm": 2.8941285486382453,
      "language_loss": 0.7948513,
      "learning_rate": 2.6220331875795466e-06,
      "loss": 0.81711447,
      "num_input_tokens_seen": 74869110,
      "step": 3471,
      "time_per_iteration": 2.638638973236084
    },
    {
      "auxiliary_loss_clip": 0.01203593,
      "auxiliary_loss_mlp": 0.01032805,
      "balance_loss_clip": 1.06262529,
      "balance_loss_mlp": 1.02311277,
      "epoch": 0.4174833162989238,
      "flos": 26685398868480.0,
      "grad_norm": 12.53893741512596,
      "language_loss": 0.75205159,
      "learning_rate": 2.62129280227276e-06,
      "loss": 0.77441561,
      "num_input_tokens_seen": 74889110,
      "step": 3472,
      "time_per_iteration": 2.724863290786743
    },
    {
      "auxiliary_loss_clip": 0.01201742,
      "auxiliary_loss_mlp": 0.01038751,
      "balance_loss_clip": 1.06152296,
      "balance_loss_mlp": 1.02974463,
      "epoch": 0.41760355918956293,
      "flos": 74739584010240.0,
      "grad_norm": 3.027167712378814,
      "language_loss": 0.68567675,
      "learning_rate": 2.62055232271635e-06,
      "loss": 0.70808172,
      "num_input_tokens_seen": 74916260,
      "step": 3473,
      "time_per_iteration": 4.034708499908447
    },
    {
      "auxiliary_loss_clip": 0.01176404,
      "auxiliary_loss_mlp": 0.01039628,
      "balance_loss_clip": 1.05613589,
      "balance_loss_mlp": 1.02951324,
      "epoch": 0.417723802080202,
      "flos": 14317513148160.0,
      "grad_norm": 2.6401374957051984,
      "language_loss": 0.87881064,
      "learning_rate": 2.619811749022646e-06,
      "loss": 0.90097106,
      "num_input_tokens_seen": 74931570,
      "step": 3474,
      "time_per_iteration": 2.602790117263794
    },
    {
      "auxiliary_loss_clip": 0.01187656,
      "auxiliary_loss_mlp": 0.01038621,
      "balance_loss_clip": 1.05697823,
      "balance_loss_mlp": 1.02845836,
      "epoch": 0.4178440449708411,
      "flos": 14643797316480.0,
      "grad_norm": 2.9273248014200837,
      "language_loss": 0.71760374,
      "learning_rate": 2.6190710813039917e-06,
      "loss": 0.7398665,
      "num_input_tokens_seen": 74944695,
      "step": 3475,
      "time_per_iteration": 2.660700798034668
    },
    {
      "auxiliary_loss_clip": 0.01172799,
      "auxiliary_loss_mlp": 0.01037431,
      "balance_loss_clip": 1.05614233,
      "balance_loss_mlp": 1.02762628,
      "epoch": 0.4179642878614802,
      "flos": 21507296094720.0,
      "grad_norm": 2.8723581412340433,
      "language_loss": 0.83838463,
      "learning_rate": 2.618330319672747e-06,
      "loss": 0.86048687,
      "num_input_tokens_seen": 74964115,
      "step": 3476,
      "time_per_iteration": 2.708752155303955
    },
    {
      "auxiliary_loss_clip": 0.01188507,
      "auxiliary_loss_mlp": 0.01030529,
      "balance_loss_clip": 1.05865967,
      "balance_loss_mlp": 1.02115917,
      "epoch": 0.41808453075211927,
      "flos": 18441997257600.0,
      "grad_norm": 2.701138956300698,
      "language_loss": 0.92142606,
      "learning_rate": 2.617589464241284e-06,
      "loss": 0.94361639,
      "num_input_tokens_seen": 74978515,
      "step": 3477,
      "time_per_iteration": 2.7300984859466553
    },
    {
      "auxiliary_loss_clip": 0.01186583,
      "auxiliary_loss_mlp": 0.01040902,
      "balance_loss_clip": 1.05674112,
      "balance_loss_mlp": 1.03177071,
      "epoch": 0.4182047736427584,
      "flos": 20301020628480.0,
      "grad_norm": 3.9306115230522782,
      "language_loss": 0.74589217,
      "learning_rate": 2.6168485151219914e-06,
      "loss": 0.76816696,
      "num_input_tokens_seen": 74998135,
      "step": 3478,
      "time_per_iteration": 2.697974681854248
    },
    {
      "auxiliary_loss_clip": 0.01201781,
      "auxiliary_loss_mlp": 0.01033261,
      "balance_loss_clip": 1.05947423,
      "balance_loss_mlp": 1.02336097,
      "epoch": 0.4183250165333975,
      "flos": 18876623823360.0,
      "grad_norm": 6.708681776635611,
      "language_loss": 0.71764469,
      "learning_rate": 2.616107472427269e-06,
      "loss": 0.73999512,
      "num_input_tokens_seen": 75012830,
      "step": 3479,
      "time_per_iteration": 2.6415131092071533
    },
    {
      "auxiliary_loss_clip": 0.01199692,
      "auxiliary_loss_mlp": 0.01029185,
      "balance_loss_clip": 1.05848074,
      "balance_loss_mlp": 1.01992834,
      "epoch": 0.41844525942403654,
      "flos": 17740050698880.0,
      "grad_norm": 2.4983895856643668,
      "language_loss": 0.76841509,
      "learning_rate": 2.615366336269533e-06,
      "loss": 0.79070389,
      "num_input_tokens_seen": 75026495,
      "step": 3480,
      "time_per_iteration": 2.638580799102783
    },
    {
      "auxiliary_loss_clip": 0.0117967,
      "auxiliary_loss_mlp": 0.01034396,
      "balance_loss_clip": 1.05759478,
      "balance_loss_mlp": 1.02454972,
      "epoch": 0.41856550231467565,
      "flos": 18361377181440.0,
      "grad_norm": 2.4759988595615807,
      "language_loss": 0.80556262,
      "learning_rate": 2.6146251067612126e-06,
      "loss": 0.8277033,
      "num_input_tokens_seen": 75041970,
      "step": 3481,
      "time_per_iteration": 2.6246747970581055
    },
    {
      "auxiliary_loss_clip": 0.0119075,
      "auxiliary_loss_mlp": 0.01032456,
      "balance_loss_clip": 1.05844462,
      "balance_loss_mlp": 1.02263331,
      "epoch": 0.41868574520531476,
      "flos": 22781801445120.0,
      "grad_norm": 1.7162453030877298,
      "language_loss": 0.82914054,
      "learning_rate": 2.6138837840147525e-06,
      "loss": 0.8513726,
      "num_input_tokens_seen": 75061005,
      "step": 3482,
      "time_per_iteration": 2.691859722137451
    },
    {
      "auxiliary_loss_clip": 0.01174022,
      "auxiliary_loss_mlp": 0.01033059,
      "balance_loss_clip": 1.05355835,
      "balance_loss_mlp": 1.02305174,
      "epoch": 0.4188059880959538,
      "flos": 13699167494400.0,
      "grad_norm": 2.5417335315586675,
      "language_loss": 0.77090591,
      "learning_rate": 2.6131423681426103e-06,
      "loss": 0.79297668,
      "num_input_tokens_seen": 75076920,
      "step": 3483,
      "time_per_iteration": 2.6403567790985107
    },
    {
      "auxiliary_loss_clip": 0.0119586,
      "auxiliary_loss_mlp": 0.01034148,
      "balance_loss_clip": 1.05781794,
      "balance_loss_mlp": 1.02506995,
      "epoch": 0.41892623098659293,
      "flos": 37818281220480.0,
      "grad_norm": 1.9014840205960724,
      "language_loss": 0.73091722,
      "learning_rate": 2.6124008592572587e-06,
      "loss": 0.75321722,
      "num_input_tokens_seen": 75100905,
      "step": 3484,
      "time_per_iteration": 2.8591439723968506
    },
    {
      "auxiliary_loss_clip": 0.01200142,
      "auxiliary_loss_mlp": 0.01035766,
      "balance_loss_clip": 1.05760002,
      "balance_loss_mlp": 1.02615213,
      "epoch": 0.419046473877232,
      "flos": 23258874908160.0,
      "grad_norm": 15.939313933288314,
      "language_loss": 0.81894481,
      "learning_rate": 2.6116592574711835e-06,
      "loss": 0.84130388,
      "num_input_tokens_seen": 75119205,
      "step": 3485,
      "time_per_iteration": 2.6351749897003174
    },
    {
      "auxiliary_loss_clip": 0.01202177,
      "auxiliary_loss_mlp": 0.01030287,
      "balance_loss_clip": 1.0588429,
      "balance_loss_mlp": 1.02040458,
      "epoch": 0.4191667167678711,
      "flos": 20741034234240.0,
      "grad_norm": 2.0660903929486376,
      "language_loss": 0.84057182,
      "learning_rate": 2.6109175628968853e-06,
      "loss": 0.8628965,
      "num_input_tokens_seen": 75138970,
      "step": 3486,
      "time_per_iteration": 2.642157793045044
    },
    {
      "auxiliary_loss_clip": 0.01196069,
      "auxiliary_loss_mlp": 0.01032879,
      "balance_loss_clip": 1.05818057,
      "balance_loss_mlp": 1.02392042,
      "epoch": 0.4192869596585102,
      "flos": 23586416052480.0,
      "grad_norm": 1.9286727204786622,
      "language_loss": 0.83234262,
      "learning_rate": 2.610175775646878e-06,
      "loss": 0.85463214,
      "num_input_tokens_seen": 75157550,
      "step": 3487,
      "time_per_iteration": 2.6382412910461426
    },
    {
      "auxiliary_loss_clip": 0.01169973,
      "auxiliary_loss_mlp": 0.01037189,
      "balance_loss_clip": 1.05659938,
      "balance_loss_mlp": 1.02735996,
      "epoch": 0.41940720254914926,
      "flos": 25081269384960.0,
      "grad_norm": 2.118667726295638,
      "language_loss": 0.73020595,
      "learning_rate": 2.6094338958336907e-06,
      "loss": 0.75227755,
      "num_input_tokens_seen": 75176220,
      "step": 3488,
      "time_per_iteration": 2.752854108810425
    },
    {
      "auxiliary_loss_clip": 0.01196653,
      "auxiliary_loss_mlp": 0.01026784,
      "balance_loss_clip": 1.05932379,
      "balance_loss_mlp": 1.01826692,
      "epoch": 0.41952744543978837,
      "flos": 15554132628480.0,
      "grad_norm": 2.958152511243012,
      "language_loss": 0.82472682,
      "learning_rate": 2.608691923569867e-06,
      "loss": 0.84696126,
      "num_input_tokens_seen": 75193095,
      "step": 3489,
      "time_per_iteration": 2.532405376434326
    },
    {
      "auxiliary_loss_clip": 0.0120161,
      "auxiliary_loss_mlp": 0.01034007,
      "balance_loss_clip": 1.06266284,
      "balance_loss_mlp": 1.02457726,
      "epoch": 0.4196476883304275,
      "flos": 24644775312000.0,
      "grad_norm": 1.6460003922427466,
      "language_loss": 0.76141512,
      "learning_rate": 2.6079498589679616e-06,
      "loss": 0.78377128,
      "num_input_tokens_seen": 75214185,
      "step": 3490,
      "time_per_iteration": 2.6751694679260254
    },
    {
      "auxiliary_loss_clip": 0.01168848,
      "auxiliary_loss_mlp": 0.0103798,
      "balance_loss_clip": 1.05356908,
      "balance_loss_mlp": 1.02753139,
      "epoch": 0.41976793122106654,
      "flos": 24531333183360.0,
      "grad_norm": 2.531686644916055,
      "language_loss": 0.76280063,
      "learning_rate": 2.6072077021405465e-06,
      "loss": 0.78486896,
      "num_input_tokens_seen": 75233020,
      "step": 3491,
      "time_per_iteration": 2.676279067993164
    },
    {
      "auxiliary_loss_clip": 0.01174828,
      "auxiliary_loss_mlp": 0.01033226,
      "balance_loss_clip": 1.05529344,
      "balance_loss_mlp": 1.02405918,
      "epoch": 0.41988817411170565,
      "flos": 21175301664000.0,
      "grad_norm": 1.9079893785073379,
      "language_loss": 0.69601095,
      "learning_rate": 2.6064654532002054e-06,
      "loss": 0.71809149,
      "num_input_tokens_seen": 75252030,
      "step": 3492,
      "time_per_iteration": 4.573726654052734
    },
    {
      "auxiliary_loss_clip": 0.01182855,
      "auxiliary_loss_mlp": 0.0103417,
      "balance_loss_clip": 1.05773866,
      "balance_loss_mlp": 1.02516389,
      "epoch": 0.42000841700234476,
      "flos": 31649402626560.0,
      "grad_norm": 2.2523786216973183,
      "language_loss": 0.76121521,
      "learning_rate": 2.6057231122595375e-06,
      "loss": 0.78338552,
      "num_input_tokens_seen": 75273340,
      "step": 3493,
      "time_per_iteration": 2.7799463272094727
    },
    {
      "auxiliary_loss_clip": 0.01198422,
      "auxiliary_loss_mlp": 0.01036537,
      "balance_loss_clip": 1.05952632,
      "balance_loss_mlp": 1.02783477,
      "epoch": 0.4201286598929838,
      "flos": 21281525159040.0,
      "grad_norm": 2.200845008269174,
      "language_loss": 0.73040944,
      "learning_rate": 2.604980679431154e-06,
      "loss": 0.75275898,
      "num_input_tokens_seen": 75291580,
      "step": 3494,
      "time_per_iteration": 2.607879877090454
    },
    {
      "auxiliary_loss_clip": 0.01187274,
      "auxiliary_loss_mlp": 0.01032589,
      "balance_loss_clip": 1.05635965,
      "balance_loss_mlp": 1.02305245,
      "epoch": 0.4202489027836229,
      "flos": 18546532813440.0,
      "grad_norm": 2.3255282057642788,
      "language_loss": 0.74307144,
      "learning_rate": 2.604238154827684e-06,
      "loss": 0.76526999,
      "num_input_tokens_seen": 75308205,
      "step": 3495,
      "time_per_iteration": 3.6661651134490967
    },
    {
      "auxiliary_loss_clip": 0.01160263,
      "auxiliary_loss_mlp": 0.01026814,
      "balance_loss_clip": 1.05608737,
      "balance_loss_mlp": 1.01775408,
      "epoch": 0.42036914567426203,
      "flos": 19317643009920.0,
      "grad_norm": 2.2547820792554636,
      "language_loss": 0.73016214,
      "learning_rate": 2.6034955385617656e-06,
      "loss": 0.75203288,
      "num_input_tokens_seen": 75326535,
      "step": 3496,
      "time_per_iteration": 2.7457711696624756
    },
    {
      "auxiliary_loss_clip": 0.01079844,
      "auxiliary_loss_mlp": 0.0100174,
      "balance_loss_clip": 1.01761174,
      "balance_loss_mlp": 1.00035751,
      "epoch": 0.4204893885649011,
      "flos": 67842942935040.0,
      "grad_norm": 0.7161048436568048,
      "language_loss": 0.61554861,
      "learning_rate": 2.6027528307460544e-06,
      "loss": 0.6363644,
      "num_input_tokens_seen": 75390540,
      "step": 3497,
      "time_per_iteration": 3.305985689163208
    },
    {
      "auxiliary_loss_clip": 0.01153785,
      "auxiliary_loss_mlp": 0.0102777,
      "balance_loss_clip": 1.05594218,
      "balance_loss_mlp": 1.01916909,
      "epoch": 0.4206096314555402,
      "flos": 21908777385600.0,
      "grad_norm": 1.8289721811093411,
      "language_loss": 0.86912668,
      "learning_rate": 2.602010031493217e-06,
      "loss": 0.89094222,
      "num_input_tokens_seen": 75408770,
      "step": 3498,
      "time_per_iteration": 2.8250818252563477
    },
    {
      "auxiliary_loss_clip": 0.01159985,
      "auxiliary_loss_mlp": 0.01030155,
      "balance_loss_clip": 1.05403304,
      "balance_loss_mlp": 1.02145243,
      "epoch": 0.42072987434617926,
      "flos": 29278185269760.0,
      "grad_norm": 2.6632430814204877,
      "language_loss": 0.87325931,
      "learning_rate": 2.6012671409159367e-06,
      "loss": 0.89516073,
      "num_input_tokens_seen": 75430105,
      "step": 3499,
      "time_per_iteration": 3.742661476135254
    },
    {
      "auxiliary_loss_clip": 0.0118193,
      "auxiliary_loss_mlp": 0.01032235,
      "balance_loss_clip": 1.05731618,
      "balance_loss_mlp": 1.02213836,
      "epoch": 0.42085011723681837,
      "flos": 27600726170880.0,
      "grad_norm": 1.8317436137677827,
      "language_loss": 0.81790137,
      "learning_rate": 2.6005241591269097e-06,
      "loss": 0.84004307,
      "num_input_tokens_seen": 75449475,
      "step": 3500,
      "time_per_iteration": 2.6642231941223145
    },
    {
      "auxiliary_loss_clip": 0.01195896,
      "auxiliary_loss_mlp": 0.01038907,
      "balance_loss_clip": 1.05960464,
      "balance_loss_mlp": 1.02994263,
      "epoch": 0.4209703601274575,
      "flos": 27818632028160.0,
      "grad_norm": 1.969572300423845,
      "language_loss": 0.80171466,
      "learning_rate": 2.5997810862388454e-06,
      "loss": 0.82406271,
      "num_input_tokens_seen": 75469315,
      "step": 3501,
      "time_per_iteration": 2.7102012634277344
    },
    {
      "auxiliary_loss_clip": 0.01180904,
      "auxiliary_loss_mlp": 0.01033635,
      "balance_loss_clip": 1.057374,
      "balance_loss_mlp": 1.02371085,
      "epoch": 0.42109060301809653,
      "flos": 27525529048320.0,
      "grad_norm": 3.958166286125514,
      "language_loss": 0.76213658,
      "learning_rate": 2.599037922364467e-06,
      "loss": 0.78428197,
      "num_input_tokens_seen": 75488215,
      "step": 3502,
      "time_per_iteration": 2.6797704696655273
    },
    {
      "auxiliary_loss_clip": 0.01167597,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.05357838,
      "balance_loss_mlp": 1.02240694,
      "epoch": 0.42121084590873564,
      "flos": 29314275459840.0,
      "grad_norm": 2.2497865998511393,
      "language_loss": 0.75946474,
      "learning_rate": 2.5982946676165112e-06,
      "loss": 0.78145933,
      "num_input_tokens_seen": 75507985,
      "step": 3503,
      "time_per_iteration": 2.805089235305786
    },
    {
      "auxiliary_loss_clip": 0.01088286,
      "auxiliary_loss_mlp": 0.01002016,
      "balance_loss_clip": 1.01736689,
      "balance_loss_mlp": 1.00072849,
      "epoch": 0.42133108879937475,
      "flos": 67398835178880.0,
      "grad_norm": 0.725057601799523,
      "language_loss": 0.57502711,
      "learning_rate": 2.5975513221077313e-06,
      "loss": 0.59593016,
      "num_input_tokens_seen": 75571955,
      "step": 3504,
      "time_per_iteration": 3.2931528091430664
    },
    {
      "auxiliary_loss_clip": 0.01196129,
      "auxiliary_loss_mlp": 0.01035901,
      "balance_loss_clip": 1.05836892,
      "balance_loss_mlp": 1.02714467,
      "epoch": 0.4214513316900138,
      "flos": 23106038538240.0,
      "grad_norm": 4.234592234025074,
      "language_loss": 0.89338976,
      "learning_rate": 2.5968078859508897e-06,
      "loss": 0.91571009,
      "num_input_tokens_seen": 75589155,
      "step": 3505,
      "time_per_iteration": 2.714346408843994
    },
    {
      "auxiliary_loss_clip": 0.01184947,
      "auxiliary_loss_mlp": 0.01025542,
      "balance_loss_clip": 1.05756688,
      "balance_loss_mlp": 1.01692057,
      "epoch": 0.4215715745806529,
      "flos": 15336190857600.0,
      "grad_norm": 2.33094523404308,
      "language_loss": 0.79952621,
      "learning_rate": 2.5960643592587673e-06,
      "loss": 0.82163113,
      "num_input_tokens_seen": 75606565,
      "step": 3506,
      "time_per_iteration": 2.7553110122680664
    },
    {
      "auxiliary_loss_clip": 0.01181723,
      "auxiliary_loss_mlp": 0.01035975,
      "balance_loss_clip": 1.05844903,
      "balance_loss_mlp": 1.0273385,
      "epoch": 0.42169181747129203,
      "flos": 22127257860480.0,
      "grad_norm": 2.140226558860202,
      "language_loss": 0.81584251,
      "learning_rate": 2.5953207421441553e-06,
      "loss": 0.83801949,
      "num_input_tokens_seen": 75625165,
      "step": 3507,
      "time_per_iteration": 2.6381547451019287
    },
    {
      "auxiliary_loss_clip": 0.01179058,
      "auxiliary_loss_mlp": 0.00762576,
      "balance_loss_clip": 1.05830181,
      "balance_loss_mlp": 1.00080812,
      "epoch": 0.4218120603619311,
      "flos": 22630724841600.0,
      "grad_norm": 3.312493232956557,
      "language_loss": 0.7570858,
      "learning_rate": 2.5945770347198603e-06,
      "loss": 0.77650213,
      "num_input_tokens_seen": 75643320,
      "step": 3508,
      "time_per_iteration": 2.710047721862793
    },
    {
      "auxiliary_loss_clip": 0.01178175,
      "auxiliary_loss_mlp": 0.01030447,
      "balance_loss_clip": 1.05725408,
      "balance_loss_mlp": 1.02150607,
      "epoch": 0.4219323032525702,
      "flos": 19682818629120.0,
      "grad_norm": 1.887330623350238,
      "language_loss": 0.8224197,
      "learning_rate": 2.593833237098701e-06,
      "loss": 0.84450591,
      "num_input_tokens_seen": 75660920,
      "step": 3509,
      "time_per_iteration": 2.690528154373169
    },
    {
      "auxiliary_loss_clip": 0.01198093,
      "auxiliary_loss_mlp": 0.01032204,
      "balance_loss_clip": 1.05726075,
      "balance_loss_mlp": 1.02284586,
      "epoch": 0.4220525461432093,
      "flos": 30190747224960.0,
      "grad_norm": 10.588705107999548,
      "language_loss": 0.62933964,
      "learning_rate": 2.593089349393512e-06,
      "loss": 0.65164256,
      "num_input_tokens_seen": 75681410,
      "step": 3510,
      "time_per_iteration": 2.718693256378174
    },
    {
      "auxiliary_loss_clip": 0.01185231,
      "auxiliary_loss_mlp": 0.01034712,
      "balance_loss_clip": 1.05627561,
      "balance_loss_mlp": 1.02613544,
      "epoch": 0.42217278903384836,
      "flos": 24315941278080.0,
      "grad_norm": 2.2808578052511295,
      "language_loss": 0.84342051,
      "learning_rate": 2.592345371717141e-06,
      "loss": 0.86562002,
      "num_input_tokens_seen": 75700940,
      "step": 3511,
      "time_per_iteration": 2.66144061088562
    },
    {
      "auxiliary_loss_clip": 0.01172511,
      "auxiliary_loss_mlp": 0.01037065,
      "balance_loss_clip": 1.05622256,
      "balance_loss_mlp": 1.02760553,
      "epoch": 0.42229303192448747,
      "flos": 17092474352640.0,
      "grad_norm": 2.202624755379754,
      "language_loss": 0.7216686,
      "learning_rate": 2.591601304182448e-06,
      "loss": 0.74376428,
      "num_input_tokens_seen": 75718910,
      "step": 3512,
      "time_per_iteration": 2.7339634895324707
    },
    {
      "auxiliary_loss_clip": 0.01197703,
      "auxiliary_loss_mlp": 0.010285,
      "balance_loss_clip": 1.06129718,
      "balance_loss_mlp": 1.02020311,
      "epoch": 0.4224132748151266,
      "flos": 22784530878720.0,
      "grad_norm": 1.8557141046293621,
      "language_loss": 0.79424,
      "learning_rate": 2.5908571469023067e-06,
      "loss": 0.81650198,
      "num_input_tokens_seen": 75738395,
      "step": 3513,
      "time_per_iteration": 2.6149353981018066
    },
    {
      "auxiliary_loss_clip": 0.01194598,
      "auxiliary_loss_mlp": 0.01035876,
      "balance_loss_clip": 1.05773735,
      "balance_loss_mlp": 1.02689981,
      "epoch": 0.42253351770576564,
      "flos": 17819090576640.0,
      "grad_norm": 2.4713750830845,
      "language_loss": 0.75696766,
      "learning_rate": 2.5901128999896067e-06,
      "loss": 0.77927244,
      "num_input_tokens_seen": 75753825,
      "step": 3514,
      "time_per_iteration": 2.6830527782440186
    },
    {
      "auxiliary_loss_clip": 0.01172515,
      "auxiliary_loss_mlp": 0.0102704,
      "balance_loss_clip": 1.05627787,
      "balance_loss_mlp": 1.01855218,
      "epoch": 0.42265376059640475,
      "flos": 28512390286080.0,
      "grad_norm": 1.6399856673172033,
      "language_loss": 0.68372715,
      "learning_rate": 2.5893685635572487e-06,
      "loss": 0.70572269,
      "num_input_tokens_seen": 75774675,
      "step": 3515,
      "time_per_iteration": 2.722865104675293
    },
    {
      "auxiliary_loss_clip": 0.01176436,
      "auxiliary_loss_mlp": 0.01032439,
      "balance_loss_clip": 1.06061578,
      "balance_loss_mlp": 1.0230875,
      "epoch": 0.4227740034870438,
      "flos": 16253349753600.0,
      "grad_norm": 2.37565633963694,
      "language_loss": 0.69938797,
      "learning_rate": 2.5886241377181483e-06,
      "loss": 0.72147667,
      "num_input_tokens_seen": 75793545,
      "step": 3516,
      "time_per_iteration": 2.687217950820923
    },
    {
      "auxiliary_loss_clip": 0.01202247,
      "auxiliary_loss_mlp": 0.01039529,
      "balance_loss_clip": 1.06190133,
      "balance_loss_mlp": 1.02956355,
      "epoch": 0.4228942463776829,
      "flos": 25295691623040.0,
      "grad_norm": 2.8616358319225106,
      "language_loss": 0.81362593,
      "learning_rate": 2.587879622585234e-06,
      "loss": 0.83604372,
      "num_input_tokens_seen": 75812145,
      "step": 3517,
      "time_per_iteration": 2.765030860900879
    },
    {
      "auxiliary_loss_clip": 0.01199311,
      "auxiliary_loss_mlp": 0.01036469,
      "balance_loss_clip": 1.06189108,
      "balance_loss_mlp": 1.02755189,
      "epoch": 0.423014489268322,
      "flos": 26395779507840.0,
      "grad_norm": 2.5806544437785144,
      "language_loss": 0.76254767,
      "learning_rate": 2.5871350182714486e-06,
      "loss": 0.78490543,
      "num_input_tokens_seen": 75833025,
      "step": 3518,
      "time_per_iteration": 2.6706533432006836
    },
    {
      "auxiliary_loss_clip": 0.0115437,
      "auxiliary_loss_mlp": 0.01034037,
      "balance_loss_clip": 1.05351138,
      "balance_loss_mlp": 1.02542448,
      "epoch": 0.4231347321589611,
      "flos": 17274002711040.0,
      "grad_norm": 4.484655155493598,
      "language_loss": 0.80737174,
      "learning_rate": 2.586390324889748e-06,
      "loss": 0.82925582,
      "num_input_tokens_seen": 75848925,
      "step": 3519,
      "time_per_iteration": 4.706530332565308
    },
    {
      "auxiliary_loss_clip": 0.01181201,
      "auxiliary_loss_mlp": 0.01028669,
      "balance_loss_clip": 1.05812228,
      "balance_loss_mlp": 1.02000225,
      "epoch": 0.4232549750496002,
      "flos": 22999635475200.0,
      "grad_norm": 1.8488400612059825,
      "language_loss": 0.67630506,
      "learning_rate": 2.5856455425531003e-06,
      "loss": 0.69840372,
      "num_input_tokens_seen": 75870400,
      "step": 3520,
      "time_per_iteration": 2.6900556087493896
    },
    {
      "auxiliary_loss_clip": 0.01185345,
      "auxiliary_loss_mlp": 0.01031404,
      "balance_loss_clip": 1.06196666,
      "balance_loss_mlp": 1.02246296,
      "epoch": 0.4233752179402393,
      "flos": 21248343970560.0,
      "grad_norm": 3.965283428084203,
      "language_loss": 0.80964947,
      "learning_rate": 2.5849006713744902e-06,
      "loss": 0.83181703,
      "num_input_tokens_seen": 75889195,
      "step": 3521,
      "time_per_iteration": 2.667351484298706
    },
    {
      "auxiliary_loss_clip": 0.01172578,
      "auxiliary_loss_mlp": 0.01027376,
      "balance_loss_clip": 1.05982542,
      "balance_loss_mlp": 1.0182029,
      "epoch": 0.42349546083087836,
      "flos": 20704297599360.0,
      "grad_norm": 2.423537563618174,
      "language_loss": 0.73207945,
      "learning_rate": 2.5841557114669135e-06,
      "loss": 0.75407898,
      "num_input_tokens_seen": 75906055,
      "step": 3522,
      "time_per_iteration": 3.7031078338623047
    },
    {
      "auxiliary_loss_clip": 0.01191451,
      "auxiliary_loss_mlp": 0.01035225,
      "balance_loss_clip": 1.05950809,
      "balance_loss_mlp": 1.02558053,
      "epoch": 0.42361570372151747,
      "flos": 18585065128320.0,
      "grad_norm": 3.5807839860757364,
      "language_loss": 0.67517638,
      "learning_rate": 2.58341066294338e-06,
      "loss": 0.69744313,
      "num_input_tokens_seen": 75922720,
      "step": 3523,
      "time_per_iteration": 2.625399589538574
    },
    {
      "auxiliary_loss_clip": 0.01190648,
      "auxiliary_loss_mlp": 0.00763339,
      "balance_loss_clip": 1.06226134,
      "balance_loss_mlp": 1.0008707,
      "epoch": 0.4237359466121566,
      "flos": 20959478795520.0,
      "grad_norm": 2.202703109704965,
      "language_loss": 0.85352576,
      "learning_rate": 2.5826655259169124e-06,
      "loss": 0.87306565,
      "num_input_tokens_seen": 75941375,
      "step": 3524,
      "time_per_iteration": 2.745460033416748
    },
    {
      "auxiliary_loss_clip": 0.01200668,
      "auxiliary_loss_mlp": 0.01034471,
      "balance_loss_clip": 1.06262755,
      "balance_loss_mlp": 1.0255003,
      "epoch": 0.42385618950279563,
      "flos": 18038181582720.0,
      "grad_norm": 1.8096817134665961,
      "language_loss": 0.90758091,
      "learning_rate": 2.5819203005005475e-06,
      "loss": 0.9299323,
      "num_input_tokens_seen": 75958710,
      "step": 3525,
      "time_per_iteration": 3.5973403453826904
    },
    {
      "auxiliary_loss_clip": 0.01188788,
      "auxiliary_loss_mlp": 0.01033831,
      "balance_loss_clip": 1.05930185,
      "balance_loss_mlp": 1.0248425,
      "epoch": 0.42397643239343474,
      "flos": 23769129559680.0,
      "grad_norm": 2.148288358169162,
      "language_loss": 0.78996217,
      "learning_rate": 2.581174986807336e-06,
      "loss": 0.81218839,
      "num_input_tokens_seen": 75978945,
      "step": 3526,
      "time_per_iteration": 2.6578218936920166
    },
    {
      "auxiliary_loss_clip": 0.01197698,
      "auxiliary_loss_mlp": 0.01030746,
      "balance_loss_clip": 1.05989194,
      "balance_loss_mlp": 1.02172208,
      "epoch": 0.42409667528407385,
      "flos": 16545088016640.0,
      "grad_norm": 2.8238394008310976,
      "language_loss": 0.91419375,
      "learning_rate": 2.580429584950341e-06,
      "loss": 0.93647826,
      "num_input_tokens_seen": 75994695,
      "step": 3527,
      "time_per_iteration": 2.671123743057251
    },
    {
      "auxiliary_loss_clip": 0.01199814,
      "auxiliary_loss_mlp": 0.01035398,
      "balance_loss_clip": 1.05951858,
      "balance_loss_mlp": 1.02598643,
      "epoch": 0.4242169181747129,
      "flos": 16034186920320.0,
      "grad_norm": 2.388958243349373,
      "language_loss": 0.66965789,
      "learning_rate": 2.5796840950426397e-06,
      "loss": 0.69201005,
      "num_input_tokens_seen": 76011780,
      "step": 3528,
      "time_per_iteration": 2.5981860160827637
    },
    {
      "auxiliary_loss_clip": 0.01197676,
      "auxiliary_loss_mlp": 0.01032656,
      "balance_loss_clip": 1.06092381,
      "balance_loss_mlp": 1.02409053,
      "epoch": 0.424337161065352,
      "flos": 20084012611200.0,
      "grad_norm": 4.197624439957267,
      "language_loss": 0.66234541,
      "learning_rate": 2.578938517197322e-06,
      "loss": 0.68464863,
      "num_input_tokens_seen": 76029875,
      "step": 3529,
      "time_per_iteration": 2.6655924320220947
    },
    {
      "auxiliary_loss_clip": 0.0118076,
      "auxiliary_loss_mlp": 0.01030225,
      "balance_loss_clip": 1.05989289,
      "balance_loss_mlp": 1.02164829,
      "epoch": 0.4244574039559911,
      "flos": 23878369797120.0,
      "grad_norm": 3.1664893894215713,
      "language_loss": 0.63158834,
      "learning_rate": 2.5781928515274916e-06,
      "loss": 0.65369821,
      "num_input_tokens_seen": 76048595,
      "step": 3530,
      "time_per_iteration": 2.6689293384552
    },
    {
      "auxiliary_loss_clip": 0.01171722,
      "auxiliary_loss_mlp": 0.01044343,
      "balance_loss_clip": 1.05757999,
      "balance_loss_mlp": 1.03562331,
      "epoch": 0.4245776468466302,
      "flos": 17565920542080.0,
      "grad_norm": 2.2404680479561945,
      "language_loss": 0.67757899,
      "learning_rate": 2.577447098146265e-06,
      "loss": 0.69973963,
      "num_input_tokens_seen": 76065770,
      "step": 3531,
      "time_per_iteration": 2.7147750854492188
    },
    {
      "auxiliary_loss_clip": 0.01178016,
      "auxiliary_loss_mlp": 0.01037085,
      "balance_loss_clip": 1.06103742,
      "balance_loss_mlp": 1.02810848,
      "epoch": 0.4246978897372693,
      "flos": 27776256958080.0,
      "grad_norm": 1.8612662338074044,
      "language_loss": 0.79114294,
      "learning_rate": 2.5767012571667724e-06,
      "loss": 0.81329393,
      "num_input_tokens_seen": 76085250,
      "step": 3532,
      "time_per_iteration": 2.6709494590759277
    },
    {
      "auxiliary_loss_clip": 0.01174766,
      "auxiliary_loss_mlp": 0.01034006,
      "balance_loss_clip": 1.05525827,
      "balance_loss_mlp": 1.02447486,
      "epoch": 0.42481813262790835,
      "flos": 15596615439360.0,
      "grad_norm": 2.06997653168032,
      "language_loss": 0.68776798,
      "learning_rate": 2.5759553287021587e-06,
      "loss": 0.70985574,
      "num_input_tokens_seen": 76103580,
      "step": 3533,
      "time_per_iteration": 2.668377161026001
    },
    {
      "auxiliary_loss_clip": 0.0120121,
      "auxiliary_loss_mlp": 0.01038868,
      "balance_loss_clip": 1.06140769,
      "balance_loss_mlp": 1.03013551,
      "epoch": 0.42493837551854746,
      "flos": 23951088881280.0,
      "grad_norm": 1.9077248076538882,
      "language_loss": 0.77735448,
      "learning_rate": 2.5752093128655786e-06,
      "loss": 0.79975528,
      "num_input_tokens_seen": 76121825,
      "step": 3534,
      "time_per_iteration": 2.6442763805389404
    },
    {
      "auxiliary_loss_clip": 0.01170914,
      "auxiliary_loss_mlp": 0.0103278,
      "balance_loss_clip": 1.05411172,
      "balance_loss_mlp": 1.02448273,
      "epoch": 0.4250586184091866,
      "flos": 20813466009600.0,
      "grad_norm": 3.4952775235465103,
      "language_loss": 0.74281448,
      "learning_rate": 2.574463209770204e-06,
      "loss": 0.76485145,
      "num_input_tokens_seen": 76141140,
      "step": 3535,
      "time_per_iteration": 2.68278431892395
    },
    {
      "auxiliary_loss_clip": 0.01200666,
      "auxiliary_loss_mlp": 0.01038566,
      "balance_loss_clip": 1.06056559,
      "balance_loss_mlp": 1.02965522,
      "epoch": 0.42517886129982563,
      "flos": 30371018607360.0,
      "grad_norm": 1.6832068744264426,
      "language_loss": 0.79687011,
      "learning_rate": 2.5737170195292165e-06,
      "loss": 0.81926244,
      "num_input_tokens_seen": 76164475,
      "step": 3536,
      "time_per_iteration": 2.6842222213745117
    },
    {
      "auxiliary_loss_clip": 0.01186428,
      "auxiliary_loss_mlp": 0.01034257,
      "balance_loss_clip": 1.05711222,
      "balance_loss_mlp": 1.02477956,
      "epoch": 0.42529910419046474,
      "flos": 20080636732800.0,
      "grad_norm": 3.3993954926702927,
      "language_loss": 0.78413969,
      "learning_rate": 2.572970742255814e-06,
      "loss": 0.80634654,
      "num_input_tokens_seen": 76182965,
      "step": 3537,
      "time_per_iteration": 2.7087533473968506
    },
    {
      "auxiliary_loss_clip": 0.01184741,
      "auxiliary_loss_mlp": 0.01036987,
      "balance_loss_clip": 1.060884,
      "balance_loss_mlp": 1.02792704,
      "epoch": 0.42541934708110385,
      "flos": 22632448694400.0,
      "grad_norm": 1.9079029302489992,
      "language_loss": 0.8155688,
      "learning_rate": 2.5722243780632046e-06,
      "loss": 0.83778602,
      "num_input_tokens_seen": 76201230,
      "step": 3538,
      "time_per_iteration": 2.722921133041382
    },
    {
      "auxiliary_loss_clip": 0.01071461,
      "auxiliary_loss_mlp": 0.01002202,
      "balance_loss_clip": 1.01943064,
      "balance_loss_mlp": 1.00070024,
      "epoch": 0.4255395899717429,
      "flos": 66200676186240.0,
      "grad_norm": 0.7533634526823585,
      "language_loss": 0.60503167,
      "learning_rate": 2.5714779270646125e-06,
      "loss": 0.6257683,
      "num_input_tokens_seen": 76262000,
      "step": 3539,
      "time_per_iteration": 3.2524805068969727
    },
    {
      "auxiliary_loss_clip": 0.01189781,
      "auxiliary_loss_mlp": 0.01034445,
      "balance_loss_clip": 1.0608865,
      "balance_loss_mlp": 1.02498007,
      "epoch": 0.425659832862382,
      "flos": 17931814433280.0,
      "grad_norm": 4.2160695331778575,
      "language_loss": 0.775702,
      "learning_rate": 2.5707313893732735e-06,
      "loss": 0.79794431,
      "num_input_tokens_seen": 76280540,
      "step": 3540,
      "time_per_iteration": 2.8729169368743896
    },
    {
      "auxiliary_loss_clip": 0.01187547,
      "auxiliary_loss_mlp": 0.01032345,
      "balance_loss_clip": 1.06177306,
      "balance_loss_mlp": 1.02310073,
      "epoch": 0.4257800757530211,
      "flos": 24022550989440.0,
      "grad_norm": 1.7251406240955045,
      "language_loss": 0.77276134,
      "learning_rate": 2.5699847651024364e-06,
      "loss": 0.79496026,
      "num_input_tokens_seen": 76301180,
      "step": 3541,
      "time_per_iteration": 2.727776050567627
    },
    {
      "auxiliary_loss_clip": 0.01196595,
      "auxiliary_loss_mlp": 0.01036532,
      "balance_loss_clip": 1.06094718,
      "balance_loss_mlp": 1.02851486,
      "epoch": 0.4259003186436602,
      "flos": 23696015425920.0,
      "grad_norm": 3.9960102945046643,
      "language_loss": 0.77514386,
      "learning_rate": 2.5692380543653627e-06,
      "loss": 0.79747516,
      "num_input_tokens_seen": 76319335,
      "step": 3542,
      "time_per_iteration": 2.6203112602233887
    },
    {
      "auxiliary_loss_clip": 0.01169769,
      "auxiliary_loss_mlp": 0.0076331,
      "balance_loss_clip": 1.05740261,
      "balance_loss_mlp": 1.00071836,
      "epoch": 0.4260205615342993,
      "flos": 15259772672640.0,
      "grad_norm": 1.9308386573022054,
      "language_loss": 0.69956309,
      "learning_rate": 2.5684912572753293e-06,
      "loss": 0.71889389,
      "num_input_tokens_seen": 76335010,
      "step": 3543,
      "time_per_iteration": 2.6897690296173096
    },
    {
      "auxiliary_loss_clip": 0.01177089,
      "auxiliary_loss_mlp": 0.01035881,
      "balance_loss_clip": 1.05406201,
      "balance_loss_mlp": 1.02782869,
      "epoch": 0.4261408044249384,
      "flos": 30665306736000.0,
      "grad_norm": 5.584249624883566,
      "language_loss": 0.84590894,
      "learning_rate": 2.5677443739456245e-06,
      "loss": 0.86803865,
      "num_input_tokens_seen": 76356670,
      "step": 3544,
      "time_per_iteration": 3.6833224296569824
    },
    {
      "auxiliary_loss_clip": 0.01183196,
      "auxiliary_loss_mlp": 0.00762926,
      "balance_loss_clip": 1.05932784,
      "balance_loss_mlp": 1.00092077,
      "epoch": 0.42626104731557746,
      "flos": 23257905240960.0,
      "grad_norm": 3.008609824427484,
      "language_loss": 0.79695213,
      "learning_rate": 2.5669974044895495e-06,
      "loss": 0.81641328,
      "num_input_tokens_seen": 76373065,
      "step": 3545,
      "time_per_iteration": 3.636242628097534
    },
    {
      "auxiliary_loss_clip": 0.01180282,
      "auxiliary_loss_mlp": 0.01032076,
      "balance_loss_clip": 1.05740285,
      "balance_loss_mlp": 1.02334452,
      "epoch": 0.42638129020621657,
      "flos": 25884770670720.0,
      "grad_norm": 2.7699366523996005,
      "language_loss": 0.79778922,
      "learning_rate": 2.5662503490204187e-06,
      "loss": 0.81991279,
      "num_input_tokens_seen": 76393230,
      "step": 3546,
      "time_per_iteration": 2.6930651664733887
    },
    {
      "auxiliary_loss_clip": 0.01197501,
      "auxiliary_loss_mlp": 0.01030726,
      "balance_loss_clip": 1.05957937,
      "balance_loss_mlp": 1.0214994,
      "epoch": 0.4265015330968556,
      "flos": 26502362138880.0,
      "grad_norm": 2.1718057118541525,
      "language_loss": 0.76247305,
      "learning_rate": 2.5655032076515603e-06,
      "loss": 0.78475535,
      "num_input_tokens_seen": 76412555,
      "step": 3547,
      "time_per_iteration": 3.6259543895721436
    },
    {
      "auxiliary_loss_clip": 0.01199006,
      "auxiliary_loss_mlp": 0.01033277,
      "balance_loss_clip": 1.06154251,
      "balance_loss_mlp": 1.02440178,
      "epoch": 0.42662177598749473,
      "flos": 24389522288640.0,
      "grad_norm": 3.490032233399255,
      "language_loss": 0.82070148,
      "learning_rate": 2.5647559804963155e-06,
      "loss": 0.84302425,
      "num_input_tokens_seen": 76432485,
      "step": 3548,
      "time_per_iteration": 2.6748783588409424
    },
    {
      "auxiliary_loss_clip": 0.011636,
      "auxiliary_loss_mlp": 0.01033439,
      "balance_loss_clip": 1.05800223,
      "balance_loss_mlp": 1.02456439,
      "epoch": 0.42674201887813384,
      "flos": 23148629089920.0,
      "grad_norm": 1.91822010601034,
      "language_loss": 0.78862911,
      "learning_rate": 2.5640086676680364e-06,
      "loss": 0.81059951,
      "num_input_tokens_seen": 76453980,
      "step": 3549,
      "time_per_iteration": 2.704233407974243
    },
    {
      "auxiliary_loss_clip": 0.01189444,
      "auxiliary_loss_mlp": 0.01029871,
      "balance_loss_clip": 1.05906391,
      "balance_loss_mlp": 1.02031028,
      "epoch": 0.4268622617687729,
      "flos": 21689614552320.0,
      "grad_norm": 2.7203946136880908,
      "language_loss": 0.81217325,
      "learning_rate": 2.5632612692800923e-06,
      "loss": 0.83436632,
      "num_input_tokens_seen": 76473045,
      "step": 3550,
      "time_per_iteration": 2.728198289871216
    },
    {
      "auxiliary_loss_clip": 0.0119111,
      "auxiliary_loss_mlp": 0.01037604,
      "balance_loss_clip": 1.06087017,
      "balance_loss_mlp": 1.02775121,
      "epoch": 0.426982504659412,
      "flos": 23440151871360.0,
      "grad_norm": 2.2755422942062133,
      "language_loss": 0.75488001,
      "learning_rate": 2.5625137854458603e-06,
      "loss": 0.7771672,
      "num_input_tokens_seen": 76492060,
      "step": 3551,
      "time_per_iteration": 2.6499581336975098
    },
    {
      "auxiliary_loss_clip": 0.01171208,
      "auxiliary_loss_mlp": 0.01031761,
      "balance_loss_clip": 1.05446708,
      "balance_loss_mlp": 1.02341032,
      "epoch": 0.4271027475500511,
      "flos": 18916556768640.0,
      "grad_norm": 4.100137874338114,
      "language_loss": 0.80615735,
      "learning_rate": 2.561766216278735e-06,
      "loss": 0.82818705,
      "num_input_tokens_seen": 76509655,
      "step": 3552,
      "time_per_iteration": 3.685493230819702
    },
    {
      "auxiliary_loss_clip": 0.01185333,
      "auxiliary_loss_mlp": 0.01029294,
      "balance_loss_clip": 1.05879855,
      "balance_loss_mlp": 1.02089012,
      "epoch": 0.4272229904406902,
      "flos": 26870554500480.0,
      "grad_norm": 2.188805737520951,
      "language_loss": 0.81562424,
      "learning_rate": 2.561018561892121e-06,
      "loss": 0.83777058,
      "num_input_tokens_seen": 76528795,
      "step": 3553,
      "time_per_iteration": 2.793710947036743
    },
    {
      "auxiliary_loss_clip": 0.01180557,
      "auxiliary_loss_mlp": 0.01031621,
      "balance_loss_clip": 1.05974233,
      "balance_loss_mlp": 1.02281713,
      "epoch": 0.4273432333313293,
      "flos": 23951376190080.0,
      "grad_norm": 1.5997644130704105,
      "language_loss": 0.76633334,
      "learning_rate": 2.5602708223994363e-06,
      "loss": 0.78845513,
      "num_input_tokens_seen": 76550660,
      "step": 3554,
      "time_per_iteration": 2.741751194000244
    },
    {
      "auxiliary_loss_clip": 0.01198114,
      "auxiliary_loss_mlp": 0.0103954,
      "balance_loss_clip": 1.05980313,
      "balance_loss_mlp": 1.03053975,
      "epoch": 0.4274634762219684,
      "flos": 29570354496000.0,
      "grad_norm": 2.2249702041387445,
      "language_loss": 0.6742847,
      "learning_rate": 2.559522997914115e-06,
      "loss": 0.69666123,
      "num_input_tokens_seen": 76570240,
      "step": 3555,
      "time_per_iteration": 2.7435317039489746
    },
    {
      "auxiliary_loss_clip": 0.01198664,
      "auxiliary_loss_mlp": 0.01029503,
      "balance_loss_clip": 1.06289208,
      "balance_loss_mlp": 1.02103925,
      "epoch": 0.42758371911260745,
      "flos": 21434146047360.0,
      "grad_norm": 2.062200105055078,
      "language_loss": 0.84615153,
      "learning_rate": 2.558775088549599e-06,
      "loss": 0.86843318,
      "num_input_tokens_seen": 76589820,
      "step": 3556,
      "time_per_iteration": 2.5956761837005615
    },
    {
      "auxiliary_loss_clip": 0.01201035,
      "auxiliary_loss_mlp": 0.01030962,
      "balance_loss_clip": 1.06103003,
      "balance_loss_mlp": 1.02103758,
      "epoch": 0.42770396200324656,
      "flos": 14752822072320.0,
      "grad_norm": 2.787861095693443,
      "language_loss": 0.67239225,
      "learning_rate": 2.5580270944193467e-06,
      "loss": 0.69471228,
      "num_input_tokens_seen": 76606640,
      "step": 3557,
      "time_per_iteration": 2.6246883869171143
    },
    {
      "auxiliary_loss_clip": 0.01053699,
      "auxiliary_loss_mlp": 0.01002149,
      "balance_loss_clip": 1.01474619,
      "balance_loss_mlp": 1.00074196,
      "epoch": 0.4278242048938857,
      "flos": 70654712601600.0,
      "grad_norm": 0.7555190824691286,
      "language_loss": 0.5543521,
      "learning_rate": 2.557279015636827e-06,
      "loss": 0.57491058,
      "num_input_tokens_seen": 76667050,
      "step": 3558,
      "time_per_iteration": 3.2826087474823
    },
    {
      "auxiliary_loss_clip": 0.01088393,
      "auxiliary_loss_mlp": 0.01003971,
      "balance_loss_clip": 1.01824856,
      "balance_loss_mlp": 1.00261772,
      "epoch": 0.42794444778452473,
      "flos": 69366165033600.0,
      "grad_norm": 0.7692314333193131,
      "language_loss": 0.61249512,
      "learning_rate": 2.5565308523155245e-06,
      "loss": 0.63341874,
      "num_input_tokens_seen": 76726650,
      "step": 3559,
      "time_per_iteration": 3.1037161350250244
    },
    {
      "auxiliary_loss_clip": 0.01154367,
      "auxiliary_loss_mlp": 0.01039122,
      "balance_loss_clip": 1.05469847,
      "balance_loss_mlp": 1.03019643,
      "epoch": 0.42806469067516384,
      "flos": 18215328481920.0,
      "grad_norm": 2.6984983185158513,
      "language_loss": 0.81985843,
      "learning_rate": 2.5557826045689336e-06,
      "loss": 0.8417933,
      "num_input_tokens_seen": 76742890,
      "step": 3560,
      "time_per_iteration": 2.7094414234161377
    },
    {
      "auxiliary_loss_clip": 0.01079174,
      "auxiliary_loss_mlp": 0.0100518,
      "balance_loss_clip": 1.01808786,
      "balance_loss_mlp": 1.00370169,
      "epoch": 0.4281849335658029,
      "flos": 54535814432640.0,
      "grad_norm": 0.817448613880517,
      "language_loss": 0.58750242,
      "learning_rate": 2.5550342725105643e-06,
      "loss": 0.60834599,
      "num_input_tokens_seen": 76801055,
      "step": 3561,
      "time_per_iteration": 3.193030834197998
    },
    {
      "auxiliary_loss_clip": 0.01201285,
      "auxiliary_loss_mlp": 0.01030996,
      "balance_loss_clip": 1.06196773,
      "balance_loss_mlp": 1.02171588,
      "epoch": 0.428305176456442,
      "flos": 17274828723840.0,
      "grad_norm": 5.464456455925295,
      "language_loss": 0.81055284,
      "learning_rate": 2.554285856253937e-06,
      "loss": 0.83287561,
      "num_input_tokens_seen": 76819890,
      "step": 3562,
      "time_per_iteration": 2.687479019165039
    },
    {
      "auxiliary_loss_clip": 0.01198266,
      "auxiliary_loss_mlp": 0.01038015,
      "balance_loss_clip": 1.06123519,
      "balance_loss_mlp": 1.02925968,
      "epoch": 0.4284254193470811,
      "flos": 26359509749760.0,
      "grad_norm": 2.6533913600197643,
      "language_loss": 0.77448851,
      "learning_rate": 2.5535373559125855e-06,
      "loss": 0.79685134,
      "num_input_tokens_seen": 76840255,
      "step": 3563,
      "time_per_iteration": 2.7013535499572754
    },
    {
      "auxiliary_loss_clip": 0.01150085,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.05449915,
      "balance_loss_mlp": 1.02015018,
      "epoch": 0.42854566223772017,
      "flos": 29714248379520.0,
      "grad_norm": 1.6637685150513044,
      "language_loss": 0.8202275,
      "learning_rate": 2.552788771600057e-06,
      "loss": 0.84202814,
      "num_input_tokens_seen": 76860565,
      "step": 3564,
      "time_per_iteration": 2.916154146194458
    },
    {
      "auxiliary_loss_clip": 0.01159945,
      "auxiliary_loss_mlp": 0.01033194,
      "balance_loss_clip": 1.05491877,
      "balance_loss_mlp": 1.02416396,
      "epoch": 0.4286659051283593,
      "flos": 22018161277440.0,
      "grad_norm": 1.8873955702400298,
      "language_loss": 0.82096994,
      "learning_rate": 2.5520401034299118e-06,
      "loss": 0.84290129,
      "num_input_tokens_seen": 76878325,
      "step": 3565,
      "time_per_iteration": 2.696732759475708
    },
    {
      "auxiliary_loss_clip": 0.01181987,
      "auxiliary_loss_mlp": 0.0103488,
      "balance_loss_clip": 1.05673456,
      "balance_loss_mlp": 1.02450848,
      "epoch": 0.4287861480189984,
      "flos": 13334422838400.0,
      "grad_norm": 3.0002681492447465,
      "language_loss": 0.87845564,
      "learning_rate": 2.551291351515722e-06,
      "loss": 0.90062428,
      "num_input_tokens_seen": 76895340,
      "step": 3566,
      "time_per_iteration": 2.6322438716888428
    },
    {
      "auxiliary_loss_clip": 0.01163572,
      "auxiliary_loss_mlp": 0.01028307,
      "balance_loss_clip": 1.05082023,
      "balance_loss_mlp": 1.01913357,
      "epoch": 0.42890639090963745,
      "flos": 26651535321600.0,
      "grad_norm": 1.7568281061687419,
      "language_loss": 0.86074615,
      "learning_rate": 2.5505425159710726e-06,
      "loss": 0.88266492,
      "num_input_tokens_seen": 76915150,
      "step": 3567,
      "time_per_iteration": 2.7070446014404297
    },
    {
      "auxiliary_loss_clip": 0.01183571,
      "auxiliary_loss_mlp": 0.01038597,
      "balance_loss_clip": 1.05824208,
      "balance_loss_mlp": 1.02940047,
      "epoch": 0.42902663380027656,
      "flos": 24055768091520.0,
      "grad_norm": 3.417254366804985,
      "language_loss": 0.8364318,
      "learning_rate": 2.549793596909561e-06,
      "loss": 0.85865349,
      "num_input_tokens_seen": 76933770,
      "step": 3568,
      "time_per_iteration": 2.6951069831848145
    },
    {
      "auxiliary_loss_clip": 0.01182347,
      "auxiliary_loss_mlp": 0.01030661,
      "balance_loss_clip": 1.05397952,
      "balance_loss_mlp": 1.02121985,
      "epoch": 0.42914687669091567,
      "flos": 15632561975040.0,
      "grad_norm": 2.4167389710500373,
      "language_loss": 0.6617856,
      "learning_rate": 2.5490445944447976e-06,
      "loss": 0.68391573,
      "num_input_tokens_seen": 76952265,
      "step": 3569,
      "time_per_iteration": 2.5941851139068604
    },
    {
      "auxiliary_loss_clip": 0.01182288,
      "auxiliary_loss_mlp": 0.01032726,
      "balance_loss_clip": 1.055022,
      "balance_loss_mlp": 1.02377915,
      "epoch": 0.4292671195815547,
      "flos": 31467802440960.0,
      "grad_norm": 2.644998150179643,
      "language_loss": 0.65618145,
      "learning_rate": 2.548295508690406e-06,
      "loss": 0.67833161,
      "num_input_tokens_seen": 76973560,
      "step": 3570,
      "time_per_iteration": 3.735474109649658
    },
    {
      "auxiliary_loss_clip": 0.01163662,
      "auxiliary_loss_mlp": 0.01038144,
      "balance_loss_clip": 1.05278778,
      "balance_loss_mlp": 1.02878559,
      "epoch": 0.42938736247219383,
      "flos": 30257756046720.0,
      "grad_norm": 1.954468908911921,
      "language_loss": 0.76526564,
      "learning_rate": 2.5475463397600217e-06,
      "loss": 0.78728372,
      "num_input_tokens_seen": 76993640,
      "step": 3571,
      "time_per_iteration": 3.98801851272583
    },
    {
      "auxiliary_loss_clip": 0.01202651,
      "auxiliary_loss_mlp": 0.01028848,
      "balance_loss_clip": 1.06194878,
      "balance_loss_mlp": 1.01974607,
      "epoch": 0.42950760536283294,
      "flos": 29349683291520.0,
      "grad_norm": 2.3187114196596887,
      "language_loss": 0.77458715,
      "learning_rate": 2.546797087767293e-06,
      "loss": 0.79690206,
      "num_input_tokens_seen": 77013765,
      "step": 3572,
      "time_per_iteration": 2.6997029781341553
    },
    {
      "auxiliary_loss_clip": 0.01194392,
      "auxiliary_loss_mlp": 0.01034937,
      "balance_loss_clip": 1.05715299,
      "balance_loss_mlp": 1.02504277,
      "epoch": 0.429627848253472,
      "flos": 26869943969280.0,
      "grad_norm": 2.0302994228884588,
      "language_loss": 0.87506086,
      "learning_rate": 2.546047752825881e-06,
      "loss": 0.89735413,
      "num_input_tokens_seen": 77034370,
      "step": 3573,
      "time_per_iteration": 3.611482620239258
    },
    {
      "auxiliary_loss_clip": 0.01180482,
      "auxiliary_loss_mlp": 0.01029,
      "balance_loss_clip": 1.05735672,
      "balance_loss_mlp": 1.02005303,
      "epoch": 0.4297480911441111,
      "flos": 13881270470400.0,
      "grad_norm": 3.107754462186435,
      "language_loss": 0.93400776,
      "learning_rate": 2.5452983350494595e-06,
      "loss": 0.95610249,
      "num_input_tokens_seen": 77049925,
      "step": 3574,
      "time_per_iteration": 2.6593875885009766
    },
    {
      "auxiliary_loss_clip": 0.0114901,
      "auxiliary_loss_mlp": 0.01035216,
      "balance_loss_clip": 1.05491471,
      "balance_loss_mlp": 1.02630496,
      "epoch": 0.4298683340347502,
      "flos": 20741141975040.0,
      "grad_norm": 2.7594701006781,
      "language_loss": 0.65671396,
      "learning_rate": 2.544548834551713e-06,
      "loss": 0.67855626,
      "num_input_tokens_seen": 77068930,
      "step": 3575,
      "time_per_iteration": 2.7098872661590576
    },
    {
      "auxiliary_loss_clip": 0.01179954,
      "auxiliary_loss_mlp": 0.01034769,
      "balance_loss_clip": 1.05912828,
      "balance_loss_mlp": 1.0258882,
      "epoch": 0.4299885769253893,
      "flos": 20882126856960.0,
      "grad_norm": 2.2061364056546955,
      "language_loss": 0.94314522,
      "learning_rate": 2.5437992514463424e-06,
      "loss": 0.96529245,
      "num_input_tokens_seen": 77082255,
      "step": 3576,
      "time_per_iteration": 2.6763854026794434
    },
    {
      "auxiliary_loss_clip": 0.01198428,
      "auxiliary_loss_mlp": 0.01031465,
      "balance_loss_clip": 1.05829215,
      "balance_loss_mlp": 1.02217317,
      "epoch": 0.4301088198160284,
      "flos": 25484618183040.0,
      "grad_norm": 2.173851235526867,
      "language_loss": 0.88343811,
      "learning_rate": 2.5430495858470565e-06,
      "loss": 0.90573704,
      "num_input_tokens_seen": 77101725,
      "step": 3577,
      "time_per_iteration": 3.7185683250427246
    },
    {
      "auxiliary_loss_clip": 0.01175039,
      "auxiliary_loss_mlp": 0.01024611,
      "balance_loss_clip": 1.05500984,
      "balance_loss_mlp": 1.01586688,
      "epoch": 0.43022906270666744,
      "flos": 18259427404800.0,
      "grad_norm": 3.1739477196672086,
      "language_loss": 0.77852046,
      "learning_rate": 2.54229983786758e-06,
      "loss": 0.8005169,
      "num_input_tokens_seen": 77119670,
      "step": 3578,
      "time_per_iteration": 2.619114875793457
    },
    {
      "auxiliary_loss_clip": 0.01194962,
      "auxiliary_loss_mlp": 0.01037521,
      "balance_loss_clip": 1.05820131,
      "balance_loss_mlp": 1.02822232,
      "epoch": 0.43034930559730655,
      "flos": 23399536567680.0,
      "grad_norm": 2.1441252915297957,
      "language_loss": 0.85152936,
      "learning_rate": 2.541550007621651e-06,
      "loss": 0.87385416,
      "num_input_tokens_seen": 77138160,
      "step": 3579,
      "time_per_iteration": 2.657625675201416
    },
    {
      "auxiliary_loss_clip": 0.01183415,
      "auxiliary_loss_mlp": 0.01027844,
      "balance_loss_clip": 1.05786467,
      "balance_loss_mlp": 1.01948774,
      "epoch": 0.43046954848794566,
      "flos": 28184382264960.0,
      "grad_norm": 2.302765240406539,
      "language_loss": 0.80447769,
      "learning_rate": 2.5408000952230156e-06,
      "loss": 0.8265903,
      "num_input_tokens_seen": 77156950,
      "step": 3580,
      "time_per_iteration": 2.7065446376800537
    },
    {
      "auxiliary_loss_clip": 0.0119739,
      "auxiliary_loss_mlp": 0.01034498,
      "balance_loss_clip": 1.05735838,
      "balance_loss_mlp": 1.02529478,
      "epoch": 0.4305897913785847,
      "flos": 28580476515840.0,
      "grad_norm": 2.688954903051144,
      "language_loss": 0.90930951,
      "learning_rate": 2.5400501007854357e-06,
      "loss": 0.93162841,
      "num_input_tokens_seen": 77176395,
      "step": 3581,
      "time_per_iteration": 2.6892611980438232
    },
    {
      "auxiliary_loss_clip": 0.01194792,
      "auxiliary_loss_mlp": 0.01036705,
      "balance_loss_clip": 1.0594542,
      "balance_loss_mlp": 1.0277884,
      "epoch": 0.43071003426922383,
      "flos": 20448721353600.0,
      "grad_norm": 2.1108598456670156,
      "language_loss": 0.75838292,
      "learning_rate": 2.539300024422685e-06,
      "loss": 0.78069782,
      "num_input_tokens_seen": 77194340,
      "step": 3582,
      "time_per_iteration": 2.593695640563965
    },
    {
      "auxiliary_loss_clip": 0.01069064,
      "auxiliary_loss_mlp": 0.01004702,
      "balance_loss_clip": 1.01563764,
      "balance_loss_mlp": 1.0032295,
      "epoch": 0.43083027715986294,
      "flos": 51997969883520.0,
      "grad_norm": 0.8354183542807252,
      "language_loss": 0.60903233,
      "learning_rate": 2.538549866248549e-06,
      "loss": 0.62976998,
      "num_input_tokens_seen": 77249320,
      "step": 3583,
      "time_per_iteration": 3.126816987991333
    },
    {
      "auxiliary_loss_clip": 0.01181609,
      "auxiliary_loss_mlp": 0.01031554,
      "balance_loss_clip": 1.05693471,
      "balance_loss_mlp": 1.02303064,
      "epoch": 0.430950520050502,
      "flos": 16690885320960.0,
      "grad_norm": 2.172048421876102,
      "language_loss": 0.81099927,
      "learning_rate": 2.5377996263768274e-06,
      "loss": 0.83313096,
      "num_input_tokens_seen": 77267400,
      "step": 3584,
      "time_per_iteration": 2.571611166000366
    },
    {
      "auxiliary_loss_clip": 0.01197732,
      "auxiliary_loss_mlp": 0.01036811,
      "balance_loss_clip": 1.06058979,
      "balance_loss_mlp": 1.0279839,
      "epoch": 0.4310707629411411,
      "flos": 24608433726720.0,
      "grad_norm": 2.1545459124471464,
      "language_loss": 0.68737555,
      "learning_rate": 2.5370493049213293e-06,
      "loss": 0.70972097,
      "num_input_tokens_seen": 77287045,
      "step": 3585,
      "time_per_iteration": 2.705486297607422
    },
    {
      "auxiliary_loss_clip": 0.01196141,
      "auxiliary_loss_mlp": 0.01035858,
      "balance_loss_clip": 1.05579233,
      "balance_loss_mlp": 1.02576137,
      "epoch": 0.4311910058317802,
      "flos": 26432983019520.0,
      "grad_norm": 2.346715181727409,
      "language_loss": 0.79820991,
      "learning_rate": 2.536298901995878e-06,
      "loss": 0.82052994,
      "num_input_tokens_seen": 77306255,
      "step": 3586,
      "time_per_iteration": 2.6403071880340576
    },
    {
      "auxiliary_loss_clip": 0.0119822,
      "auxiliary_loss_mlp": 0.01030724,
      "balance_loss_clip": 1.05986512,
      "balance_loss_mlp": 1.0211699,
      "epoch": 0.43131124872241927,
      "flos": 25155891889920.0,
      "grad_norm": 1.9453408405162531,
      "language_loss": 0.80227172,
      "learning_rate": 2.535548417714311e-06,
      "loss": 0.82456112,
      "num_input_tokens_seen": 77325555,
      "step": 3587,
      "time_per_iteration": 2.723257064819336
    },
    {
      "auxiliary_loss_clip": 0.01196992,
      "auxiliary_loss_mlp": 0.01030759,
      "balance_loss_clip": 1.05856633,
      "balance_loss_mlp": 1.02218831,
      "epoch": 0.4314314916130584,
      "flos": 21614812479360.0,
      "grad_norm": 1.7198055551961549,
      "language_loss": 0.87354511,
      "learning_rate": 2.534797852190474e-06,
      "loss": 0.89582258,
      "num_input_tokens_seen": 77345735,
      "step": 3588,
      "time_per_iteration": 2.648099660873413
    },
    {
      "auxiliary_loss_clip": 0.01194638,
      "auxiliary_loss_mlp": 0.01039629,
      "balance_loss_clip": 1.0573988,
      "balance_loss_mlp": 1.03061044,
      "epoch": 0.4315517345036975,
      "flos": 19275016544640.0,
      "grad_norm": 3.566667249842902,
      "language_loss": 0.81833804,
      "learning_rate": 2.5340472055382283e-06,
      "loss": 0.84068066,
      "num_input_tokens_seen": 77361765,
      "step": 3589,
      "time_per_iteration": 2.591543674468994
    },
    {
      "auxiliary_loss_clip": 0.01193521,
      "auxiliary_loss_mlp": 0.00762508,
      "balance_loss_clip": 1.05861354,
      "balance_loss_mlp": 1.00083554,
      "epoch": 0.43167197739433655,
      "flos": 24273853516800.0,
      "grad_norm": 2.627099008637973,
      "language_loss": 0.81426823,
      "learning_rate": 2.5332964778714468e-06,
      "loss": 0.83382851,
      "num_input_tokens_seen": 77378950,
      "step": 3590,
      "time_per_iteration": 2.682359218597412
    },
    {
      "auxiliary_loss_clip": 0.0119674,
      "auxiliary_loss_mlp": 0.01038057,
      "balance_loss_clip": 1.05925393,
      "balance_loss_mlp": 1.02943254,
      "epoch": 0.43179222028497566,
      "flos": 16867816738560.0,
      "grad_norm": 2.292673981562363,
      "language_loss": 0.66354442,
      "learning_rate": 2.5325456693040123e-06,
      "loss": 0.68589234,
      "num_input_tokens_seen": 77396145,
      "step": 3591,
      "time_per_iteration": 2.6055665016174316
    },
    {
      "auxiliary_loss_clip": 0.01188659,
      "auxiliary_loss_mlp": 0.01026694,
      "balance_loss_clip": 1.05599475,
      "balance_loss_mlp": 1.01663828,
      "epoch": 0.43191246317561477,
      "flos": 17639214243840.0,
      "grad_norm": 2.611609410131861,
      "language_loss": 0.75086498,
      "learning_rate": 2.531794779949824e-06,
      "loss": 0.77301848,
      "num_input_tokens_seen": 77414045,
      "step": 3592,
      "time_per_iteration": 2.660191059112549
    },
    {
      "auxiliary_loss_clip": 0.01129394,
      "auxiliary_loss_mlp": 0.01031007,
      "balance_loss_clip": 1.05135775,
      "balance_loss_mlp": 1.02272224,
      "epoch": 0.4320327060662538,
      "flos": 23878800760320.0,
      "grad_norm": 1.8350074805078391,
      "language_loss": 0.884027,
      "learning_rate": 2.5310438099227903e-06,
      "loss": 0.90563095,
      "num_input_tokens_seen": 77431310,
      "step": 3593,
      "time_per_iteration": 2.760456085205078
    },
    {
      "auxiliary_loss_clip": 0.01084212,
      "auxiliary_loss_mlp": 0.00999982,
      "balance_loss_clip": 1.01544738,
      "balance_loss_mlp": 0.99852794,
      "epoch": 0.43215294895689293,
      "flos": 66394917959040.0,
      "grad_norm": 0.8053949336305092,
      "language_loss": 0.5335201,
      "learning_rate": 2.530292759336833e-06,
      "loss": 0.55436206,
      "num_input_tokens_seen": 77492045,
      "step": 3594,
      "time_per_iteration": 3.259648084640503
    },
    {
      "auxiliary_loss_clip": 0.01199866,
      "auxiliary_loss_mlp": 0.00763893,
      "balance_loss_clip": 1.05920339,
      "balance_loss_mlp": 1.00081253,
      "epoch": 0.432273191847532,
      "flos": 20594267262720.0,
      "grad_norm": 2.190160955257522,
      "language_loss": 0.70095086,
      "learning_rate": 2.5295416283058855e-06,
      "loss": 0.72058845,
      "num_input_tokens_seen": 77510910,
      "step": 3595,
      "time_per_iteration": 2.6244301795959473
    },
    {
      "auxiliary_loss_clip": 0.01182159,
      "auxiliary_loss_mlp": 0.01024742,
      "balance_loss_clip": 1.05999827,
      "balance_loss_mlp": 1.01704681,
      "epoch": 0.4323934347381711,
      "flos": 19282127437440.0,
      "grad_norm": 2.145555777211377,
      "language_loss": 0.66407168,
      "learning_rate": 2.5287904169438943e-06,
      "loss": 0.68614066,
      "num_input_tokens_seen": 77530115,
      "step": 3596,
      "time_per_iteration": 3.617952346801758
    },
    {
      "auxiliary_loss_clip": 0.0120296,
      "auxiliary_loss_mlp": 0.01038341,
      "balance_loss_clip": 1.06135786,
      "balance_loss_mlp": 1.02938819,
      "epoch": 0.4325136776288102,
      "flos": 21726315273600.0,
      "grad_norm": 2.740045673924808,
      "language_loss": 0.63722265,
      "learning_rate": 2.528039125364817e-06,
      "loss": 0.65963572,
      "num_input_tokens_seen": 77548920,
      "step": 3597,
      "time_per_iteration": 3.527810573577881
    },
    {
      "auxiliary_loss_clip": 0.01181886,
      "auxiliary_loss_mlp": 0.01035756,
      "balance_loss_clip": 1.05771494,
      "balance_loss_mlp": 1.02549851,
      "epoch": 0.43263392051944927,
      "flos": 22340746344960.0,
      "grad_norm": 4.692508930246249,
      "language_loss": 0.76279002,
      "learning_rate": 2.5272877536826246e-06,
      "loss": 0.78496647,
      "num_input_tokens_seen": 77567715,
      "step": 3598,
      "time_per_iteration": 2.6336781978607178
    },
    {
      "auxiliary_loss_clip": 0.01194514,
      "auxiliary_loss_mlp": 0.01032817,
      "balance_loss_clip": 1.05672741,
      "balance_loss_mlp": 1.0232265,
      "epoch": 0.4327541634100884,
      "flos": 29168406328320.0,
      "grad_norm": 3.8320161537887,
      "language_loss": 0.70740938,
      "learning_rate": 2.5265363020112986e-06,
      "loss": 0.72968268,
      "num_input_tokens_seen": 77588035,
      "step": 3599,
      "time_per_iteration": 2.7562737464904785
    },
    {
      "auxiliary_loss_clip": 0.01161922,
      "auxiliary_loss_mlp": 0.01041518,
      "balance_loss_clip": 1.05655646,
      "balance_loss_mlp": 1.03282118,
      "epoch": 0.4328744063007275,
      "flos": 26067448264320.0,
      "grad_norm": 2.2692550006520404,
      "language_loss": 0.841048,
      "learning_rate": 2.5257847704648344e-06,
      "loss": 0.86308241,
      "num_input_tokens_seen": 77609265,
      "step": 3600,
      "time_per_iteration": 3.658175230026245
    },
    {
      "auxiliary_loss_clip": 0.01157965,
      "auxiliary_loss_mlp": 0.01032575,
      "balance_loss_clip": 1.05391955,
      "balance_loss_mlp": 1.0239805,
      "epoch": 0.43299464919136654,
      "flos": 16581357774720.0,
      "grad_norm": 2.3558321314369706,
      "language_loss": 0.7556088,
      "learning_rate": 2.525033159157239e-06,
      "loss": 0.77751422,
      "num_input_tokens_seen": 77625580,
      "step": 3601,
      "time_per_iteration": 2.6389198303222656
    },
    {
      "auxiliary_loss_clip": 0.01184443,
      "auxiliary_loss_mlp": 0.01029382,
      "balance_loss_clip": 1.06042647,
      "balance_loss_mlp": 1.01974416,
      "epoch": 0.43311489208200565,
      "flos": 16107265140480.0,
      "grad_norm": 4.863254937272786,
      "language_loss": 0.77079165,
      "learning_rate": 2.52428146820253e-06,
      "loss": 0.79293001,
      "num_input_tokens_seen": 77643835,
      "step": 3602,
      "time_per_iteration": 2.6702117919921875
    },
    {
      "auxiliary_loss_clip": 0.0119766,
      "auxiliary_loss_mlp": 0.0103579,
      "balance_loss_clip": 1.05855155,
      "balance_loss_mlp": 1.02586007,
      "epoch": 0.43323513497264476,
      "flos": 22930220442240.0,
      "grad_norm": 2.2454931596583405,
      "language_loss": 0.82249653,
      "learning_rate": 2.52352969771474e-06,
      "loss": 0.84483111,
      "num_input_tokens_seen": 77663060,
      "step": 3603,
      "time_per_iteration": 2.6332507133483887
    },
    {
      "auxiliary_loss_clip": 0.01196897,
      "auxiliary_loss_mlp": 0.01029825,
      "balance_loss_clip": 1.06074965,
      "balance_loss_mlp": 1.02187991,
      "epoch": 0.4333553778632838,
      "flos": 25299031587840.0,
      "grad_norm": 2.8476392373400743,
      "language_loss": 0.88120294,
      "learning_rate": 2.5227778478079106e-06,
      "loss": 0.90347016,
      "num_input_tokens_seen": 77682470,
      "step": 3604,
      "time_per_iteration": 3.915952682495117
    },
    {
      "auxiliary_loss_clip": 0.0118333,
      "auxiliary_loss_mlp": 0.01036618,
      "balance_loss_clip": 1.05745077,
      "balance_loss_mlp": 1.02823162,
      "epoch": 0.43347562075392293,
      "flos": 19387165783680.0,
      "grad_norm": 1.8102083067009938,
      "language_loss": 0.77078831,
      "learning_rate": 2.522025918596098e-06,
      "loss": 0.79298776,
      "num_input_tokens_seen": 77700770,
      "step": 3605,
      "time_per_iteration": 2.717790365219116
    },
    {
      "auxiliary_loss_clip": 0.01192967,
      "auxiliary_loss_mlp": 0.01031419,
      "balance_loss_clip": 1.05925465,
      "balance_loss_mlp": 1.02301431,
      "epoch": 0.43359586364456204,
      "flos": 26325969425280.0,
      "grad_norm": 1.7618492869080662,
      "language_loss": 0.65731013,
      "learning_rate": 2.521273910193368e-06,
      "loss": 0.67955399,
      "num_input_tokens_seen": 77723950,
      "step": 3606,
      "time_per_iteration": 2.6941187381744385
    },
    {
      "auxiliary_loss_clip": 0.01184594,
      "auxiliary_loss_mlp": 0.01038197,
      "balance_loss_clip": 1.06028199,
      "balance_loss_mlp": 1.0291667,
      "epoch": 0.4337161065352011,
      "flos": 15989261984640.0,
      "grad_norm": 2.1943945017818964,
      "language_loss": 0.8725996,
      "learning_rate": 2.5205218227138006e-06,
      "loss": 0.89482749,
      "num_input_tokens_seen": 77736905,
      "step": 3607,
      "time_per_iteration": 2.662612199783325
    },
    {
      "auxiliary_loss_clip": 0.01182412,
      "auxiliary_loss_mlp": 0.01027115,
      "balance_loss_clip": 1.05795884,
      "balance_loss_mlp": 1.01819777,
      "epoch": 0.4338363494258402,
      "flos": 20224710184320.0,
      "grad_norm": 2.1326834253271802,
      "language_loss": 0.79434013,
      "learning_rate": 2.519769656271486e-06,
      "loss": 0.8164354,
      "num_input_tokens_seen": 77754325,
      "step": 3608,
      "time_per_iteration": 2.628711700439453
    },
    {
      "auxiliary_loss_clip": 0.01175736,
      "auxiliary_loss_mlp": 0.01028128,
      "balance_loss_clip": 1.05572736,
      "balance_loss_mlp": 1.01850235,
      "epoch": 0.43395659231647926,
      "flos": 20083904870400.0,
      "grad_norm": 2.014573571641292,
      "language_loss": 0.68115819,
      "learning_rate": 2.5190174109805285e-06,
      "loss": 0.70319688,
      "num_input_tokens_seen": 77774150,
      "step": 3609,
      "time_per_iteration": 2.682858943939209
    },
    {
      "auxiliary_loss_clip": 0.01194481,
      "auxiliary_loss_mlp": 0.0103348,
      "balance_loss_clip": 1.05765462,
      "balance_loss_mlp": 1.02476549,
      "epoch": 0.43407683520711837,
      "flos": 19901801894400.0,
      "grad_norm": 2.4200207713256066,
      "language_loss": 0.64076626,
      "learning_rate": 2.518265086955042e-06,
      "loss": 0.66304588,
      "num_input_tokens_seen": 77791870,
      "step": 3610,
      "time_per_iteration": 2.629099130630493
    },
    {
      "auxiliary_loss_clip": 0.01196575,
      "auxiliary_loss_mlp": 0.0103161,
      "balance_loss_clip": 1.05775273,
      "balance_loss_mlp": 1.02296138,
      "epoch": 0.4341970780977575,
      "flos": 23108732058240.0,
      "grad_norm": 1.9188151347439166,
      "language_loss": 0.83696425,
      "learning_rate": 2.5175126843091534e-06,
      "loss": 0.85924602,
      "num_input_tokens_seen": 77811240,
      "step": 3611,
      "time_per_iteration": 2.6950738430023193
    },
    {
      "auxiliary_loss_clip": 0.01167153,
      "auxiliary_loss_mlp": 0.01034882,
      "balance_loss_clip": 1.05565763,
      "balance_loss_mlp": 1.02591157,
      "epoch": 0.43431732098839654,
      "flos": 37408288406400.0,
      "grad_norm": 2.099501349515304,
      "language_loss": 0.75390363,
      "learning_rate": 2.5167602031570034e-06,
      "loss": 0.77592397,
      "num_input_tokens_seen": 77831425,
      "step": 3612,
      "time_per_iteration": 2.7982349395751953
    },
    {
      "auxiliary_loss_clip": 0.01180942,
      "auxiliary_loss_mlp": 0.010265,
      "balance_loss_clip": 1.05720627,
      "balance_loss_mlp": 1.01780367,
      "epoch": 0.43443756387903565,
      "flos": 31868206323840.0,
      "grad_norm": 1.9440813257035314,
      "language_loss": 0.73668635,
      "learning_rate": 2.51600764361274e-06,
      "loss": 0.75876081,
      "num_input_tokens_seen": 77852950,
      "step": 3613,
      "time_per_iteration": 2.759389877319336
    },
    {
      "auxiliary_loss_clip": 0.01185601,
      "auxiliary_loss_mlp": 0.01034154,
      "balance_loss_clip": 1.05699301,
      "balance_loss_mlp": 1.02552307,
      "epoch": 0.43455780676967476,
      "flos": 23477139901440.0,
      "grad_norm": 4.305542568742036,
      "language_loss": 0.79006225,
      "learning_rate": 2.5152550057905283e-06,
      "loss": 0.81225979,
      "num_input_tokens_seen": 77872840,
      "step": 3614,
      "time_per_iteration": 2.699469804763794
    },
    {
      "auxiliary_loss_clip": 0.01176309,
      "auxiliary_loss_mlp": 0.01031274,
      "balance_loss_clip": 1.05853236,
      "balance_loss_mlp": 1.02187467,
      "epoch": 0.4346780496603138,
      "flos": 24207060176640.0,
      "grad_norm": 2.44372034632876,
      "language_loss": 0.76750547,
      "learning_rate": 2.5145022898045415e-06,
      "loss": 0.7895813,
      "num_input_tokens_seen": 77892025,
      "step": 3615,
      "time_per_iteration": 2.759835958480835
    },
    {
      "auxiliary_loss_clip": 0.01196948,
      "auxiliary_loss_mlp": 0.01035327,
      "balance_loss_clip": 1.05760169,
      "balance_loss_mlp": 1.02608252,
      "epoch": 0.4347982925509529,
      "flos": 17092366611840.0,
      "grad_norm": 4.458022191651743,
      "language_loss": 0.90217221,
      "learning_rate": 2.5137494957689664e-06,
      "loss": 0.92449486,
      "num_input_tokens_seen": 77907635,
      "step": 3616,
      "time_per_iteration": 2.5874581336975098
    },
    {
      "auxiliary_loss_clip": 0.01082088,
      "auxiliary_loss_mlp": 0.01003614,
      "balance_loss_clip": 1.01370811,
      "balance_loss_mlp": 1.00219524,
      "epoch": 0.43491853544159204,
      "flos": 60945544696320.0,
      "grad_norm": 0.7661567765734042,
      "language_loss": 0.57325733,
      "learning_rate": 2.5129966237980016e-06,
      "loss": 0.5941143,
      "num_input_tokens_seen": 77970630,
      "step": 3617,
      "time_per_iteration": 3.2450356483459473
    },
    {
      "auxiliary_loss_clip": 0.01195022,
      "auxiliary_loss_mlp": 0.0103952,
      "balance_loss_clip": 1.05813646,
      "balance_loss_mlp": 1.03027487,
      "epoch": 0.4350387783322311,
      "flos": 21944652094080.0,
      "grad_norm": 1.9623601038070346,
      "language_loss": 0.78412384,
      "learning_rate": 2.512243674005857e-06,
      "loss": 0.8064692,
      "num_input_tokens_seen": 77989995,
      "step": 3618,
      "time_per_iteration": 2.647317886352539
    },
    {
      "auxiliary_loss_clip": 0.01173214,
      "auxiliary_loss_mlp": 0.01028472,
      "balance_loss_clip": 1.05891967,
      "balance_loss_mlp": 1.01956725,
      "epoch": 0.4351590212228702,
      "flos": 25082705928960.0,
      "grad_norm": 1.885584347384447,
      "language_loss": 0.86117297,
      "learning_rate": 2.5114906465067537e-06,
      "loss": 0.8831898,
      "num_input_tokens_seen": 78010980,
      "step": 3619,
      "time_per_iteration": 2.741595983505249
    },
    {
      "auxiliary_loss_clip": 0.01178088,
      "auxiliary_loss_mlp": 0.01035521,
      "balance_loss_clip": 1.05640399,
      "balance_loss_mlp": 1.02625859,
      "epoch": 0.4352792641135093,
      "flos": 21506541909120.0,
      "grad_norm": 5.465861919629209,
      "language_loss": 0.75206274,
      "learning_rate": 2.5107375414149264e-06,
      "loss": 0.77419889,
      "num_input_tokens_seen": 78030225,
      "step": 3620,
      "time_per_iteration": 2.7424161434173584
    },
    {
      "auxiliary_loss_clip": 0.01197414,
      "auxiliary_loss_mlp": 0.01040315,
      "balance_loss_clip": 1.05797362,
      "balance_loss_mlp": 1.03113556,
      "epoch": 0.43539950700414837,
      "flos": 16253457494400.0,
      "grad_norm": 2.683781957756064,
      "language_loss": 0.71913588,
      "learning_rate": 2.5099843588446197e-06,
      "loss": 0.74151313,
      "num_input_tokens_seen": 78048545,
      "step": 3621,
      "time_per_iteration": 2.557609796524048
    },
    {
      "auxiliary_loss_clip": 0.01195396,
      "auxiliary_loss_mlp": 0.01031361,
      "balance_loss_clip": 1.0585649,
      "balance_loss_mlp": 1.02270091,
      "epoch": 0.4355197498947875,
      "flos": 16691819074560.0,
      "grad_norm": 1.7827574894042526,
      "language_loss": 0.6191268,
      "learning_rate": 2.509231098910091e-06,
      "loss": 0.64139438,
      "num_input_tokens_seen": 78068415,
      "step": 3622,
      "time_per_iteration": 3.5928127765655518
    },
    {
      "auxiliary_loss_clip": 0.01198175,
      "auxiliary_loss_mlp": 0.01026917,
      "balance_loss_clip": 1.060323,
      "balance_loss_mlp": 1.01777339,
      "epoch": 0.4356399927854266,
      "flos": 16362733645440.0,
      "grad_norm": 3.3890358814741957,
      "language_loss": 0.75372505,
      "learning_rate": 2.508477761725611e-06,
      "loss": 0.77597594,
      "num_input_tokens_seen": 78086690,
      "step": 3623,
      "time_per_iteration": 3.5387134552001953
    },
    {
      "auxiliary_loss_clip": 0.01179078,
      "auxiliary_loss_mlp": 0.01033477,
      "balance_loss_clip": 1.05515265,
      "balance_loss_mlp": 1.024441,
      "epoch": 0.43576023567606564,
      "flos": 17202037812480.0,
      "grad_norm": 4.1284491471715095,
      "language_loss": 0.81456447,
      "learning_rate": 2.507724347405458e-06,
      "loss": 0.83669007,
      "num_input_tokens_seen": 78104640,
      "step": 3624,
      "time_per_iteration": 2.6221652030944824
    },
    {
      "auxiliary_loss_clip": 0.01177664,
      "auxiliary_loss_mlp": 0.01031282,
      "balance_loss_clip": 1.05234456,
      "balance_loss_mlp": 1.02272248,
      "epoch": 0.43588047856670475,
      "flos": 15917656222080.0,
      "grad_norm": 2.1780911345230387,
      "language_loss": 0.81971371,
      "learning_rate": 2.5069708560639243e-06,
      "loss": 0.84180319,
      "num_input_tokens_seen": 78122550,
      "step": 3625,
      "time_per_iteration": 3.627504348754883
    },
    {
      "auxiliary_loss_clip": 0.01195033,
      "auxiliary_loss_mlp": 0.01028318,
      "balance_loss_clip": 1.05683613,
      "balance_loss_mlp": 1.01932359,
      "epoch": 0.4360007214573438,
      "flos": 23659566099840.0,
      "grad_norm": 8.725496020652757,
      "language_loss": 0.62091708,
      "learning_rate": 2.5062172878153158e-06,
      "loss": 0.64315063,
      "num_input_tokens_seen": 78141825,
      "step": 3626,
      "time_per_iteration": 2.631094217300415
    },
    {
      "auxiliary_loss_clip": 0.0119701,
      "auxiliary_loss_mlp": 0.01037618,
      "balance_loss_clip": 1.0579648,
      "balance_loss_mlp": 1.02830791,
      "epoch": 0.4361209643479829,
      "flos": 21978767036160.0,
      "grad_norm": 2.0996192364737905,
      "language_loss": 0.87548339,
      "learning_rate": 2.505463642773947e-06,
      "loss": 0.89782965,
      "num_input_tokens_seen": 78161790,
      "step": 3627,
      "time_per_iteration": 2.671057939529419
    },
    {
      "auxiliary_loss_clip": 0.0118156,
      "auxiliary_loss_mlp": 0.01034213,
      "balance_loss_clip": 1.05639994,
      "balance_loss_mlp": 1.02477765,
      "epoch": 0.43624120723862203,
      "flos": 17420159151360.0,
      "grad_norm": 2.404697069369866,
      "language_loss": 0.75458443,
      "learning_rate": 2.504709921054146e-06,
      "loss": 0.7767421,
      "num_input_tokens_seen": 78178605,
      "step": 3628,
      "time_per_iteration": 2.7101409435272217
    },
    {
      "auxiliary_loss_clip": 0.01160152,
      "auxiliary_loss_mlp": 0.01028481,
      "balance_loss_clip": 1.05355978,
      "balance_loss_mlp": 1.01950514,
      "epoch": 0.4363614501292611,
      "flos": 17895293280000.0,
      "grad_norm": 2.226648406298736,
      "language_loss": 0.83985871,
      "learning_rate": 2.50395612277025e-06,
      "loss": 0.86174506,
      "num_input_tokens_seen": 78194460,
      "step": 3629,
      "time_per_iteration": 2.6520659923553467
    },
    {
      "auxiliary_loss_clip": 0.01173403,
      "auxiliary_loss_mlp": 0.0103251,
      "balance_loss_clip": 1.05223346,
      "balance_loss_mlp": 1.02331865,
      "epoch": 0.4364816930199002,
      "flos": 20302888135680.0,
      "grad_norm": 2.069122175852786,
      "language_loss": 0.73348391,
      "learning_rate": 2.503202248036612e-06,
      "loss": 0.75554305,
      "num_input_tokens_seen": 78213315,
      "step": 3630,
      "time_per_iteration": 3.5693020820617676
    },
    {
      "auxiliary_loss_clip": 0.01194703,
      "auxiliary_loss_mlp": 0.01035102,
      "balance_loss_clip": 1.05767775,
      "balance_loss_mlp": 1.02620912,
      "epoch": 0.4366019359105393,
      "flos": 24061334699520.0,
      "grad_norm": 1.9239424808017727,
      "language_loss": 0.7368322,
      "learning_rate": 2.5024482969675927e-06,
      "loss": 0.75913024,
      "num_input_tokens_seen": 78233270,
      "step": 3631,
      "time_per_iteration": 2.641449451446533
    },
    {
      "auxiliary_loss_clip": 0.01158515,
      "auxiliary_loss_mlp": 0.0103615,
      "balance_loss_clip": 1.0518645,
      "balance_loss_mlp": 1.0284611,
      "epoch": 0.43672217880117836,
      "flos": 21754109422080.0,
      "grad_norm": 3.263985079389551,
      "language_loss": 0.8443588,
      "learning_rate": 2.501694269677566e-06,
      "loss": 0.86630535,
      "num_input_tokens_seen": 78251040,
      "step": 3632,
      "time_per_iteration": 2.661961317062378
    },
    {
      "auxiliary_loss_clip": 0.01167251,
      "auxiliary_loss_mlp": 0.01031813,
      "balance_loss_clip": 1.05259132,
      "balance_loss_mlp": 1.02300322,
      "epoch": 0.4368424216918175,
      "flos": 18035200753920.0,
      "grad_norm": 2.950853192757257,
      "language_loss": 0.80730987,
      "learning_rate": 2.500940166280918e-06,
      "loss": 0.8293004,
      "num_input_tokens_seen": 78269470,
      "step": 3633,
      "time_per_iteration": 2.6543943881988525
    },
    {
      "auxiliary_loss_clip": 0.01164677,
      "auxiliary_loss_mlp": 0.01030266,
      "balance_loss_clip": 1.05680776,
      "balance_loss_mlp": 1.02207661,
      "epoch": 0.4369626645824566,
      "flos": 25447127362560.0,
      "grad_norm": 2.075225675639116,
      "language_loss": 0.79224259,
      "learning_rate": 2.500185986892045e-06,
      "loss": 0.81419206,
      "num_input_tokens_seen": 78288955,
      "step": 3634,
      "time_per_iteration": 2.752668857574463
    },
    {
      "auxiliary_loss_clip": 0.01179773,
      "auxiliary_loss_mlp": 0.01030429,
      "balance_loss_clip": 1.05588782,
      "balance_loss_mlp": 1.02161336,
      "epoch": 0.43708290747309564,
      "flos": 25302694775040.0,
      "grad_norm": 2.040170684556536,
      "language_loss": 0.77511126,
      "learning_rate": 2.499431731625355e-06,
      "loss": 0.79721332,
      "num_input_tokens_seen": 78307980,
      "step": 3635,
      "time_per_iteration": 2.6904890537261963
    },
    {
      "auxiliary_loss_clip": 0.0116027,
      "auxiliary_loss_mlp": 0.01030379,
      "balance_loss_clip": 1.05250275,
      "balance_loss_mlp": 1.02209353,
      "epoch": 0.43720315036373475,
      "flos": 31575103344000.0,
      "grad_norm": 1.9465038333013713,
      "language_loss": 0.79368144,
      "learning_rate": 2.4986774005952686e-06,
      "loss": 0.81558788,
      "num_input_tokens_seen": 78330355,
      "step": 3636,
      "time_per_iteration": 2.8612353801727295
    },
    {
      "auxiliary_loss_clip": 0.01193741,
      "auxiliary_loss_mlp": 0.01026664,
      "balance_loss_clip": 1.05783677,
      "balance_loss_mlp": 1.01834977,
      "epoch": 0.43732339325437386,
      "flos": 23112000195840.0,
      "grad_norm": 58.63292100847719,
      "language_loss": 0.85017598,
      "learning_rate": 2.4979229939162166e-06,
      "loss": 0.87238002,
      "num_input_tokens_seen": 78349135,
      "step": 3637,
      "time_per_iteration": 2.6936604976654053
    },
    {
      "auxiliary_loss_clip": 0.01168293,
      "auxiliary_loss_mlp": 0.01029431,
      "balance_loss_clip": 1.05818176,
      "balance_loss_mlp": 1.02092552,
      "epoch": 0.4374436361450129,
      "flos": 27746272080000.0,
      "grad_norm": 1.7118710313291021,
      "language_loss": 0.80392283,
      "learning_rate": 2.4971685117026433e-06,
      "loss": 0.82590008,
      "num_input_tokens_seen": 78368900,
      "step": 3638,
      "time_per_iteration": 2.703301429748535
    },
    {
      "auxiliary_loss_clip": 0.01147493,
      "auxiliary_loss_mlp": 0.01042524,
      "balance_loss_clip": 1.05316889,
      "balance_loss_mlp": 1.03398871,
      "epoch": 0.437563879035652,
      "flos": 24172370616960.0,
      "grad_norm": 4.106309171474949,
      "language_loss": 0.76947135,
      "learning_rate": 2.4964139540690018e-06,
      "loss": 0.79137158,
      "num_input_tokens_seen": 78392235,
      "step": 3639,
      "time_per_iteration": 2.8102266788482666
    },
    {
      "auxiliary_loss_clip": 0.01195555,
      "auxiliary_loss_mlp": 0.01036234,
      "balance_loss_clip": 1.05774713,
      "balance_loss_mlp": 1.02634525,
      "epoch": 0.4376841219262911,
      "flos": 23477211728640.0,
      "grad_norm": 2.647297274965163,
      "language_loss": 0.72883701,
      "learning_rate": 2.495659321129758e-06,
      "loss": 0.7511549,
      "num_input_tokens_seen": 78409980,
      "step": 3640,
      "time_per_iteration": 2.5910847187042236
    },
    {
      "auxiliary_loss_clip": 0.01178867,
      "auxiliary_loss_mlp": 0.01035595,
      "balance_loss_clip": 1.05752861,
      "balance_loss_mlp": 1.02697039,
      "epoch": 0.4378043648169302,
      "flos": 25447809720960.0,
      "grad_norm": 1.9320675149868805,
      "language_loss": 0.7557075,
      "learning_rate": 2.494904612999389e-06,
      "loss": 0.77785212,
      "num_input_tokens_seen": 78428690,
      "step": 3641,
      "time_per_iteration": 2.7018167972564697
    },
    {
      "auxiliary_loss_clip": 0.01073118,
      "auxiliary_loss_mlp": 0.01001765,
      "balance_loss_clip": 1.01381004,
      "balance_loss_mlp": 1.0003525,
      "epoch": 0.4379246077075693,
      "flos": 53914056986880.0,
      "grad_norm": 0.7451475633919763,
      "language_loss": 0.56467837,
      "learning_rate": 2.4941498297923843e-06,
      "loss": 0.58542722,
      "num_input_tokens_seen": 78489260,
      "step": 3642,
      "time_per_iteration": 3.1674351692199707
    },
    {
      "auxiliary_loss_clip": 0.01194515,
      "auxiliary_loss_mlp": 0.01033016,
      "balance_loss_clip": 1.05847812,
      "balance_loss_mlp": 1.02453458,
      "epoch": 0.43804485059820836,
      "flos": 20588305605120.0,
      "grad_norm": 4.221080185327925,
      "language_loss": 0.70137799,
      "learning_rate": 2.4933949716232424e-06,
      "loss": 0.72365326,
      "num_input_tokens_seen": 78506785,
      "step": 3643,
      "time_per_iteration": 2.6686387062072754
    },
    {
      "auxiliary_loss_clip": 0.01193403,
      "auxiliary_loss_mlp": 0.01033424,
      "balance_loss_clip": 1.05763173,
      "balance_loss_mlp": 1.02475142,
      "epoch": 0.43816509348884747,
      "flos": 23876214981120.0,
      "grad_norm": 2.299199473860395,
      "language_loss": 0.73794687,
      "learning_rate": 2.492640038606476e-06,
      "loss": 0.76021516,
      "num_input_tokens_seen": 78525150,
      "step": 3644,
      "time_per_iteration": 2.6281931400299072
    },
    {
      "auxiliary_loss_clip": 0.01178307,
      "auxiliary_loss_mlp": 0.01032961,
      "balance_loss_clip": 1.05539346,
      "balance_loss_mlp": 1.02384722,
      "epoch": 0.4382853363794866,
      "flos": 14684448533760.0,
      "grad_norm": 2.2667794202253404,
      "language_loss": 0.78782952,
      "learning_rate": 2.491885030856608e-06,
      "loss": 0.80994213,
      "num_input_tokens_seen": 78543245,
      "step": 3645,
      "time_per_iteration": 2.672637701034546
    },
    {
      "auxiliary_loss_clip": 0.01181149,
      "auxiliary_loss_mlp": 0.0102783,
      "balance_loss_clip": 1.05851817,
      "balance_loss_mlp": 1.01835918,
      "epoch": 0.43840557927012563,
      "flos": 17165301177600.0,
      "grad_norm": 2.8263765994432672,
      "language_loss": 0.82964504,
      "learning_rate": 2.4911299484881713e-06,
      "loss": 0.85173482,
      "num_input_tokens_seen": 78560775,
      "step": 3646,
      "time_per_iteration": 2.7696688175201416
    },
    {
      "auxiliary_loss_clip": 0.01191971,
      "auxiliary_loss_mlp": 0.01031561,
      "balance_loss_clip": 1.05677199,
      "balance_loss_mlp": 1.02275777,
      "epoch": 0.43852582216076474,
      "flos": 19390685316480.0,
      "grad_norm": 2.478093757786394,
      "language_loss": 0.8126868,
      "learning_rate": 2.490374791615712e-06,
      "loss": 0.83492213,
      "num_input_tokens_seen": 78580800,
      "step": 3647,
      "time_per_iteration": 2.647642135620117
    },
    {
      "auxiliary_loss_clip": 0.01160336,
      "auxiliary_loss_mlp": 0.01036743,
      "balance_loss_clip": 1.05477595,
      "balance_loss_mlp": 1.02764702,
      "epoch": 0.43864606505140386,
      "flos": 18075133699200.0,
      "grad_norm": 4.345203664708571,
      "language_loss": 0.78180355,
      "learning_rate": 2.4896195603537867e-06,
      "loss": 0.80377436,
      "num_input_tokens_seen": 78595410,
      "step": 3648,
      "time_per_iteration": 3.627794027328491
    },
    {
      "auxiliary_loss_clip": 0.0116915,
      "auxiliary_loss_mlp": 0.01028506,
      "balance_loss_clip": 1.05550003,
      "balance_loss_mlp": 1.01912987,
      "epoch": 0.4387663079420429,
      "flos": 19644896845440.0,
      "grad_norm": 2.494563334443331,
      "language_loss": 0.73618245,
      "learning_rate": 2.488864254816964e-06,
      "loss": 0.75815898,
      "num_input_tokens_seen": 78614100,
      "step": 3649,
      "time_per_iteration": 3.6786601543426514
    },
    {
      "auxiliary_loss_clip": 0.01197332,
      "auxiliary_loss_mlp": 0.01040361,
      "balance_loss_clip": 1.05849695,
      "balance_loss_mlp": 1.03106225,
      "epoch": 0.438886550832682,
      "flos": 19719339782400.0,
      "grad_norm": 2.4425115505923194,
      "language_loss": 0.68082756,
      "learning_rate": 2.4881088751198218e-06,
      "loss": 0.70320451,
      "num_input_tokens_seen": 78632260,
      "step": 3650,
      "time_per_iteration": 2.613889694213867
    },
    {
      "auxiliary_loss_clip": 0.01181365,
      "auxiliary_loss_mlp": 0.01035413,
      "balance_loss_clip": 1.05673051,
      "balance_loss_mlp": 1.0258348,
      "epoch": 0.43900679372332113,
      "flos": 14536675981440.0,
      "grad_norm": 2.586761165026747,
      "language_loss": 0.64628732,
      "learning_rate": 2.4873534213769517e-06,
      "loss": 0.66845518,
      "num_input_tokens_seen": 78647490,
      "step": 3651,
      "time_per_iteration": 2.7063283920288086
    },
    {
      "auxiliary_loss_clip": 0.01136829,
      "auxiliary_loss_mlp": 0.01033321,
      "balance_loss_clip": 1.05431116,
      "balance_loss_mlp": 1.02495813,
      "epoch": 0.4391270366139602,
      "flos": 24056234968320.0,
      "grad_norm": 1.9415916005336673,
      "language_loss": 0.72342765,
      "learning_rate": 2.4865978937029547e-06,
      "loss": 0.74512911,
      "num_input_tokens_seen": 78666470,
      "step": 3652,
      "time_per_iteration": 3.686931848526001
    },
    {
      "auxiliary_loss_clip": 0.01194843,
      "auxiliary_loss_mlp": 0.01034123,
      "balance_loss_clip": 1.05895758,
      "balance_loss_mlp": 1.02504539,
      "epoch": 0.4392472795045993,
      "flos": 31538510363520.0,
      "grad_norm": 1.8624046222450117,
      "language_loss": 0.66824579,
      "learning_rate": 2.485842292212445e-06,
      "loss": 0.69053549,
      "num_input_tokens_seen": 78687685,
      "step": 3653,
      "time_per_iteration": 2.7762389183044434
    },
    {
      "auxiliary_loss_clip": 0.01184697,
      "auxiliary_loss_mlp": 0.01038554,
      "balance_loss_clip": 1.05688119,
      "balance_loss_mlp": 1.02949429,
      "epoch": 0.4393675223952384,
      "flos": 14866300114560.0,
      "grad_norm": 2.821098928435715,
      "language_loss": 0.80592275,
      "learning_rate": 2.485086617020045e-06,
      "loss": 0.82815534,
      "num_input_tokens_seen": 78706180,
      "step": 3654,
      "time_per_iteration": 2.6397340297698975
    },
    {
      "auxiliary_loss_clip": 0.01197452,
      "auxiliary_loss_mlp": 0.01031985,
      "balance_loss_clip": 1.05954337,
      "balance_loss_mlp": 1.02246583,
      "epoch": 0.43948776528587746,
      "flos": 14825900292480.0,
      "grad_norm": 9.964456273753656,
      "language_loss": 0.8209523,
      "learning_rate": 2.4843308682403903e-06,
      "loss": 0.8432467,
      "num_input_tokens_seen": 78723095,
      "step": 3655,
      "time_per_iteration": 2.6027588844299316
    },
    {
      "auxiliary_loss_clip": 0.01192119,
      "auxiliary_loss_mlp": 0.01029771,
      "balance_loss_clip": 1.0574317,
      "balance_loss_mlp": 1.02177775,
      "epoch": 0.4396080081765166,
      "flos": 13914523486080.0,
      "grad_norm": 2.490516299471282,
      "language_loss": 0.8317067,
      "learning_rate": 2.4835750459881294e-06,
      "loss": 0.85392559,
      "num_input_tokens_seen": 78739720,
      "step": 3656,
      "time_per_iteration": 2.587897300720215
    },
    {
      "auxiliary_loss_clip": 0.01183069,
      "auxiliary_loss_mlp": 0.0103008,
      "balance_loss_clip": 1.05589759,
      "balance_loss_mlp": 1.02115107,
      "epoch": 0.43972825106715563,
      "flos": 18222978078720.0,
      "grad_norm": 2.8391338328278346,
      "language_loss": 0.8211208,
      "learning_rate": 2.4828191503779177e-06,
      "loss": 0.8432523,
      "num_input_tokens_seen": 78757820,
      "step": 3657,
      "time_per_iteration": 3.5652570724487305
    },
    {
      "auxiliary_loss_clip": 0.01180495,
      "auxiliary_loss_mlp": 0.01033916,
      "balance_loss_clip": 1.05739021,
      "balance_loss_mlp": 1.02451634,
      "epoch": 0.43984849395779474,
      "flos": 16873239692160.0,
      "grad_norm": 2.1252283912745895,
      "language_loss": 0.90362072,
      "learning_rate": 2.482063181524425e-06,
      "loss": 0.9257648,
      "num_input_tokens_seen": 78773720,
      "step": 3658,
      "time_per_iteration": 2.648360013961792
    },
    {
      "auxiliary_loss_clip": 0.01200907,
      "auxiliary_loss_mlp": 0.01037824,
      "balance_loss_clip": 1.06098473,
      "balance_loss_mlp": 1.02729154,
      "epoch": 0.43996873684843385,
      "flos": 18691504104960.0,
      "grad_norm": 2.108211773119252,
      "language_loss": 0.81242859,
      "learning_rate": 2.4813071395423307e-06,
      "loss": 0.83481586,
      "num_input_tokens_seen": 78791285,
      "step": 3659,
      "time_per_iteration": 2.578906536102295
    },
    {
      "auxiliary_loss_clip": 0.01193914,
      "auxiliary_loss_mlp": 0.01028296,
      "balance_loss_clip": 1.05711341,
      "balance_loss_mlp": 1.01962388,
      "epoch": 0.4400889797390729,
      "flos": 23653460787840.0,
      "grad_norm": 2.3166595604338096,
      "language_loss": 0.64236927,
      "learning_rate": 2.4805510245463263e-06,
      "loss": 0.66459143,
      "num_input_tokens_seen": 78811440,
      "step": 3660,
      "time_per_iteration": 2.705359935760498
    },
    {
      "auxiliary_loss_clip": 0.01175078,
      "auxiliary_loss_mlp": 0.01038854,
      "balance_loss_clip": 1.05266023,
      "balance_loss_mlp": 1.03014624,
      "epoch": 0.440209222629712,
      "flos": 23149203707520.0,
      "grad_norm": 2.0627155113867284,
      "language_loss": 0.60642135,
      "learning_rate": 2.4797948366511137e-06,
      "loss": 0.62856066,
      "num_input_tokens_seen": 78831150,
      "step": 3661,
      "time_per_iteration": 2.6585309505462646
    },
    {
      "auxiliary_loss_clip": 0.01195724,
      "auxiliary_loss_mlp": 0.01041304,
      "balance_loss_clip": 1.05750251,
      "balance_loss_mlp": 1.03245282,
      "epoch": 0.4403294655203511,
      "flos": 24823394668800.0,
      "grad_norm": 2.3900600855798575,
      "language_loss": 0.76533931,
      "learning_rate": 2.4790385759714055e-06,
      "loss": 0.78770959,
      "num_input_tokens_seen": 78850215,
      "step": 3662,
      "time_per_iteration": 2.653925895690918
    },
    {
      "auxiliary_loss_clip": 0.01194623,
      "auxiliary_loss_mlp": 0.01037129,
      "balance_loss_clip": 1.05884707,
      "balance_loss_mlp": 1.02818847,
      "epoch": 0.4404497084109902,
      "flos": 22565080736640.0,
      "grad_norm": 1.7982434322156056,
      "language_loss": 0.71790242,
      "learning_rate": 2.478282242621926e-06,
      "loss": 0.74021995,
      "num_input_tokens_seen": 78870675,
      "step": 3663,
      "time_per_iteration": 2.6195309162139893
    },
    {
      "auxiliary_loss_clip": 0.01083544,
      "auxiliary_loss_mlp": 0.01002186,
      "balance_loss_clip": 1.01558912,
      "balance_loss_mlp": 1.00073767,
      "epoch": 0.4405699513016293,
      "flos": 64967073448320.0,
      "grad_norm": 0.9691384313449573,
      "language_loss": 0.59473538,
      "learning_rate": 2.477525836717411e-06,
      "loss": 0.6155926,
      "num_input_tokens_seen": 78938440,
      "step": 3664,
      "time_per_iteration": 3.2926371097564697
    },
    {
      "auxiliary_loss_clip": 0.01195393,
      "auxiliary_loss_mlp": 0.01035554,
      "balance_loss_clip": 1.05797911,
      "balance_loss_mlp": 1.02614284,
      "epoch": 0.4406901941922684,
      "flos": 35661952978560.0,
      "grad_norm": 3.150276572789096,
      "language_loss": 0.79537892,
      "learning_rate": 2.476769358372606e-06,
      "loss": 0.81768835,
      "num_input_tokens_seen": 78960090,
      "step": 3665,
      "time_per_iteration": 2.7470710277557373
    },
    {
      "auxiliary_loss_clip": 0.01193982,
      "auxiliary_loss_mlp": 0.01026972,
      "balance_loss_clip": 1.05993629,
      "balance_loss_mlp": 1.01833558,
      "epoch": 0.44081043708290746,
      "flos": 18040767361920.0,
      "grad_norm": 2.1237770984200486,
      "language_loss": 0.75085217,
      "learning_rate": 2.4760128077022683e-06,
      "loss": 0.77306175,
      "num_input_tokens_seen": 78978225,
      "step": 3666,
      "time_per_iteration": 2.6171891689300537
    },
    {
      "auxiliary_loss_clip": 0.01153645,
      "auxiliary_loss_mlp": 0.01028943,
      "balance_loss_clip": 1.05220628,
      "balance_loss_mlp": 1.0204556,
      "epoch": 0.44093067997354657,
      "flos": 30153507799680.0,
      "grad_norm": 1.6001093029977655,
      "language_loss": 0.6877594,
      "learning_rate": 2.4752561848211672e-06,
      "loss": 0.70958531,
      "num_input_tokens_seen": 79000625,
      "step": 3667,
      "time_per_iteration": 2.7753236293792725
    },
    {
      "auxiliary_loss_clip": 0.01160716,
      "auxiliary_loss_mlp": 0.01034856,
      "balance_loss_clip": 1.0551331,
      "balance_loss_mlp": 1.02601039,
      "epoch": 0.4410509228641857,
      "flos": 23255068066560.0,
      "grad_norm": 2.2496891136279884,
      "language_loss": 0.71625459,
      "learning_rate": 2.4744994898440797e-06,
      "loss": 0.73821026,
      "num_input_tokens_seen": 79019415,
      "step": 3668,
      "time_per_iteration": 2.7335948944091797
    },
    {
      "auxiliary_loss_clip": 0.01180498,
      "auxiliary_loss_mlp": 0.01029988,
      "balance_loss_clip": 1.05630279,
      "balance_loss_mlp": 1.02091002,
      "epoch": 0.44117116575482473,
      "flos": 19500571998720.0,
      "grad_norm": 1.9790206288268664,
      "language_loss": 0.8360405,
      "learning_rate": 2.473742722885797e-06,
      "loss": 0.85814536,
      "num_input_tokens_seen": 79038435,
      "step": 3669,
      "time_per_iteration": 2.6703174114227295
    },
    {
      "auxiliary_loss_clip": 0.01199082,
      "auxiliary_loss_mlp": 0.00762873,
      "balance_loss_clip": 1.06145692,
      "balance_loss_mlp": 1.00100613,
      "epoch": 0.44129140864546385,
      "flos": 27053124353280.0,
      "grad_norm": 2.019862264787062,
      "language_loss": 0.65151048,
      "learning_rate": 2.4729858840611197e-06,
      "loss": 0.67113006,
      "num_input_tokens_seen": 79057345,
      "step": 3670,
      "time_per_iteration": 2.623828649520874
    },
    {
      "auxiliary_loss_clip": 0.01135525,
      "auxiliary_loss_mlp": 0.01031939,
      "balance_loss_clip": 1.04747486,
      "balance_loss_mlp": 1.02339125,
      "epoch": 0.4414116515361029,
      "flos": 26102101910400.0,
      "grad_norm": 2.9464354437407043,
      "language_loss": 0.72924674,
      "learning_rate": 2.4722289734848605e-06,
      "loss": 0.75092137,
      "num_input_tokens_seen": 79077810,
      "step": 3671,
      "time_per_iteration": 2.8393523693084717
    },
    {
      "auxiliary_loss_clip": 0.01166544,
      "auxiliary_loss_mlp": 0.01026911,
      "balance_loss_clip": 1.05435038,
      "balance_loss_mlp": 1.01777959,
      "epoch": 0.441531894426742,
      "flos": 21906083865600.0,
      "grad_norm": 2.7965402357964435,
      "language_loss": 0.78264594,
      "learning_rate": 2.471471991271841e-06,
      "loss": 0.80458045,
      "num_input_tokens_seen": 79094935,
      "step": 3672,
      "time_per_iteration": 2.783482313156128
    },
    {
      "auxiliary_loss_clip": 0.01194934,
      "auxiliary_loss_mlp": 0.0103542,
      "balance_loss_clip": 1.05754137,
      "balance_loss_mlp": 1.02668774,
      "epoch": 0.4416521373173811,
      "flos": 23437099215360.0,
      "grad_norm": 2.4486609762616136,
      "language_loss": 0.79460788,
      "learning_rate": 2.470714937536896e-06,
      "loss": 0.81691146,
      "num_input_tokens_seen": 79113660,
      "step": 3673,
      "time_per_iteration": 2.684316873550415
    },
    {
      "auxiliary_loss_clip": 0.01195935,
      "auxiliary_loss_mlp": 0.01033529,
      "balance_loss_clip": 1.05761755,
      "balance_loss_mlp": 1.02433181,
      "epoch": 0.4417723802080202,
      "flos": 20334345471360.0,
      "grad_norm": 2.012612992630685,
      "language_loss": 0.71072924,
      "learning_rate": 2.469957812394868e-06,
      "loss": 0.73302388,
      "num_input_tokens_seen": 79132470,
      "step": 3674,
      "time_per_iteration": 2.5924651622772217
    },
    {
      "auxiliary_loss_clip": 0.01185303,
      "auxiliary_loss_mlp": 0.01034289,
      "balance_loss_clip": 1.05942404,
      "balance_loss_mlp": 1.0248301,
      "epoch": 0.4418926230986593,
      "flos": 18880682060160.0,
      "grad_norm": 2.0055803115782496,
      "language_loss": 0.76735115,
      "learning_rate": 2.4692006159606148e-06,
      "loss": 0.78954709,
      "num_input_tokens_seen": 79150000,
      "step": 3675,
      "time_per_iteration": 3.588818311691284
    },
    {
      "auxiliary_loss_clip": 0.01180518,
      "auxiliary_loss_mlp": 0.01037754,
      "balance_loss_clip": 1.05752373,
      "balance_loss_mlp": 1.02875972,
      "epoch": 0.4420128659892984,
      "flos": 19464409981440.0,
      "grad_norm": 1.7573877468227919,
      "language_loss": 0.78769767,
      "learning_rate": 2.468443348349e-06,
      "loss": 0.80988044,
      "num_input_tokens_seen": 79167875,
      "step": 3676,
      "time_per_iteration": 2.7744200229644775
    },
    {
      "auxiliary_loss_clip": 0.01160103,
      "auxiliary_loss_mlp": 0.01030967,
      "balance_loss_clip": 1.05397284,
      "balance_loss_mlp": 1.02179384,
      "epoch": 0.44213310887993745,
      "flos": 17894359526400.0,
      "grad_norm": 2.312776390603567,
      "language_loss": 0.82739258,
      "learning_rate": 2.467686009674902e-06,
      "loss": 0.84930325,
      "num_input_tokens_seen": 79182325,
      "step": 3677,
      "time_per_iteration": 2.6146395206451416
    },
    {
      "auxiliary_loss_clip": 0.01186178,
      "auxiliary_loss_mlp": 0.01034842,
      "balance_loss_clip": 1.05790448,
      "balance_loss_mlp": 1.02550745,
      "epoch": 0.44225335177057656,
      "flos": 19204667758080.0,
      "grad_norm": 2.5246034630604406,
      "language_loss": 0.85167009,
      "learning_rate": 2.466928600053209e-06,
      "loss": 0.87388027,
      "num_input_tokens_seen": 79197630,
      "step": 3678,
      "time_per_iteration": 3.565932512283325
    },
    {
      "auxiliary_loss_clip": 0.01168187,
      "auxiliary_loss_mlp": 0.01033763,
      "balance_loss_clip": 1.05459058,
      "balance_loss_mlp": 1.02550209,
      "epoch": 0.4423735946612157,
      "flos": 23471321898240.0,
      "grad_norm": 2.037031882183407,
      "language_loss": 0.71712768,
      "learning_rate": 2.466171119598818e-06,
      "loss": 0.73914719,
      "num_input_tokens_seen": 79217600,
      "step": 3679,
      "time_per_iteration": 2.7154626846313477
    },
    {
      "auxiliary_loss_clip": 0.01198522,
      "auxiliary_loss_mlp": 0.01038274,
      "balance_loss_clip": 1.05803537,
      "balance_loss_mlp": 1.0285821,
      "epoch": 0.44249383755185473,
      "flos": 26685398868480.0,
      "grad_norm": 2.017522017320078,
      "language_loss": 0.7757802,
      "learning_rate": 2.465413568426639e-06,
      "loss": 0.79814816,
      "num_input_tokens_seen": 79238550,
      "step": 3680,
      "time_per_iteration": 2.6785788536071777
    },
    {
      "auxiliary_loss_clip": 0.01160019,
      "auxiliary_loss_mlp": 0.01029633,
      "balance_loss_clip": 1.05037439,
      "balance_loss_mlp": 1.02151513,
      "epoch": 0.44261408044249384,
      "flos": 23147659422720.0,
      "grad_norm": 1.6342175597586657,
      "language_loss": 0.81477493,
      "learning_rate": 2.464655946651591e-06,
      "loss": 0.83667147,
      "num_input_tokens_seen": 79257555,
      "step": 3681,
      "time_per_iteration": 2.7565877437591553
    },
    {
      "auxiliary_loss_clip": 0.01183171,
      "auxiliary_loss_mlp": 0.01040038,
      "balance_loss_clip": 1.05693412,
      "balance_loss_mlp": 1.03123403,
      "epoch": 0.44273432333313295,
      "flos": 24462564595200.0,
      "grad_norm": 4.487167597051458,
      "language_loss": 0.80992019,
      "learning_rate": 2.4638982543886065e-06,
      "loss": 0.83215231,
      "num_input_tokens_seen": 79277595,
      "step": 3682,
      "time_per_iteration": 2.7012293338775635
    },
    {
      "auxiliary_loss_clip": 0.01185735,
      "auxiliary_loss_mlp": 0.01038523,
      "balance_loss_clip": 1.05762172,
      "balance_loss_mlp": 1.02924824,
      "epoch": 0.442854566223772,
      "flos": 17528932512000.0,
      "grad_norm": 9.492988313844254,
      "language_loss": 0.87725854,
      "learning_rate": 2.4631404917526254e-06,
      "loss": 0.89950109,
      "num_input_tokens_seen": 79294550,
      "step": 3683,
      "time_per_iteration": 3.5141336917877197
    },
    {
      "auxiliary_loss_clip": 0.01191678,
      "auxiliary_loss_mlp": 0.01031292,
      "balance_loss_clip": 1.05613446,
      "balance_loss_mlp": 1.02272093,
      "epoch": 0.4429748091144111,
      "flos": 24896293320960.0,
      "grad_norm": 1.9437479088074958,
      "language_loss": 0.79388475,
      "learning_rate": 2.4623826588586e-06,
      "loss": 0.81611443,
      "num_input_tokens_seen": 79314820,
      "step": 3684,
      "time_per_iteration": 2.613090991973877
    },
    {
      "auxiliary_loss_clip": 0.01164902,
      "auxiliary_loss_mlp": 0.01029468,
      "balance_loss_clip": 1.05472934,
      "balance_loss_mlp": 1.0200386,
      "epoch": 0.4430950520050502,
      "flos": 21614704738560.0,
      "grad_norm": 1.6384550676637228,
      "language_loss": 0.82941145,
      "learning_rate": 2.461624755821492e-06,
      "loss": 0.8513552,
      "num_input_tokens_seen": 79334300,
      "step": 3685,
      "time_per_iteration": 2.699206590652466
    },
    {
      "auxiliary_loss_clip": 0.01167021,
      "auxiliary_loss_mlp": 0.01034393,
      "balance_loss_clip": 1.0544517,
      "balance_loss_mlp": 1.02512503,
      "epoch": 0.4432152948956893,
      "flos": 24572271709440.0,
      "grad_norm": 2.620467610751897,
      "language_loss": 0.76910818,
      "learning_rate": 2.4608667827562763e-06,
      "loss": 0.79112232,
      "num_input_tokens_seen": 79353630,
      "step": 3686,
      "time_per_iteration": 2.6695122718811035
    },
    {
      "auxiliary_loss_clip": 0.01199053,
      "auxiliary_loss_mlp": 0.0104185,
      "balance_loss_clip": 1.05921042,
      "balance_loss_mlp": 1.03252745,
      "epoch": 0.4433355377863284,
      "flos": 21762261809280.0,
      "grad_norm": 2.4806799938940114,
      "language_loss": 0.90096724,
      "learning_rate": 2.460108739777936e-06,
      "loss": 0.9233762,
      "num_input_tokens_seen": 79372765,
      "step": 3687,
      "time_per_iteration": 2.646662473678589
    },
    {
      "auxiliary_loss_clip": 0.01193562,
      "auxiliary_loss_mlp": 0.01031617,
      "balance_loss_clip": 1.05780113,
      "balance_loss_mlp": 1.02278948,
      "epoch": 0.44345578067696745,
      "flos": 20084479488000.0,
      "grad_norm": 1.5840270848608382,
      "language_loss": 0.76561236,
      "learning_rate": 2.4593506270014656e-06,
      "loss": 0.78786421,
      "num_input_tokens_seen": 79391735,
      "step": 3688,
      "time_per_iteration": 2.5979466438293457
    },
    {
      "auxiliary_loss_clip": 0.01179012,
      "auxiliary_loss_mlp": 0.01032441,
      "balance_loss_clip": 1.0562942,
      "balance_loss_mlp": 1.02275515,
      "epoch": 0.44357602356760656,
      "flos": 24169497528960.0,
      "grad_norm": 1.5287839444875784,
      "language_loss": 0.8232463,
      "learning_rate": 2.45859244454187e-06,
      "loss": 0.84536088,
      "num_input_tokens_seen": 79411525,
      "step": 3689,
      "time_per_iteration": 2.7061655521392822
    },
    {
      "auxiliary_loss_clip": 0.01180126,
      "auxiliary_loss_mlp": 0.01032898,
      "balance_loss_clip": 1.05791712,
      "balance_loss_mlp": 1.02367735,
      "epoch": 0.44369626645824567,
      "flos": 22707717644160.0,
      "grad_norm": 1.850315704628922,
      "language_loss": 0.66776133,
      "learning_rate": 2.4578341925141655e-06,
      "loss": 0.68989158,
      "num_input_tokens_seen": 79430740,
      "step": 3690,
      "time_per_iteration": 2.674076557159424
    },
    {
      "auxiliary_loss_clip": 0.0116567,
      "auxiliary_loss_mlp": 0.01033193,
      "balance_loss_clip": 1.0565722,
      "balance_loss_mlp": 1.02319753,
      "epoch": 0.4438165093488847,
      "flos": 38030225420160.0,
      "grad_norm": 2.2656740541166807,
      "language_loss": 0.72358733,
      "learning_rate": 2.457075871033378e-06,
      "loss": 0.7455759,
      "num_input_tokens_seen": 79452615,
      "step": 3691,
      "time_per_iteration": 2.8043699264526367
    },
    {
      "auxiliary_loss_clip": 0.01172856,
      "auxiliary_loss_mlp": 0.01033841,
      "balance_loss_clip": 1.05263197,
      "balance_loss_mlp": 1.0241971,
      "epoch": 0.44393675223952384,
      "flos": 15523213996800.0,
      "grad_norm": 2.3045096466454065,
      "language_loss": 0.88812256,
      "learning_rate": 2.4563174802145445e-06,
      "loss": 0.91018951,
      "num_input_tokens_seen": 79469865,
      "step": 3692,
      "time_per_iteration": 2.7054386138916016
    },
    {
      "auxiliary_loss_clip": 0.01058707,
      "auxiliary_loss_mlp": 0.00753444,
      "balance_loss_clip": 1.01479602,
      "balance_loss_mlp": 1.00016809,
      "epoch": 0.44405699513016295,
      "flos": 64574893779840.0,
      "grad_norm": 0.6315417224230995,
      "language_loss": 0.48479038,
      "learning_rate": 2.455559020172712e-06,
      "loss": 0.50291193,
      "num_input_tokens_seen": 79537220,
      "step": 3693,
      "time_per_iteration": 3.335294723510742
    },
    {
      "auxiliary_loss_clip": 0.01190558,
      "auxiliary_loss_mlp": 0.01032064,
      "balance_loss_clip": 1.0638541,
      "balance_loss_mlp": 1.02306986,
      "epoch": 0.444177238020802,
      "flos": 23987394552960.0,
      "grad_norm": 2.3619288304783645,
      "language_loss": 0.89835596,
      "learning_rate": 2.4548004910229385e-06,
      "loss": 0.92058218,
      "num_input_tokens_seen": 79554795,
      "step": 3694,
      "time_per_iteration": 2.8175454139709473
    },
    {
      "auxiliary_loss_clip": 0.01151962,
      "auxiliary_loss_mlp": 0.0103021,
      "balance_loss_clip": 1.05148101,
      "balance_loss_mlp": 1.02134657,
      "epoch": 0.4442974809114411,
      "flos": 22563069575040.0,
      "grad_norm": 3.2971318800077265,
      "language_loss": 0.87173909,
      "learning_rate": 2.4540418928802913e-06,
      "loss": 0.89356077,
      "num_input_tokens_seen": 79573530,
      "step": 3695,
      "time_per_iteration": 2.724087953567505
    },
    {
      "auxiliary_loss_clip": 0.01162346,
      "auxiliary_loss_mlp": 0.01032549,
      "balance_loss_clip": 1.05297005,
      "balance_loss_mlp": 1.02295828,
      "epoch": 0.4444177238020802,
      "flos": 17675699483520.0,
      "grad_norm": 2.359959934060551,
      "language_loss": 0.66428638,
      "learning_rate": 2.4532832258598506e-06,
      "loss": 0.68623537,
      "num_input_tokens_seen": 79591360,
      "step": 3696,
      "time_per_iteration": 2.7054779529571533
    },
    {
      "auxiliary_loss_clip": 0.01193591,
      "auxiliary_loss_mlp": 0.01028008,
      "balance_loss_clip": 1.05857968,
      "balance_loss_mlp": 1.01894832,
      "epoch": 0.4445379666927193,
      "flos": 28621594609920.0,
      "grad_norm": 1.9800998474990512,
      "language_loss": 0.80841249,
      "learning_rate": 2.4525244900767047e-06,
      "loss": 0.83062851,
      "num_input_tokens_seen": 79612175,
      "step": 3697,
      "time_per_iteration": 2.7186367511749268
    },
    {
      "auxiliary_loss_clip": 0.01083467,
      "auxiliary_loss_mlp": 0.01007208,
      "balance_loss_clip": 1.01586688,
      "balance_loss_mlp": 1.00573564,
      "epoch": 0.4446582095833584,
      "flos": 70487370115200.0,
      "grad_norm": 0.7705486004756179,
      "language_loss": 0.60510349,
      "learning_rate": 2.4517656856459536e-06,
      "loss": 0.62601024,
      "num_input_tokens_seen": 79678020,
      "step": 3698,
      "time_per_iteration": 3.3015434741973877
    },
    {
      "auxiliary_loss_clip": 0.01190613,
      "auxiliary_loss_mlp": 0.01032007,
      "balance_loss_clip": 1.0552361,
      "balance_loss_mlp": 1.02317905,
      "epoch": 0.4447784524739975,
      "flos": 26505199313280.0,
      "grad_norm": 2.232655883634345,
      "language_loss": 0.68002552,
      "learning_rate": 2.4510068126827073e-06,
      "loss": 0.70225167,
      "num_input_tokens_seen": 79699020,
      "step": 3699,
      "time_per_iteration": 2.6184682846069336
    },
    {
      "auxiliary_loss_clip": 0.01196273,
      "auxiliary_loss_mlp": 0.01031338,
      "balance_loss_clip": 1.0603652,
      "balance_loss_mlp": 1.02232623,
      "epoch": 0.44489869536463655,
      "flos": 11656209553920.0,
      "grad_norm": 2.5267131129892024,
      "language_loss": 0.82581663,
      "learning_rate": 2.450247871302086e-06,
      "loss": 0.84809273,
      "num_input_tokens_seen": 79716795,
      "step": 3700,
      "time_per_iteration": 2.5996975898742676
    },
    {
      "auxiliary_loss_clip": 0.01192997,
      "auxiliary_loss_mlp": 0.01028173,
      "balance_loss_clip": 1.05745971,
      "balance_loss_mlp": 1.02018034,
      "epoch": 0.44501893825527566,
      "flos": 20448469958400.0,
      "grad_norm": 6.128928814114545,
      "language_loss": 0.836061,
      "learning_rate": 2.44948886161922e-06,
      "loss": 0.85827267,
      "num_input_tokens_seen": 79735810,
      "step": 3701,
      "time_per_iteration": 4.462112903594971
    },
    {
      "auxiliary_loss_clip": 0.01194303,
      "auxiliary_loss_mlp": 0.01033593,
      "balance_loss_clip": 1.05979776,
      "balance_loss_mlp": 1.02453887,
      "epoch": 0.4451391811459148,
      "flos": 18261079430400.0,
      "grad_norm": 1.9882554542220416,
      "language_loss": 0.85261202,
      "learning_rate": 2.4487297837492524e-06,
      "loss": 0.87489098,
      "num_input_tokens_seen": 79754975,
      "step": 3702,
      "time_per_iteration": 2.610018253326416
    },
    {
      "auxiliary_loss_clip": 0.01196316,
      "auxiliary_loss_mlp": 0.01027082,
      "balance_loss_clip": 1.05869222,
      "balance_loss_mlp": 1.01841497,
      "epoch": 0.44525942403655383,
      "flos": 16910155895040.0,
      "grad_norm": 1.918439550498638,
      "language_loss": 0.62330449,
      "learning_rate": 2.4479706378073323e-06,
      "loss": 0.64553845,
      "num_input_tokens_seen": 79773515,
      "step": 3703,
      "time_per_iteration": 2.6275312900543213
    },
    {
      "auxiliary_loss_clip": 0.01190824,
      "auxiliary_loss_mlp": 0.0103191,
      "balance_loss_clip": 1.05710649,
      "balance_loss_mlp": 1.02352977,
      "epoch": 0.44537966692719294,
      "flos": 23258838994560.0,
      "grad_norm": 1.5014501493461727,
      "language_loss": 0.83964574,
      "learning_rate": 2.447211423908623e-06,
      "loss": 0.86187309,
      "num_input_tokens_seen": 79793560,
      "step": 3704,
      "time_per_iteration": 3.569270610809326
    },
    {
      "auxiliary_loss_clip": 0.0115538,
      "auxiliary_loss_mlp": 0.01031797,
      "balance_loss_clip": 1.05284095,
      "balance_loss_mlp": 1.02327979,
      "epoch": 0.445499909817832,
      "flos": 21724160457600.0,
      "grad_norm": 2.020519898801277,
      "language_loss": 0.7513504,
      "learning_rate": 2.4464521421682966e-06,
      "loss": 0.77322221,
      "num_input_tokens_seen": 79811150,
      "step": 3705,
      "time_per_iteration": 2.6736414432525635
    },
    {
      "auxiliary_loss_clip": 0.0119154,
      "auxiliary_loss_mlp": 0.01026587,
      "balance_loss_clip": 1.05824709,
      "balance_loss_mlp": 1.0185287,
      "epoch": 0.4456201527084711,
      "flos": 23987969170560.0,
      "grad_norm": 1.4040249115663719,
      "language_loss": 0.87958497,
      "learning_rate": 2.4456927927015345e-06,
      "loss": 0.9017663,
      "num_input_tokens_seen": 79832190,
      "step": 3706,
      "time_per_iteration": 2.671278476715088
    },
    {
      "auxiliary_loss_clip": 0.01171969,
      "auxiliary_loss_mlp": 0.01037891,
      "balance_loss_clip": 1.05973744,
      "balance_loss_mlp": 1.02802634,
      "epoch": 0.4457403955991102,
      "flos": 18807065136000.0,
      "grad_norm": 2.3795208693944248,
      "language_loss": 0.76541644,
      "learning_rate": 2.4449333756235307e-06,
      "loss": 0.78751504,
      "num_input_tokens_seen": 79848905,
      "step": 3707,
      "time_per_iteration": 2.6760141849517822
    },
    {
      "auxiliary_loss_clip": 0.0119829,
      "auxiliary_loss_mlp": 0.01034111,
      "balance_loss_clip": 1.0619086,
      "balance_loss_mlp": 1.02505112,
      "epoch": 0.4458606384897493,
      "flos": 19207756327680.0,
      "grad_norm": 6.082089344879322,
      "language_loss": 0.78768611,
      "learning_rate": 2.4441738910494876e-06,
      "loss": 0.81001014,
      "num_input_tokens_seen": 79863640,
      "step": 3708,
      "time_per_iteration": 2.6615657806396484
    },
    {
      "auxiliary_loss_clip": 0.01174886,
      "auxiliary_loss_mlp": 0.01030492,
      "balance_loss_clip": 1.05589747,
      "balance_loss_mlp": 1.02105057,
      "epoch": 0.4459808813803884,
      "flos": 21361283308800.0,
      "grad_norm": 1.851566602134364,
      "language_loss": 0.81941462,
      "learning_rate": 2.4434143390946176e-06,
      "loss": 0.84146839,
      "num_input_tokens_seen": 79882450,
      "step": 3709,
      "time_per_iteration": 2.662964105606079
    },
    {
      "auxiliary_loss_clip": 0.01192649,
      "auxiliary_loss_mlp": 0.01033894,
      "balance_loss_clip": 1.05837882,
      "balance_loss_mlp": 1.02477479,
      "epoch": 0.4461011242710275,
      "flos": 23288967527040.0,
      "grad_norm": 2.1252131053841383,
      "language_loss": 0.85846758,
      "learning_rate": 2.4426547198741457e-06,
      "loss": 0.88073301,
      "num_input_tokens_seen": 79900655,
      "step": 3710,
      "time_per_iteration": 3.5577151775360107
    },
    {
      "auxiliary_loss_clip": 0.01169341,
      "auxiliary_loss_mlp": 0.01033441,
      "balance_loss_clip": 1.05799055,
      "balance_loss_mlp": 1.02403557,
      "epoch": 0.44622136716166655,
      "flos": 20193001453440.0,
      "grad_norm": 3.2358962979373165,
      "language_loss": 0.74514651,
      "learning_rate": 2.441895033503305e-06,
      "loss": 0.76717436,
      "num_input_tokens_seen": 79918575,
      "step": 3711,
      "time_per_iteration": 2.74237060546875
    },
    {
      "auxiliary_loss_clip": 0.01196326,
      "auxiliary_loss_mlp": 0.01031207,
      "balance_loss_clip": 1.05817246,
      "balance_loss_mlp": 1.0218308,
      "epoch": 0.44634161005230566,
      "flos": 21283033530240.0,
      "grad_norm": 2.750675162008058,
      "language_loss": 0.8225404,
      "learning_rate": 2.4411352800973375e-06,
      "loss": 0.84481573,
      "num_input_tokens_seen": 79937010,
      "step": 3712,
      "time_per_iteration": 2.6028060913085938
    },
    {
      "auxiliary_loss_clip": 0.01176544,
      "auxiliary_loss_mlp": 0.01028835,
      "balance_loss_clip": 1.05601549,
      "balance_loss_mlp": 1.02002549,
      "epoch": 0.44646185294294477,
      "flos": 22929358515840.0,
      "grad_norm": 2.869388872650263,
      "language_loss": 0.75717074,
      "learning_rate": 2.4403754597715005e-06,
      "loss": 0.77922451,
      "num_input_tokens_seen": 79956455,
      "step": 3713,
      "time_per_iteration": 2.6903464794158936
    },
    {
      "auxiliary_loss_clip": 0.01198511,
      "auxiliary_loss_mlp": 0.01032411,
      "balance_loss_clip": 1.05886948,
      "balance_loss_mlp": 1.02292192,
      "epoch": 0.4465820958335838,
      "flos": 22637692080000.0,
      "grad_norm": 2.1431411467604518,
      "language_loss": 0.93175495,
      "learning_rate": 2.4396155726410553e-06,
      "loss": 0.95406419,
      "num_input_tokens_seen": 79975065,
      "step": 3714,
      "time_per_iteration": 2.608032464981079
    },
    {
      "auxiliary_loss_clip": 0.01179417,
      "auxiliary_loss_mlp": 0.01026937,
      "balance_loss_clip": 1.0557878,
      "balance_loss_mlp": 1.01809788,
      "epoch": 0.44670233872422294,
      "flos": 22672525294080.0,
      "grad_norm": 3.7676942951852883,
      "language_loss": 0.91243768,
      "learning_rate": 2.438855618821278e-06,
      "loss": 0.93450123,
      "num_input_tokens_seen": 79990865,
      "step": 3715,
      "time_per_iteration": 2.6766345500946045
    },
    {
      "auxiliary_loss_clip": 0.01156224,
      "auxiliary_loss_mlp": 0.01029931,
      "balance_loss_clip": 1.05220497,
      "balance_loss_mlp": 1.02128267,
      "epoch": 0.44682258161486205,
      "flos": 23582178247680.0,
      "grad_norm": 1.843482765169488,
      "language_loss": 0.67832065,
      "learning_rate": 2.4380955984274517e-06,
      "loss": 0.70018214,
      "num_input_tokens_seen": 80009520,
      "step": 3716,
      "time_per_iteration": 2.6645050048828125
    },
    {
      "auxiliary_loss_clip": 0.01146971,
      "auxiliary_loss_mlp": 0.01036057,
      "balance_loss_clip": 1.05084527,
      "balance_loss_mlp": 1.02730691,
      "epoch": 0.4469428245055011,
      "flos": 26501356558080.0,
      "grad_norm": 2.0409509249857467,
      "language_loss": 0.77243084,
      "learning_rate": 2.4373355115748716e-06,
      "loss": 0.7942611,
      "num_input_tokens_seen": 80030350,
      "step": 3717,
      "time_per_iteration": 2.7638397216796875
    },
    {
      "auxiliary_loss_clip": 0.01181237,
      "auxiliary_loss_mlp": 0.01035015,
      "balance_loss_clip": 1.05621839,
      "balance_loss_mlp": 1.02641404,
      "epoch": 0.4470630673961402,
      "flos": 21504925797120.0,
      "grad_norm": 1.9394444450766621,
      "language_loss": 0.72442031,
      "learning_rate": 2.436575358378842e-06,
      "loss": 0.74658275,
      "num_input_tokens_seen": 80049840,
      "step": 3718,
      "time_per_iteration": 2.6995046138763428
    },
    {
      "auxiliary_loss_clip": 0.01161969,
      "auxiliary_loss_mlp": 0.01034567,
      "balance_loss_clip": 1.05474949,
      "balance_loss_mlp": 1.02629972,
      "epoch": 0.44718331028677927,
      "flos": 16173986653440.0,
      "grad_norm": 3.15571453738153,
      "language_loss": 0.83824557,
      "learning_rate": 2.4358151389546782e-06,
      "loss": 0.86021096,
      "num_input_tokens_seen": 80066525,
      "step": 3719,
      "time_per_iteration": 2.6656997203826904
    },
    {
      "auxiliary_loss_clip": 0.01163451,
      "auxiliary_loss_mlp": 0.01028723,
      "balance_loss_clip": 1.05639076,
      "balance_loss_mlp": 1.01982391,
      "epoch": 0.4473035531774184,
      "flos": 19681238430720.0,
      "grad_norm": 2.194633668257179,
      "language_loss": 0.75858545,
      "learning_rate": 2.4350548534177035e-06,
      "loss": 0.78050715,
      "num_input_tokens_seen": 80083355,
      "step": 3720,
      "time_per_iteration": 2.665738105773926
    },
    {
      "auxiliary_loss_clip": 0.0119126,
      "auxiliary_loss_mlp": 0.01029267,
      "balance_loss_clip": 1.05758977,
      "balance_loss_mlp": 1.02132154,
      "epoch": 0.4474237960680575,
      "flos": 41427590515200.0,
      "grad_norm": 1.7242185526945721,
      "language_loss": 0.67110479,
      "learning_rate": 2.434294501883254e-06,
      "loss": 0.69331002,
      "num_input_tokens_seen": 80106450,
      "step": 3721,
      "time_per_iteration": 2.757413625717163
    },
    {
      "auxiliary_loss_clip": 0.01167607,
      "auxiliary_loss_mlp": 0.01031076,
      "balance_loss_clip": 1.05589604,
      "balance_loss_mlp": 1.02243352,
      "epoch": 0.44754403895869654,
      "flos": 22891328991360.0,
      "grad_norm": 1.8482019461867067,
      "language_loss": 0.65958631,
      "learning_rate": 2.433534084466674e-06,
      "loss": 0.68157315,
      "num_input_tokens_seen": 80125670,
      "step": 3722,
      "time_per_iteration": 2.729578733444214
    },
    {
      "auxiliary_loss_clip": 0.01152729,
      "auxiliary_loss_mlp": 0.00763215,
      "balance_loss_clip": 1.05233514,
      "balance_loss_mlp": 1.00138628,
      "epoch": 0.44766428184933565,
      "flos": 25630271832960.0,
      "grad_norm": 1.563649751294739,
      "language_loss": 0.71135396,
      "learning_rate": 2.4327736012833178e-06,
      "loss": 0.73051339,
      "num_input_tokens_seen": 80147390,
      "step": 3723,
      "time_per_iteration": 2.737943410873413
    },
    {
      "auxiliary_loss_clip": 0.0116504,
      "auxiliary_loss_mlp": 0.01027629,
      "balance_loss_clip": 1.05251706,
      "balance_loss_mlp": 1.01908803,
      "epoch": 0.44778452473997477,
      "flos": 20448972748800.0,
      "grad_norm": 2.4400843483383152,
      "language_loss": 0.76598215,
      "learning_rate": 2.4320130524485506e-06,
      "loss": 0.78790885,
      "num_input_tokens_seen": 80166185,
      "step": 3724,
      "time_per_iteration": 2.669503927230835
    },
    {
      "auxiliary_loss_clip": 0.0117514,
      "auxiliary_loss_mlp": 0.01031855,
      "balance_loss_clip": 1.05539489,
      "balance_loss_mlp": 1.02349818,
      "epoch": 0.4479047676306138,
      "flos": 21975462984960.0,
      "grad_norm": 1.5737795198235827,
      "language_loss": 0.80038357,
      "learning_rate": 2.431252438077746e-06,
      "loss": 0.8224535,
      "num_input_tokens_seen": 80185685,
      "step": 3725,
      "time_per_iteration": 2.6690096855163574
    },
    {
      "auxiliary_loss_clip": 0.01193013,
      "auxiliary_loss_mlp": 0.01027084,
      "balance_loss_clip": 1.05780077,
      "balance_loss_mlp": 1.01804757,
      "epoch": 0.44802501052125293,
      "flos": 21467219495040.0,
      "grad_norm": 2.2902164717505715,
      "language_loss": 0.77479947,
      "learning_rate": 2.4304917582862906e-06,
      "loss": 0.79700041,
      "num_input_tokens_seen": 80204865,
      "step": 3726,
      "time_per_iteration": 2.657423734664917
    },
    {
      "auxiliary_loss_clip": 0.01170554,
      "auxiliary_loss_mlp": 0.01032256,
      "balance_loss_clip": 1.0553925,
      "balance_loss_mlp": 1.02342284,
      "epoch": 0.44814525341189204,
      "flos": 22126970551680.0,
      "grad_norm": 2.13754300809568,
      "language_loss": 0.87819159,
      "learning_rate": 2.4297310131895774e-06,
      "loss": 0.90021968,
      "num_input_tokens_seen": 80223410,
      "step": 3727,
      "time_per_iteration": 4.611078262329102
    },
    {
      "auxiliary_loss_clip": 0.01176303,
      "auxiliary_loss_mlp": 0.00763307,
      "balance_loss_clip": 1.05280709,
      "balance_loss_mlp": 1.00132132,
      "epoch": 0.4482654963025311,
      "flos": 16653933204480.0,
      "grad_norm": 8.768172178522962,
      "language_loss": 0.74495673,
      "learning_rate": 2.4289702029030113e-06,
      "loss": 0.76435286,
      "num_input_tokens_seen": 80240880,
      "step": 3728,
      "time_per_iteration": 2.6548612117767334
    },
    {
      "auxiliary_loss_clip": 0.01197212,
      "auxiliary_loss_mlp": 0.01040605,
      "balance_loss_clip": 1.05937743,
      "balance_loss_mlp": 1.03089547,
      "epoch": 0.4483857391931702,
      "flos": 18841251905280.0,
      "grad_norm": 2.1812015786876233,
      "language_loss": 0.831671,
      "learning_rate": 2.4282093275420057e-06,
      "loss": 0.85404915,
      "num_input_tokens_seen": 80259910,
      "step": 3729,
      "time_per_iteration": 3.558112144470215
    },
    {
      "auxiliary_loss_clip": 0.01194613,
      "auxiliary_loss_mlp": 0.01033483,
      "balance_loss_clip": 1.05924273,
      "balance_loss_mlp": 1.02503705,
      "epoch": 0.4485059820838093,
      "flos": 20372590477440.0,
      "grad_norm": 4.324414584034287,
      "language_loss": 0.70517623,
      "learning_rate": 2.4274483872219863e-06,
      "loss": 0.72745717,
      "num_input_tokens_seen": 80277270,
      "step": 3730,
      "time_per_iteration": 2.624007225036621
    },
    {
      "auxiliary_loss_clip": 0.01170443,
      "auxiliary_loss_mlp": 0.0103472,
      "balance_loss_clip": 1.05473948,
      "balance_loss_mlp": 1.0262208,
      "epoch": 0.4486262249744484,
      "flos": 20047742853120.0,
      "grad_norm": 1.9190618481989374,
      "language_loss": 0.93998063,
      "learning_rate": 2.426687382058386e-06,
      "loss": 0.96203226,
      "num_input_tokens_seen": 80295550,
      "step": 3731,
      "time_per_iteration": 2.641650438308716
    },
    {
      "auxiliary_loss_clip": 0.01085565,
      "auxiliary_loss_mlp": 0.01001762,
      "balance_loss_clip": 1.01819253,
      "balance_loss_mlp": 1.00037885,
      "epoch": 0.4487464678650875,
      "flos": 64595684776320.0,
      "grad_norm": 0.8814261190530253,
      "language_loss": 0.59815156,
      "learning_rate": 2.425926312166649e-06,
      "loss": 0.61902487,
      "num_input_tokens_seen": 80348425,
      "step": 3732,
      "time_per_iteration": 3.089872360229492
    },
    {
      "auxiliary_loss_clip": 0.01199014,
      "auxiliary_loss_mlp": 0.01038969,
      "balance_loss_clip": 1.06009436,
      "balance_loss_mlp": 1.02964711,
      "epoch": 0.4488667107557266,
      "flos": 20769798049920.0,
      "grad_norm": 3.299996979874431,
      "language_loss": 0.73799908,
      "learning_rate": 2.42516517766223e-06,
      "loss": 0.76037896,
      "num_input_tokens_seen": 80366505,
      "step": 3733,
      "time_per_iteration": 2.6110546588897705
    },
    {
      "auxiliary_loss_clip": 0.01196327,
      "auxiliary_loss_mlp": 0.01024827,
      "balance_loss_clip": 1.06143665,
      "balance_loss_mlp": 1.01654506,
      "epoch": 0.44898695364636565,
      "flos": 23951735326080.0,
      "grad_norm": 2.029693755310636,
      "language_loss": 0.68637466,
      "learning_rate": 2.4244039786605907e-06,
      "loss": 0.7085861,
      "num_input_tokens_seen": 80387510,
      "step": 3734,
      "time_per_iteration": 2.711467981338501
    },
    {
      "auxiliary_loss_clip": 0.01193755,
      "auxiliary_loss_mlp": 0.0103201,
      "balance_loss_clip": 1.05652726,
      "balance_loss_mlp": 1.02267551,
      "epoch": 0.44910719653700476,
      "flos": 18624351628800.0,
      "grad_norm": 4.718163593827769,
      "language_loss": 0.82796359,
      "learning_rate": 2.4236427152772055e-06,
      "loss": 0.85022128,
      "num_input_tokens_seen": 80405915,
      "step": 3735,
      "time_per_iteration": 3.5542590618133545
    },
    {
      "auxiliary_loss_clip": 0.01072868,
      "auxiliary_loss_mlp": 0.01001483,
      "balance_loss_clip": 1.0172832,
      "balance_loss_mlp": 1.0001359,
      "epoch": 0.4492274394276438,
      "flos": 57033435749760.0,
      "grad_norm": 0.8406746568327489,
      "language_loss": 0.57347429,
      "learning_rate": 2.422881387627557e-06,
      "loss": 0.59421778,
      "num_input_tokens_seen": 80458365,
      "step": 3736,
      "time_per_iteration": 2.9851582050323486
    },
    {
      "auxiliary_loss_clip": 0.01194253,
      "auxiliary_loss_mlp": 0.01026941,
      "balance_loss_clip": 1.05902195,
      "balance_loss_mlp": 1.01849484,
      "epoch": 0.4493476823182829,
      "flos": 23254888498560.0,
      "grad_norm": 1.7049100537688462,
      "language_loss": 0.77826393,
      "learning_rate": 2.422119995827139e-06,
      "loss": 0.80047584,
      "num_input_tokens_seen": 80478490,
      "step": 3737,
      "time_per_iteration": 2.6970767974853516
    },
    {
      "auxiliary_loss_clip": 0.01196022,
      "auxiliary_loss_mlp": 0.01034811,
      "balance_loss_clip": 1.06098247,
      "balance_loss_mlp": 1.02532792,
      "epoch": 0.44946792520892204,
      "flos": 15815131827840.0,
      "grad_norm": 5.8095640048343355,
      "language_loss": 0.74309564,
      "learning_rate": 2.4213585399914528e-06,
      "loss": 0.76540399,
      "num_input_tokens_seen": 80495695,
      "step": 3738,
      "time_per_iteration": 2.6887452602386475
    },
    {
      "auxiliary_loss_clip": 0.01194824,
      "auxiliary_loss_mlp": 0.01034188,
      "balance_loss_clip": 1.0600214,
      "balance_loss_mlp": 1.02553916,
      "epoch": 0.4495881680995611,
      "flos": 19610063631360.0,
      "grad_norm": 1.9383027770634011,
      "language_loss": 0.85529041,
      "learning_rate": 2.4205970202360113e-06,
      "loss": 0.87758052,
      "num_input_tokens_seen": 80515260,
      "step": 3739,
      "time_per_iteration": 2.6898348331451416
    },
    {
      "auxiliary_loss_clip": 0.01182286,
      "auxiliary_loss_mlp": 0.01032907,
      "balance_loss_clip": 1.05661428,
      "balance_loss_mlp": 1.02345991,
      "epoch": 0.4497084109902002,
      "flos": 26031465815040.0,
      "grad_norm": 1.9218735536927045,
      "language_loss": 0.78253329,
      "learning_rate": 2.4198354366763354e-06,
      "loss": 0.80468524,
      "num_input_tokens_seen": 80533900,
      "step": 3740,
      "time_per_iteration": 2.719895124435425
    },
    {
      "auxiliary_loss_clip": 0.01170306,
      "auxiliary_loss_mlp": 0.0103702,
      "balance_loss_clip": 1.05501103,
      "balance_loss_mlp": 1.02752507,
      "epoch": 0.4498286538808393,
      "flos": 14793688771200.0,
      "grad_norm": 2.0583221356377774,
      "language_loss": 0.79264855,
      "learning_rate": 2.4190737894279587e-06,
      "loss": 0.81472182,
      "num_input_tokens_seen": 80551270,
      "step": 3741,
      "time_per_iteration": 2.7133946418762207
    },
    {
      "auxiliary_loss_clip": 0.01169151,
      "auxiliary_loss_mlp": 0.01032742,
      "balance_loss_clip": 1.05702233,
      "balance_loss_mlp": 1.02446282,
      "epoch": 0.44994889677147837,
      "flos": 15450171690240.0,
      "grad_norm": 2.139665711373394,
      "language_loss": 0.80583191,
      "learning_rate": 2.4183120786064203e-06,
      "loss": 0.82785082,
      "num_input_tokens_seen": 80568145,
      "step": 3742,
      "time_per_iteration": 2.6660706996917725
    },
    {
      "auxiliary_loss_clip": 0.01154047,
      "auxiliary_loss_mlp": 0.0103449,
      "balance_loss_clip": 1.05367231,
      "balance_loss_mlp": 1.0259788,
      "epoch": 0.4500691396621175,
      "flos": 21798316085760.0,
      "grad_norm": 2.5643061253079136,
      "language_loss": 0.85919785,
      "learning_rate": 2.417550304327273e-06,
      "loss": 0.88108319,
      "num_input_tokens_seen": 80586185,
      "step": 3743,
      "time_per_iteration": 2.8184051513671875
    },
    {
      "auxiliary_loss_clip": 0.01180184,
      "auxiliary_loss_mlp": 0.01029298,
      "balance_loss_clip": 1.05800283,
      "balance_loss_mlp": 1.01977956,
      "epoch": 0.4501893825527566,
      "flos": 32382016421760.0,
      "grad_norm": 1.8878690198806967,
      "language_loss": 0.75677174,
      "learning_rate": 2.4167884667060763e-06,
      "loss": 0.77886659,
      "num_input_tokens_seen": 80608895,
      "step": 3744,
      "time_per_iteration": 2.752699851989746
    },
    {
      "auxiliary_loss_clip": 0.01158041,
      "auxiliary_loss_mlp": 0.01035433,
      "balance_loss_clip": 1.05289066,
      "balance_loss_mlp": 1.02618241,
      "epoch": 0.45030962544339564,
      "flos": 16544944362240.0,
      "grad_norm": 2.4928645267756626,
      "language_loss": 0.87488961,
      "learning_rate": 2.4160265658584e-06,
      "loss": 0.89682442,
      "num_input_tokens_seen": 80623785,
      "step": 3745,
      "time_per_iteration": 2.7345969676971436
    },
    {
      "auxiliary_loss_clip": 0.01173545,
      "auxiliary_loss_mlp": 0.01030391,
      "balance_loss_clip": 1.05594051,
      "balance_loss_mlp": 1.02141404,
      "epoch": 0.45042986833403476,
      "flos": 19573039687680.0,
      "grad_norm": 1.9370771275667311,
      "language_loss": 0.68422019,
      "learning_rate": 2.4152646018998253e-06,
      "loss": 0.70625949,
      "num_input_tokens_seen": 80642735,
      "step": 3746,
      "time_per_iteration": 2.654738426208496
    },
    {
      "auxiliary_loss_clip": 0.01180687,
      "auxiliary_loss_mlp": 0.01033154,
      "balance_loss_clip": 1.05637622,
      "balance_loss_mlp": 1.02448118,
      "epoch": 0.45055011122467387,
      "flos": 23112467072640.0,
      "grad_norm": 1.8460427425957389,
      "language_loss": 0.72325671,
      "learning_rate": 2.4145025749459403e-06,
      "loss": 0.74539506,
      "num_input_tokens_seen": 80663760,
      "step": 3747,
      "time_per_iteration": 2.6918442249298096
    },
    {
      "auxiliary_loss_clip": 0.01194223,
      "auxiliary_loss_mlp": 0.01036214,
      "balance_loss_clip": 1.05756021,
      "balance_loss_mlp": 1.02686167,
      "epoch": 0.4506703541153129,
      "flos": 19934623946880.0,
      "grad_norm": 9.34974161391686,
      "language_loss": 0.7035687,
      "learning_rate": 2.413740485112344e-06,
      "loss": 0.72587311,
      "num_input_tokens_seen": 80682100,
      "step": 3748,
      "time_per_iteration": 2.6475577354431152
    },
    {
      "auxiliary_loss_clip": 0.01156921,
      "auxiliary_loss_mlp": 0.0102944,
      "balance_loss_clip": 1.05065203,
      "balance_loss_mlp": 1.02084482,
      "epoch": 0.45079059700595203,
      "flos": 19499530504320.0,
      "grad_norm": 1.525213241246781,
      "language_loss": 0.82691598,
      "learning_rate": 2.412978332514646e-06,
      "loss": 0.84877956,
      "num_input_tokens_seen": 80700880,
      "step": 3749,
      "time_per_iteration": 2.7111613750457764
    },
    {
      "auxiliary_loss_clip": 0.01177925,
      "auxiliary_loss_mlp": 0.01036738,
      "balance_loss_clip": 1.0562222,
      "balance_loss_mlp": 1.02804208,
      "epoch": 0.4509108398965911,
      "flos": 27636313570560.0,
      "grad_norm": 2.106874939839119,
      "language_loss": 0.72686517,
      "learning_rate": 2.4122161172684623e-06,
      "loss": 0.74901181,
      "num_input_tokens_seen": 80721675,
      "step": 3750,
      "time_per_iteration": 2.7404637336730957
    },
    {
      "auxiliary_loss_clip": 0.0118054,
      "auxiliary_loss_mlp": 0.0103187,
      "balance_loss_clip": 1.05749726,
      "balance_loss_mlp": 1.02301252,
      "epoch": 0.4510310827872302,
      "flos": 20995712640000.0,
      "grad_norm": 2.100026417944261,
      "language_loss": 0.84457463,
      "learning_rate": 2.4114538394894216e-06,
      "loss": 0.86669862,
      "num_input_tokens_seen": 80739315,
      "step": 3751,
      "time_per_iteration": 2.7569632530212402
    },
    {
      "auxiliary_loss_clip": 0.01192773,
      "auxiliary_loss_mlp": 0.01027299,
      "balance_loss_clip": 1.05911422,
      "balance_loss_mlp": 1.01842368,
      "epoch": 0.4511513256778693,
      "flos": 16216684945920.0,
      "grad_norm": 2.0693817540258475,
      "language_loss": 0.83487433,
      "learning_rate": 2.410691499293161e-06,
      "loss": 0.8570751,
      "num_input_tokens_seen": 80757470,
      "step": 3752,
      "time_per_iteration": 2.6448166370391846
    },
    {
      "auxiliary_loss_clip": 0.01174701,
      "auxiliary_loss_mlp": 0.01031868,
      "balance_loss_clip": 1.05578077,
      "balance_loss_mlp": 1.02358913,
      "epoch": 0.45127156856850836,
      "flos": 25186702780800.0,
      "grad_norm": 1.5779253696907451,
      "language_loss": 0.74411118,
      "learning_rate": 2.409929096795326e-06,
      "loss": 0.76617688,
      "num_input_tokens_seen": 80777840,
      "step": 3753,
      "time_per_iteration": 4.633047819137573
    },
    {
      "auxiliary_loss_clip": 0.01180093,
      "auxiliary_loss_mlp": 0.01029587,
      "balance_loss_clip": 1.05848455,
      "balance_loss_mlp": 1.02055657,
      "epoch": 0.4513918114591475,
      "flos": 20412523422720.0,
      "grad_norm": 1.9269050027531027,
      "language_loss": 0.79110181,
      "learning_rate": 2.409166632111573e-06,
      "loss": 0.81319869,
      "num_input_tokens_seen": 80795975,
      "step": 3754,
      "time_per_iteration": 2.720308542251587
    },
    {
      "auxiliary_loss_clip": 0.01187952,
      "auxiliary_loss_mlp": 0.01036269,
      "balance_loss_clip": 1.05761623,
      "balance_loss_mlp": 1.02664256,
      "epoch": 0.4515120543497866,
      "flos": 26648482665600.0,
      "grad_norm": 2.0577600306707966,
      "language_loss": 0.8046487,
      "learning_rate": 2.4084041053575674e-06,
      "loss": 0.82689095,
      "num_input_tokens_seen": 80815395,
      "step": 3755,
      "time_per_iteration": 3.8876116275787354
    },
    {
      "auxiliary_loss_clip": 0.01198472,
      "auxiliary_loss_mlp": 0.01030856,
      "balance_loss_clip": 1.06083679,
      "balance_loss_mlp": 1.0217185,
      "epoch": 0.45163229724042564,
      "flos": 20595093275520.0,
      "grad_norm": 2.3806572883912924,
      "language_loss": 0.72743744,
      "learning_rate": 2.4076415166489834e-06,
      "loss": 0.74973071,
      "num_input_tokens_seen": 80834805,
      "step": 3756,
      "time_per_iteration": 2.7071428298950195
    },
    {
      "auxiliary_loss_clip": 0.01193863,
      "auxiliary_loss_mlp": 0.01037172,
      "balance_loss_clip": 1.05757499,
      "balance_loss_mlp": 1.0272243,
      "epoch": 0.45175254013106475,
      "flos": 21689004021120.0,
      "grad_norm": 1.7895937993341362,
      "language_loss": 0.79275453,
      "learning_rate": 2.406878866101506e-06,
      "loss": 0.81506491,
      "num_input_tokens_seen": 80853770,
      "step": 3757,
      "time_per_iteration": 2.644467830657959
    },
    {
      "auxiliary_loss_clip": 0.01168117,
      "auxiliary_loss_mlp": 0.01038907,
      "balance_loss_clip": 1.05873895,
      "balance_loss_mlp": 1.0298506,
      "epoch": 0.45187278302170386,
      "flos": 18878850466560.0,
      "grad_norm": 2.965168457196855,
      "language_loss": 0.78397942,
      "learning_rate": 2.4061161538308273e-06,
      "loss": 0.80604964,
      "num_input_tokens_seen": 80870615,
      "step": 3758,
      "time_per_iteration": 2.7411367893218994
    },
    {
      "auxiliary_loss_clip": 0.01192588,
      "auxiliary_loss_mlp": 0.01026271,
      "balance_loss_clip": 1.05768943,
      "balance_loss_mlp": 1.01766348,
      "epoch": 0.4519930259123429,
      "flos": 18582479349120.0,
      "grad_norm": 2.2501389911516725,
      "language_loss": 0.89919174,
      "learning_rate": 2.4053533799526523e-06,
      "loss": 0.92138028,
      "num_input_tokens_seen": 80886335,
      "step": 3759,
      "time_per_iteration": 2.6443583965301514
    },
    {
      "auxiliary_loss_clip": 0.01191686,
      "auxiliary_loss_mlp": 0.01030342,
      "balance_loss_clip": 1.05722737,
      "balance_loss_mlp": 1.02097225,
      "epoch": 0.452113268802982,
      "flos": 25192377129600.0,
      "grad_norm": 1.86010088330828,
      "language_loss": 0.86713147,
      "learning_rate": 2.404590544582691e-06,
      "loss": 0.88935173,
      "num_input_tokens_seen": 80904570,
      "step": 3760,
      "time_per_iteration": 2.71012282371521
    },
    {
      "auxiliary_loss_clip": 0.01171908,
      "auxiliary_loss_mlp": 0.01035577,
      "balance_loss_clip": 1.05534291,
      "balance_loss_mlp": 1.02660036,
      "epoch": 0.45223351169362114,
      "flos": 39378922312320.0,
      "grad_norm": 1.7105824018374487,
      "language_loss": 0.81145084,
      "learning_rate": 2.403827647836666e-06,
      "loss": 0.83352566,
      "num_input_tokens_seen": 80925125,
      "step": 3761,
      "time_per_iteration": 2.8975260257720947
    },
    {
      "auxiliary_loss_clip": 0.01193222,
      "auxiliary_loss_mlp": 0.01035633,
      "balance_loss_clip": 1.05645013,
      "balance_loss_mlp": 1.0265851,
      "epoch": 0.4523537545842602,
      "flos": 21582169994880.0,
      "grad_norm": 3.4670246328576817,
      "language_loss": 0.69776976,
      "learning_rate": 2.4030646898303075e-06,
      "loss": 0.72005832,
      "num_input_tokens_seen": 80946615,
      "step": 3762,
      "time_per_iteration": 3.6192564964294434
    },
    {
      "auxiliary_loss_clip": 0.01190948,
      "auxiliary_loss_mlp": 0.01037687,
      "balance_loss_clip": 1.05612338,
      "balance_loss_mlp": 1.02915144,
      "epoch": 0.4524739974748993,
      "flos": 28439527547520.0,
      "grad_norm": 2.4220095093456475,
      "language_loss": 0.8263768,
      "learning_rate": 2.4023016706793566e-06,
      "loss": 0.84866315,
      "num_input_tokens_seen": 80966410,
      "step": 3763,
      "time_per_iteration": 2.6863152980804443
    },
    {
      "auxiliary_loss_clip": 0.01084434,
      "auxiliary_loss_mlp": 0.00999427,
      "balance_loss_clip": 1.01692188,
      "balance_loss_mlp": 0.9978537,
      "epoch": 0.4525942403655384,
      "flos": 61556492148480.0,
      "grad_norm": 0.7882836075925704,
      "language_loss": 0.56842822,
      "learning_rate": 2.401538590499561e-06,
      "loss": 0.58926684,
      "num_input_tokens_seen": 81026865,
      "step": 3764,
      "time_per_iteration": 3.2580347061157227
    },
    {
      "auxiliary_loss_clip": 0.01168823,
      "auxiliary_loss_mlp": 0.01039355,
      "balance_loss_clip": 1.0566318,
      "balance_loss_mlp": 1.03034866,
      "epoch": 0.45271448325617747,
      "flos": 27529838680320.0,
      "grad_norm": 2.0420982836676775,
      "language_loss": 0.71820796,
      "learning_rate": 2.400775449406682e-06,
      "loss": 0.74028981,
      "num_input_tokens_seen": 81050060,
      "step": 3765,
      "time_per_iteration": 2.818162679672241
    },
    {
      "auxiliary_loss_clip": 0.0119004,
      "auxiliary_loss_mlp": 0.01030121,
      "balance_loss_clip": 1.05558991,
      "balance_loss_mlp": 1.02131104,
      "epoch": 0.4528347261468166,
      "flos": 22452608275200.0,
      "grad_norm": 1.989871968101558,
      "language_loss": 0.73427457,
      "learning_rate": 2.400012247516485e-06,
      "loss": 0.75647616,
      "num_input_tokens_seen": 81070625,
      "step": 3766,
      "time_per_iteration": 2.633502244949341
    },
    {
      "auxiliary_loss_clip": 0.01166514,
      "auxiliary_loss_mlp": 0.00763588,
      "balance_loss_clip": 1.05756235,
      "balance_loss_mlp": 1.00141573,
      "epoch": 0.45295496903745563,
      "flos": 21103875469440.0,
      "grad_norm": 1.7211855102195168,
      "language_loss": 0.90306801,
      "learning_rate": 2.3992489849447484e-06,
      "loss": 0.922369,
      "num_input_tokens_seen": 81089080,
      "step": 3767,
      "time_per_iteration": 2.748457908630371
    },
    {
      "auxiliary_loss_clip": 0.01181225,
      "auxiliary_loss_mlp": 0.01028888,
      "balance_loss_clip": 1.05693889,
      "balance_loss_mlp": 1.02026939,
      "epoch": 0.45307521192809475,
      "flos": 23221168606080.0,
      "grad_norm": 6.291814678381155,
      "language_loss": 0.79152542,
      "learning_rate": 2.3984856618072584e-06,
      "loss": 0.81362653,
      "num_input_tokens_seen": 81109115,
      "step": 3768,
      "time_per_iteration": 2.6789422035217285
    },
    {
      "auxiliary_loss_clip": 0.01182809,
      "auxiliary_loss_mlp": 0.01031931,
      "balance_loss_clip": 1.06006134,
      "balance_loss_mlp": 1.02337456,
      "epoch": 0.45319545481873386,
      "flos": 15560094286080.0,
      "grad_norm": 1.9484424192941279,
      "language_loss": 0.74536127,
      "learning_rate": 2.3977222782198098e-06,
      "loss": 0.76750869,
      "num_input_tokens_seen": 81127750,
      "step": 3769,
      "time_per_iteration": 2.6932132244110107
    },
    {
      "auxiliary_loss_clip": 0.01180265,
      "auxiliary_loss_mlp": 0.01036437,
      "balance_loss_clip": 1.05760908,
      "balance_loss_mlp": 1.02674496,
      "epoch": 0.4533156977093729,
      "flos": 21944759834880.0,
      "grad_norm": 1.7114747194793074,
      "language_loss": 0.75541794,
      "learning_rate": 2.3969588342982077e-06,
      "loss": 0.77758503,
      "num_input_tokens_seen": 81147125,
      "step": 3770,
      "time_per_iteration": 2.6876351833343506
    },
    {
      "auxiliary_loss_clip": 0.01178639,
      "auxiliary_loss_mlp": 0.01034018,
      "balance_loss_clip": 1.05859947,
      "balance_loss_mlp": 1.02522063,
      "epoch": 0.453435940600012,
      "flos": 24242180699520.0,
      "grad_norm": 2.4561202839433025,
      "language_loss": 0.73065209,
      "learning_rate": 2.396195330158267e-06,
      "loss": 0.75277865,
      "num_input_tokens_seen": 81167015,
      "step": 3771,
      "time_per_iteration": 2.789945125579834
    },
    {
      "auxiliary_loss_clip": 0.01164654,
      "auxiliary_loss_mlp": 0.01033435,
      "balance_loss_clip": 1.05704057,
      "balance_loss_mlp": 1.0243454,
      "epoch": 0.45355618349065113,
      "flos": 23440367352960.0,
      "grad_norm": 4.627604427256918,
      "language_loss": 0.79663634,
      "learning_rate": 2.3954317659158094e-06,
      "loss": 0.81861722,
      "num_input_tokens_seen": 81187350,
      "step": 3772,
      "time_per_iteration": 2.9019205570220947
    },
    {
      "auxiliary_loss_clip": 0.01083313,
      "auxiliary_loss_mlp": 0.01001336,
      "balance_loss_clip": 1.01590943,
      "balance_loss_mlp": 0.99984008,
      "epoch": 0.4536764263812902,
      "flos": 66903161448960.0,
      "grad_norm": 0.8884498849083624,
      "language_loss": 0.56872952,
      "learning_rate": 2.394668141686667e-06,
      "loss": 0.58957601,
      "num_input_tokens_seen": 81249315,
      "step": 3773,
      "time_per_iteration": 3.225715398788452
    },
    {
      "auxiliary_loss_clip": 0.01193704,
      "auxiliary_loss_mlp": 0.01026347,
      "balance_loss_clip": 1.05847955,
      "balance_loss_mlp": 1.01809192,
      "epoch": 0.4537966692719293,
      "flos": 42739766254080.0,
      "grad_norm": 2.213612153969597,
      "language_loss": 0.69889879,
      "learning_rate": 2.3939044575866813e-06,
      "loss": 0.72109926,
      "num_input_tokens_seen": 81272065,
      "step": 3774,
      "time_per_iteration": 2.8542284965515137
    },
    {
      "auxiliary_loss_clip": 0.01194073,
      "auxiliary_loss_mlp": 0.01030119,
      "balance_loss_clip": 1.05861068,
      "balance_loss_mlp": 1.02058852,
      "epoch": 0.4539169121625684,
      "flos": 35549480517120.0,
      "grad_norm": 4.48154743811657,
      "language_loss": 0.75891411,
      "learning_rate": 2.3931407137317024e-06,
      "loss": 0.78115606,
      "num_input_tokens_seen": 81292220,
      "step": 3775,
      "time_per_iteration": 2.714080810546875
    },
    {
      "auxiliary_loss_clip": 0.01172231,
      "auxiliary_loss_mlp": 0.01033367,
      "balance_loss_clip": 1.0555954,
      "balance_loss_mlp": 1.02445614,
      "epoch": 0.45403715505320746,
      "flos": 18514716341760.0,
      "grad_norm": 1.743071253661371,
      "language_loss": 0.85043848,
      "learning_rate": 2.3923769102375907e-06,
      "loss": 0.87249446,
      "num_input_tokens_seen": 81311085,
      "step": 3776,
      "time_per_iteration": 2.7162656784057617
    },
    {
      "auxiliary_loss_clip": 0.01194974,
      "auxiliary_loss_mlp": 0.01030091,
      "balance_loss_clip": 1.05830443,
      "balance_loss_mlp": 1.0210427,
      "epoch": 0.4541573979438466,
      "flos": 25045825639680.0,
      "grad_norm": 1.9702395630578753,
      "language_loss": 0.7870748,
      "learning_rate": 2.391613047220213e-06,
      "loss": 0.80932546,
      "num_input_tokens_seen": 81330985,
      "step": 3777,
      "time_per_iteration": 2.5980730056762695
    },
    {
      "auxiliary_loss_clip": 0.01194605,
      "auxiliary_loss_mlp": 0.01033056,
      "balance_loss_clip": 1.05906987,
      "balance_loss_mlp": 1.024616,
      "epoch": 0.4542776408344857,
      "flos": 18332397884160.0,
      "grad_norm": 2.908782979802358,
      "language_loss": 0.79145956,
      "learning_rate": 2.390849124795447e-06,
      "loss": 0.8137362,
      "num_input_tokens_seen": 81346985,
      "step": 3778,
      "time_per_iteration": 2.651641607284546
    },
    {
      "auxiliary_loss_clip": 0.01182345,
      "auxiliary_loss_mlp": 0.01029658,
      "balance_loss_clip": 1.05664814,
      "balance_loss_mlp": 1.02033615,
      "epoch": 0.45439788372512474,
      "flos": 20701173116160.0,
      "grad_norm": 3.00068762526536,
      "language_loss": 0.84154713,
      "learning_rate": 2.3900851430791804e-06,
      "loss": 0.86366713,
      "num_input_tokens_seen": 81365005,
      "step": 3779,
      "time_per_iteration": 3.6065995693206787
    },
    {
      "auxiliary_loss_clip": 0.01185873,
      "auxiliary_loss_mlp": 0.01037716,
      "balance_loss_clip": 1.05711067,
      "balance_loss_mlp": 1.02833462,
      "epoch": 0.45451812661576385,
      "flos": 22309432663680.0,
      "grad_norm": 1.9489374617786392,
      "language_loss": 0.84915292,
      "learning_rate": 2.389321102187307e-06,
      "loss": 0.87138879,
      "num_input_tokens_seen": 81383785,
      "step": 3780,
      "time_per_iteration": 3.578979969024658
    },
    {
      "auxiliary_loss_clip": 0.01185945,
      "auxiliary_loss_mlp": 0.01031511,
      "balance_loss_clip": 1.05742788,
      "balance_loss_mlp": 1.02215326,
      "epoch": 0.4546383695064029,
      "flos": 21763303303680.0,
      "grad_norm": 1.906824091929148,
      "language_loss": 0.82054871,
      "learning_rate": 2.3885570022357326e-06,
      "loss": 0.84272325,
      "num_input_tokens_seen": 81402915,
      "step": 3781,
      "time_per_iteration": 3.6545369625091553
    },
    {
      "auxiliary_loss_clip": 0.01032564,
      "auxiliary_loss_mlp": 0.01010937,
      "balance_loss_clip": 1.01780915,
      "balance_loss_mlp": 1.00944138,
      "epoch": 0.454758612397042,
      "flos": 64242755694720.0,
      "grad_norm": 0.8062802086293879,
      "language_loss": 0.60843468,
      "learning_rate": 2.38779284334037e-06,
      "loss": 0.62886965,
      "num_input_tokens_seen": 81467890,
      "step": 3782,
      "time_per_iteration": 3.5557518005371094
    },
    {
      "auxiliary_loss_clip": 0.0115614,
      "auxiliary_loss_mlp": 0.01030424,
      "balance_loss_clip": 1.05314803,
      "balance_loss_mlp": 1.02126861,
      "epoch": 0.4548788552876811,
      "flos": 27304175485440.0,
      "grad_norm": 2.0433675317975513,
      "language_loss": 0.78662485,
      "learning_rate": 2.387028625617141e-06,
      "loss": 0.80849051,
      "num_input_tokens_seen": 81487105,
      "step": 3783,
      "time_per_iteration": 2.943434476852417
    },
    {
      "auxiliary_loss_clip": 0.01178033,
      "auxiliary_loss_mlp": 0.01028585,
      "balance_loss_clip": 1.05310619,
      "balance_loss_mlp": 1.0191853,
      "epoch": 0.4549990981783202,
      "flos": 22857142222080.0,
      "grad_norm": 3.012263905934659,
      "language_loss": 0.84921432,
      "learning_rate": 2.3862643491819766e-06,
      "loss": 0.87128055,
      "num_input_tokens_seen": 81505670,
      "step": 3784,
      "time_per_iteration": 2.70564341545105
    },
    {
      "auxiliary_loss_clip": 0.0115904,
      "auxiliary_loss_mlp": 0.01027414,
      "balance_loss_clip": 1.05550122,
      "balance_loss_mlp": 1.01888502,
      "epoch": 0.4551193410689593,
      "flos": 23258587599360.0,
      "grad_norm": 1.796178606489892,
      "language_loss": 0.84361172,
      "learning_rate": 2.3855000141508186e-06,
      "loss": 0.86547625,
      "num_input_tokens_seen": 81525825,
      "step": 3785,
      "time_per_iteration": 2.7188165187835693
    },
    {
      "auxiliary_loss_clip": 0.01183506,
      "auxiliary_loss_mlp": 0.01031966,
      "balance_loss_clip": 1.05667377,
      "balance_loss_mlp": 1.02271557,
      "epoch": 0.4552395839595984,
      "flos": 20777519473920.0,
      "grad_norm": 2.3338922512741624,
      "language_loss": 0.841627,
      "learning_rate": 2.3847356206396143e-06,
      "loss": 0.86378181,
      "num_input_tokens_seen": 81543135,
      "step": 3786,
      "time_per_iteration": 2.6379854679107666
    },
    {
      "auxiliary_loss_clip": 0.01181517,
      "auxiliary_loss_mlp": 0.01038116,
      "balance_loss_clip": 1.05936229,
      "balance_loss_mlp": 1.028651,
      "epoch": 0.45535982685023746,
      "flos": 23257510191360.0,
      "grad_norm": 1.7383015498815146,
      "language_loss": 0.78455931,
      "learning_rate": 2.3839711687643227e-06,
      "loss": 0.80675566,
      "num_input_tokens_seen": 81564360,
      "step": 3787,
      "time_per_iteration": 2.6938438415527344
    },
    {
      "auxiliary_loss_clip": 0.01157326,
      "auxiliary_loss_mlp": 0.01031082,
      "balance_loss_clip": 1.05145121,
      "balance_loss_mlp": 1.02153325,
      "epoch": 0.45548006974087657,
      "flos": 19646117907840.0,
      "grad_norm": 2.5563227497778946,
      "language_loss": 0.7447536,
      "learning_rate": 2.38320665864091e-06,
      "loss": 0.76663768,
      "num_input_tokens_seen": 81583710,
      "step": 3788,
      "time_per_iteration": 3.6079916954040527
    },
    {
      "auxiliary_loss_clip": 0.01175759,
      "auxiliary_loss_mlp": 0.01032588,
      "balance_loss_clip": 1.05331695,
      "balance_loss_mlp": 1.02210331,
      "epoch": 0.4556003126315157,
      "flos": 20047778766720.0,
      "grad_norm": 1.9340701998365681,
      "language_loss": 0.81782991,
      "learning_rate": 2.3824420903853516e-06,
      "loss": 0.83991337,
      "num_input_tokens_seen": 81602175,
      "step": 3789,
      "time_per_iteration": 2.694093704223633
    },
    {
      "auxiliary_loss_clip": 0.01157484,
      "auxiliary_loss_mlp": 0.01039094,
      "balance_loss_clip": 1.0556277,
      "balance_loss_mlp": 1.03065372,
      "epoch": 0.45572055552215474,
      "flos": 22959738443520.0,
      "grad_norm": 2.327169334029541,
      "language_loss": 0.82334197,
      "learning_rate": 2.3816774641136324e-06,
      "loss": 0.84530777,
      "num_input_tokens_seen": 81619430,
      "step": 3790,
      "time_per_iteration": 2.6803159713745117
    },
    {
      "auxiliary_loss_clip": 0.01176629,
      "auxiliary_loss_mlp": 0.01034131,
      "balance_loss_clip": 1.05763686,
      "balance_loss_mlp": 1.02544689,
      "epoch": 0.45584079841279385,
      "flos": 33109925535360.0,
      "grad_norm": 1.7437861290489813,
      "language_loss": 0.71584249,
      "learning_rate": 2.380912779941745e-06,
      "loss": 0.73795009,
      "num_input_tokens_seen": 81642550,
      "step": 3791,
      "time_per_iteration": 2.7170629501342773
    },
    {
      "auxiliary_loss_clip": 0.01187214,
      "auxiliary_loss_mlp": 0.01034342,
      "balance_loss_clip": 1.05678058,
      "balance_loss_mlp": 1.02487659,
      "epoch": 0.45596104130343296,
      "flos": 27272179445760.0,
      "grad_norm": 2.395435522511629,
      "language_loss": 0.8317461,
      "learning_rate": 2.3801480379856918e-06,
      "loss": 0.85396165,
      "num_input_tokens_seen": 81664260,
      "step": 3792,
      "time_per_iteration": 2.6636850833892822
    },
    {
      "auxiliary_loss_clip": 0.01195238,
      "auxiliary_loss_mlp": 0.01034659,
      "balance_loss_clip": 1.05875826,
      "balance_loss_mlp": 1.02625513,
      "epoch": 0.456081284194072,
      "flos": 21579799697280.0,
      "grad_norm": 1.641203733202709,
      "language_loss": 0.83693886,
      "learning_rate": 2.379383238361484e-06,
      "loss": 0.85923779,
      "num_input_tokens_seen": 81683620,
      "step": 3793,
      "time_per_iteration": 2.7306056022644043
    },
    {
      "auxiliary_loss_clip": 0.01178057,
      "auxiliary_loss_mlp": 0.01030874,
      "balance_loss_clip": 1.05531883,
      "balance_loss_mlp": 1.02214813,
      "epoch": 0.4562015270847111,
      "flos": 35918822113920.0,
      "grad_norm": 2.3697163317342955,
      "language_loss": 0.79679966,
      "learning_rate": 2.3786183811851407e-06,
      "loss": 0.8188889,
      "num_input_tokens_seen": 81704325,
      "step": 3794,
      "time_per_iteration": 2.7734904289245605
    },
    {
      "auxiliary_loss_clip": 0.01173639,
      "auxiliary_loss_mlp": 0.01028475,
      "balance_loss_clip": 1.05581105,
      "balance_loss_mlp": 1.02008903,
      "epoch": 0.45632176997535023,
      "flos": 13589783602560.0,
      "grad_norm": 11.868636200859351,
      "language_loss": 0.7991842,
      "learning_rate": 2.3778534665726892e-06,
      "loss": 0.82120538,
      "num_input_tokens_seen": 81721155,
      "step": 3795,
      "time_per_iteration": 2.6258578300476074
    },
    {
      "auxiliary_loss_clip": 0.01192424,
      "auxiliary_loss_mlp": 0.01032502,
      "balance_loss_clip": 1.05930066,
      "balance_loss_mlp": 1.02399635,
      "epoch": 0.4564420128659893,
      "flos": 32635401937920.0,
      "grad_norm": 1.7168599786498344,
      "language_loss": 0.72748518,
      "learning_rate": 2.377088494640168e-06,
      "loss": 0.7497344,
      "num_input_tokens_seen": 81742905,
      "step": 3796,
      "time_per_iteration": 2.791386842727661
    },
    {
      "auxiliary_loss_clip": 0.01177097,
      "auxiliary_loss_mlp": 0.01033784,
      "balance_loss_clip": 1.05741656,
      "balance_loss_mlp": 1.02555811,
      "epoch": 0.4565622557566284,
      "flos": 20377690208640.0,
      "grad_norm": 1.9392874028113654,
      "language_loss": 0.77661312,
      "learning_rate": 2.3763234655036216e-06,
      "loss": 0.79872191,
      "num_input_tokens_seen": 81762105,
      "step": 3797,
      "time_per_iteration": 2.672680616378784
    },
    {
      "auxiliary_loss_clip": 0.01193657,
      "auxiliary_loss_mlp": 0.01031087,
      "balance_loss_clip": 1.05939138,
      "balance_loss_mlp": 1.02191997,
      "epoch": 0.45668249864726745,
      "flos": 25374372364800.0,
      "grad_norm": 2.301807699461474,
      "language_loss": 0.87100196,
      "learning_rate": 2.3755583792791046e-06,
      "loss": 0.89324939,
      "num_input_tokens_seen": 81781975,
      "step": 3798,
      "time_per_iteration": 2.6710565090179443
    },
    {
      "auxiliary_loss_clip": 0.01176469,
      "auxiliary_loss_mlp": 0.01034644,
      "balance_loss_clip": 1.05585682,
      "balance_loss_mlp": 1.02570915,
      "epoch": 0.45680274153790656,
      "flos": 15559806977280.0,
      "grad_norm": 2.569989093523244,
      "language_loss": 0.74379802,
      "learning_rate": 2.3747932360826803e-06,
      "loss": 0.76590908,
      "num_input_tokens_seen": 81798905,
      "step": 3799,
      "time_per_iteration": 2.6006968021392822
    },
    {
      "auxiliary_loss_clip": 0.01180589,
      "auxiliary_loss_mlp": 0.00763744,
      "balance_loss_clip": 1.05763745,
      "balance_loss_mlp": 1.00119233,
      "epoch": 0.4569229844285457,
      "flos": 19792884879360.0,
      "grad_norm": 1.9125842684440948,
      "language_loss": 0.82472962,
      "learning_rate": 2.3740280360304205e-06,
      "loss": 0.84417295,
      "num_input_tokens_seen": 81816630,
      "step": 3800,
      "time_per_iteration": 2.6904702186584473
    },
    {
      "auxiliary_loss_clip": 0.01157803,
      "auxiliary_loss_mlp": 0.01024434,
      "balance_loss_clip": 1.05250835,
      "balance_loss_mlp": 1.0155232,
      "epoch": 0.45704322731918473,
      "flos": 24093941270400.0,
      "grad_norm": 2.5120822556632727,
      "language_loss": 0.68300521,
      "learning_rate": 2.3732627792384038e-06,
      "loss": 0.70482761,
      "num_input_tokens_seen": 81837700,
      "step": 3801,
      "time_per_iteration": 2.665558338165283
    },
    {
      "auxiliary_loss_clip": 0.01192527,
      "auxiliary_loss_mlp": 0.01032583,
      "balance_loss_clip": 1.05780637,
      "balance_loss_mlp": 1.02273631,
      "epoch": 0.45716347020982384,
      "flos": 31317803245440.0,
      "grad_norm": 1.8243310981009384,
      "language_loss": 0.75661993,
      "learning_rate": 2.3724974658227207e-06,
      "loss": 0.77887106,
      "num_input_tokens_seen": 81858490,
      "step": 3802,
      "time_per_iteration": 2.7347004413604736
    },
    {
      "auxiliary_loss_clip": 0.01169768,
      "auxiliary_loss_mlp": 0.01032812,
      "balance_loss_clip": 1.05673814,
      "balance_loss_mlp": 1.02349007,
      "epoch": 0.45728371310046295,
      "flos": 26501392471680.0,
      "grad_norm": 1.8226287687284901,
      "language_loss": 0.71580434,
      "learning_rate": 2.3717320958994687e-06,
      "loss": 0.7378301,
      "num_input_tokens_seen": 81876050,
      "step": 3803,
      "time_per_iteration": 2.6970231533050537
    },
    {
      "auxiliary_loss_clip": 0.01191763,
      "auxiliary_loss_mlp": 0.01038148,
      "balance_loss_clip": 1.0568018,
      "balance_loss_mlp": 1.02868843,
      "epoch": 0.457403955991102,
      "flos": 17929408222080.0,
      "grad_norm": 3.1044300588180507,
      "language_loss": 0.7063055,
      "learning_rate": 2.3709666695847534e-06,
      "loss": 0.72860461,
      "num_input_tokens_seen": 81894230,
      "step": 3804,
      "time_per_iteration": 2.6263999938964844
    },
    {
      "auxiliary_loss_clip": 0.01190547,
      "auxiliary_loss_mlp": 0.01033786,
      "balance_loss_clip": 1.05816686,
      "balance_loss_mlp": 1.02492309,
      "epoch": 0.4575241988817411,
      "flos": 42230660837760.0,
      "grad_norm": 2.005270709681734,
      "language_loss": 0.70499396,
      "learning_rate": 2.370201186994689e-06,
      "loss": 0.72723734,
      "num_input_tokens_seen": 81917915,
      "step": 3805,
      "time_per_iteration": 3.8013362884521484
    },
    {
      "auxiliary_loss_clip": 0.01194403,
      "auxiliary_loss_mlp": 0.01041749,
      "balance_loss_clip": 1.06003046,
      "balance_loss_mlp": 1.03276658,
      "epoch": 0.45764444177238023,
      "flos": 30117309868800.0,
      "grad_norm": 3.685149832778251,
      "language_loss": 0.7020992,
      "learning_rate": 2.369435648245399e-06,
      "loss": 0.72446066,
      "num_input_tokens_seen": 81938130,
      "step": 3806,
      "time_per_iteration": 2.6977226734161377
    },
    {
      "auxiliary_loss_clip": 0.01193724,
      "auxiliary_loss_mlp": 0.01038579,
      "balance_loss_clip": 1.05927062,
      "balance_loss_mlp": 1.02948391,
      "epoch": 0.4577646846630193,
      "flos": 24060293205120.0,
      "grad_norm": 2.710281515780454,
      "language_loss": 0.85014653,
      "learning_rate": 2.368670053453015e-06,
      "loss": 0.87246954,
      "num_input_tokens_seen": 81959820,
      "step": 3807,
      "time_per_iteration": 3.6449410915374756
    },
    {
      "auxiliary_loss_clip": 0.01194609,
      "auxiliary_loss_mlp": 0.01038424,
      "balance_loss_clip": 1.05885482,
      "balance_loss_mlp": 1.02900028,
      "epoch": 0.4578849275536584,
      "flos": 17418578952960.0,
      "grad_norm": 2.567544158049267,
      "language_loss": 0.74719542,
      "learning_rate": 2.3679044027336757e-06,
      "loss": 0.76952577,
      "num_input_tokens_seen": 81975710,
      "step": 3808,
      "time_per_iteration": 2.6203904151916504
    },
    {
      "auxiliary_loss_clip": 0.0119629,
      "auxiliary_loss_mlp": 0.01036968,
      "balance_loss_clip": 1.05900693,
      "balance_loss_mlp": 1.02770591,
      "epoch": 0.4580051704442975,
      "flos": 13510169107200.0,
      "grad_norm": 6.7298193125738885,
      "language_loss": 0.69536382,
      "learning_rate": 2.3671386962035326e-06,
      "loss": 0.71769637,
      "num_input_tokens_seen": 81993180,
      "step": 3809,
      "time_per_iteration": 2.6551551818847656
    },
    {
      "auxiliary_loss_clip": 0.01182036,
      "auxiliary_loss_mlp": 0.01029669,
      "balance_loss_clip": 1.05677748,
      "balance_loss_mlp": 1.02016854,
      "epoch": 0.45812541333493656,
      "flos": 18037606965120.0,
      "grad_norm": 2.3984512174561163,
      "language_loss": 0.68952656,
      "learning_rate": 2.3663729339787405e-06,
      "loss": 0.71164364,
      "num_input_tokens_seen": 82010115,
      "step": 3810,
      "time_per_iteration": 2.700488328933716
    },
    {
      "auxiliary_loss_clip": 0.01176369,
      "auxiliary_loss_mlp": 0.01040017,
      "balance_loss_clip": 1.054075,
      "balance_loss_mlp": 1.03056979,
      "epoch": 0.45824565622557567,
      "flos": 20222196232320.0,
      "grad_norm": 2.548398044735349,
      "language_loss": 0.73380703,
      "learning_rate": 2.365607116175466e-06,
      "loss": 0.75597084,
      "num_input_tokens_seen": 82025540,
      "step": 3811,
      "time_per_iteration": 2.6262805461883545
    },
    {
      "auxiliary_loss_clip": 0.01195641,
      "auxiliary_loss_mlp": 0.01030368,
      "balance_loss_clip": 1.05937791,
      "balance_loss_mlp": 1.0217495,
      "epoch": 0.4583658991162148,
      "flos": 19864885691520.0,
      "grad_norm": 2.7909729496291584,
      "language_loss": 0.66994482,
      "learning_rate": 2.3648412429098825e-06,
      "loss": 0.69220489,
      "num_input_tokens_seen": 82043890,
      "step": 3812,
      "time_per_iteration": 2.6617159843444824
    },
    {
      "auxiliary_loss_clip": 0.01198191,
      "auxiliary_loss_mlp": 0.01038794,
      "balance_loss_clip": 1.05929232,
      "balance_loss_mlp": 1.02913165,
      "epoch": 0.45848614200685384,
      "flos": 21029935322880.0,
      "grad_norm": 1.8176250338724416,
      "language_loss": 0.8209815,
      "learning_rate": 2.364075314298172e-06,
      "loss": 0.84335136,
      "num_input_tokens_seen": 82061345,
      "step": 3813,
      "time_per_iteration": 2.683412790298462
    },
    {
      "auxiliary_loss_clip": 0.01142492,
      "auxiliary_loss_mlp": 0.01030713,
      "balance_loss_clip": 1.05536389,
      "balance_loss_mlp": 1.02161717,
      "epoch": 0.45860638489749295,
      "flos": 21069293650560.0,
      "grad_norm": 2.064341333673032,
      "language_loss": 0.70443845,
      "learning_rate": 2.3633093304565267e-06,
      "loss": 0.72617054,
      "num_input_tokens_seen": 82080400,
      "step": 3814,
      "time_per_iteration": 3.707383394241333
    },
    {
      "auxiliary_loss_clip": 0.01198636,
      "auxiliary_loss_mlp": 0.01032153,
      "balance_loss_clip": 1.0610249,
      "balance_loss_mlp": 1.02288449,
      "epoch": 0.458726627788132,
      "flos": 26833889692800.0,
      "grad_norm": 2.123914647030383,
      "language_loss": 0.62878168,
      "learning_rate": 2.3625432915011443e-06,
      "loss": 0.65108961,
      "num_input_tokens_seen": 82102310,
      "step": 3815,
      "time_per_iteration": 2.8399598598480225
    },
    {
      "auxiliary_loss_clip": 0.01191908,
      "auxiliary_loss_mlp": 0.01034143,
      "balance_loss_clip": 1.0575974,
      "balance_loss_mlp": 1.02524972,
      "epoch": 0.4588468706787711,
      "flos": 24097927680000.0,
      "grad_norm": 2.071671985846698,
      "language_loss": 0.6545614,
      "learning_rate": 2.3617771975482334e-06,
      "loss": 0.67682195,
      "num_input_tokens_seen": 82121140,
      "step": 3816,
      "time_per_iteration": 2.7809128761291504
    },
    {
      "auxiliary_loss_clip": 0.01181415,
      "auxiliary_loss_mlp": 0.01030959,
      "balance_loss_clip": 1.05932474,
      "balance_loss_mlp": 1.02251315,
      "epoch": 0.4589671135694102,
      "flos": 17889331622400.0,
      "grad_norm": 1.7719812123104057,
      "language_loss": 0.74520802,
      "learning_rate": 2.3610110487140083e-06,
      "loss": 0.76733172,
      "num_input_tokens_seen": 82139575,
      "step": 3817,
      "time_per_iteration": 2.676736354827881
    },
    {
      "auxiliary_loss_clip": 0.01173326,
      "auxiliary_loss_mlp": 0.01030868,
      "balance_loss_clip": 1.05624199,
      "balance_loss_mlp": 1.02239215,
      "epoch": 0.4590873564600493,
      "flos": 25626967781760.0,
      "grad_norm": 2.220033436258404,
      "language_loss": 0.81160533,
      "learning_rate": 2.360244845114695e-06,
      "loss": 0.83364725,
      "num_input_tokens_seen": 82159195,
      "step": 3818,
      "time_per_iteration": 2.68965744972229
    },
    {
      "auxiliary_loss_clip": 0.01167557,
      "auxiliary_loss_mlp": 0.01031476,
      "balance_loss_clip": 1.05235863,
      "balance_loss_mlp": 1.02207673,
      "epoch": 0.4592075993506884,
      "flos": 18514788168960.0,
      "grad_norm": 5.601000170258729,
      "language_loss": 0.68810689,
      "learning_rate": 2.3594785868665245e-06,
      "loss": 0.71009725,
      "num_input_tokens_seen": 82175500,
      "step": 3819,
      "time_per_iteration": 2.740760564804077
    },
    {
      "auxiliary_loss_clip": 0.01197357,
      "auxiliary_loss_mlp": 0.01038543,
      "balance_loss_clip": 1.06046498,
      "balance_loss_mlp": 1.02987099,
      "epoch": 0.4593278422413275,
      "flos": 20631111638400.0,
      "grad_norm": 2.214277057806175,
      "language_loss": 0.80477631,
      "learning_rate": 2.3587122740857386e-06,
      "loss": 0.82713532,
      "num_input_tokens_seen": 82192600,
      "step": 3820,
      "time_per_iteration": 2.6569862365722656
    },
    {
      "auxiliary_loss_clip": 0.01177216,
      "auxiliary_loss_mlp": 0.00762986,
      "balance_loss_clip": 1.05465698,
      "balance_loss_mlp": 1.00122881,
      "epoch": 0.45944808513196655,
      "flos": 21358517961600.0,
      "grad_norm": 1.7149270722507182,
      "language_loss": 0.7833209,
      "learning_rate": 2.357945906888586e-06,
      "loss": 0.80272293,
      "num_input_tokens_seen": 82212040,
      "step": 3821,
      "time_per_iteration": 2.7432472705841064
    },
    {
      "auxiliary_loss_clip": 0.01172253,
      "auxiliary_loss_mlp": 0.01026776,
      "balance_loss_clip": 1.05338907,
      "balance_loss_mlp": 1.01740026,
      "epoch": 0.45956832802260567,
      "flos": 21427789340160.0,
      "grad_norm": 2.900322529176417,
      "language_loss": 0.80043364,
      "learning_rate": 2.357179485391324e-06,
      "loss": 0.82242388,
      "num_input_tokens_seen": 82229895,
      "step": 3822,
      "time_per_iteration": 2.679597854614258
    },
    {
      "auxiliary_loss_clip": 0.01191719,
      "auxiliary_loss_mlp": 0.0102722,
      "balance_loss_clip": 1.05769193,
      "balance_loss_mlp": 1.01822519,
      "epoch": 0.4596885709132448,
      "flos": 22382654538240.0,
      "grad_norm": 2.170025681797781,
      "language_loss": 0.86510116,
      "learning_rate": 2.3564130097102173e-06,
      "loss": 0.88729054,
      "num_input_tokens_seen": 82249550,
      "step": 3823,
      "time_per_iteration": 2.6932384967803955
    },
    {
      "auxiliary_loss_clip": 0.01179715,
      "auxiliary_loss_mlp": 0.01026266,
      "balance_loss_clip": 1.0554359,
      "balance_loss_mlp": 1.01744413,
      "epoch": 0.45980881380388383,
      "flos": 28981957806720.0,
      "grad_norm": 2.616092529546901,
      "language_loss": 0.75392878,
      "learning_rate": 2.355646479961541e-06,
      "loss": 0.77598858,
      "num_input_tokens_seen": 82268860,
      "step": 3824,
      "time_per_iteration": 2.695368528366089
    },
    {
      "auxiliary_loss_clip": 0.01192551,
      "auxiliary_loss_mlp": 0.01037996,
      "balance_loss_clip": 1.05752468,
      "balance_loss_mlp": 1.0290494,
      "epoch": 0.45992905669452294,
      "flos": 33396599980800.0,
      "grad_norm": 1.8673999009412017,
      "language_loss": 0.71831274,
      "learning_rate": 2.354879896261576e-06,
      "loss": 0.74061823,
      "num_input_tokens_seen": 82289070,
      "step": 3825,
      "time_per_iteration": 2.7293436527252197
    },
    {
      "auxiliary_loss_clip": 0.01136785,
      "auxiliary_loss_mlp": 0.01034352,
      "balance_loss_clip": 1.05139482,
      "balance_loss_mlp": 1.02499413,
      "epoch": 0.46004929958516205,
      "flos": 36318184502400.0,
      "grad_norm": 2.613615232071332,
      "language_loss": 0.56562543,
      "learning_rate": 2.3541132587266133e-06,
      "loss": 0.58733678,
      "num_input_tokens_seen": 82311790,
      "step": 3826,
      "time_per_iteration": 2.792416572570801
    },
    {
      "auxiliary_loss_clip": 0.01159144,
      "auxiliary_loss_mlp": 0.01032829,
      "balance_loss_clip": 1.05384696,
      "balance_loss_mlp": 1.02354884,
      "epoch": 0.4601695424758011,
      "flos": 17238451224960.0,
      "grad_norm": 1.8160207412551743,
      "language_loss": 0.69438148,
      "learning_rate": 2.3533465674729515e-06,
      "loss": 0.7163012,
      "num_input_tokens_seen": 82329020,
      "step": 3827,
      "time_per_iteration": 2.6863036155700684
    },
    {
      "auxiliary_loss_clip": 0.01193626,
      "auxiliary_loss_mlp": 0.01029532,
      "balance_loss_clip": 1.05871487,
      "balance_loss_mlp": 1.02079439,
      "epoch": 0.4602897853664402,
      "flos": 15888425529600.0,
      "grad_norm": 1.9938891473084877,
      "language_loss": 0.73290515,
      "learning_rate": 2.352579822616895e-06,
      "loss": 0.75513673,
      "num_input_tokens_seen": 82346455,
      "step": 3828,
      "time_per_iteration": 2.646028757095337
    },
    {
      "auxiliary_loss_clip": 0.01179687,
      "auxiliary_loss_mlp": 0.01029256,
      "balance_loss_clip": 1.0579524,
      "balance_loss_mlp": 1.02054214,
      "epoch": 0.4604100282570793,
      "flos": 25412617370880.0,
      "grad_norm": 1.9260404480246438,
      "language_loss": 0.78198528,
      "learning_rate": 2.351813024274761e-06,
      "loss": 0.8040747,
      "num_input_tokens_seen": 82367810,
      "step": 3829,
      "time_per_iteration": 2.731769323348999
    },
    {
      "auxiliary_loss_clip": 0.01179254,
      "auxiliary_loss_mlp": 0.01030761,
      "balance_loss_clip": 1.05901372,
      "balance_loss_mlp": 1.02258301,
      "epoch": 0.4605302711477184,
      "flos": 27630711048960.0,
      "grad_norm": 2.0238772758196544,
      "language_loss": 0.73826599,
      "learning_rate": 2.3510461725628693e-06,
      "loss": 0.76036614,
      "num_input_tokens_seen": 82388275,
      "step": 3830,
      "time_per_iteration": 2.7598838806152344
    },
    {
      "auxiliary_loss_clip": 0.01190319,
      "auxiliary_loss_mlp": 0.01028557,
      "balance_loss_clip": 1.05675673,
      "balance_loss_mlp": 1.0197475,
      "epoch": 0.4606505140383575,
      "flos": 23839657914240.0,
      "grad_norm": 1.9548887670652455,
      "language_loss": 0.71647239,
      "learning_rate": 2.350279267597554e-06,
      "loss": 0.73866117,
      "num_input_tokens_seen": 82408915,
      "step": 3831,
      "time_per_iteration": 2.6933913230895996
    },
    {
      "auxiliary_loss_clip": 0.01179923,
      "auxiliary_loss_mlp": 0.01030901,
      "balance_loss_clip": 1.05855393,
      "balance_loss_mlp": 1.02130437,
      "epoch": 0.46077075692899655,
      "flos": 16107013745280.0,
      "grad_norm": 4.397875884380476,
      "language_loss": 0.83427072,
      "learning_rate": 2.3495123094951515e-06,
      "loss": 0.85637891,
      "num_input_tokens_seen": 82427260,
      "step": 3832,
      "time_per_iteration": 3.655311346054077
    },
    {
      "auxiliary_loss_clip": 0.01159329,
      "auxiliary_loss_mlp": 0.00762667,
      "balance_loss_clip": 1.05426455,
      "balance_loss_mlp": 1.00089908,
      "epoch": 0.46089099981963566,
      "flos": 48798147634560.0,
      "grad_norm": 2.253018483435096,
      "language_loss": 0.76122963,
      "learning_rate": 2.34874529837201e-06,
      "loss": 0.78044963,
      "num_input_tokens_seen": 82450805,
      "step": 3833,
      "time_per_iteration": 3.8595492839813232
    },
    {
      "auxiliary_loss_clip": 0.0116775,
      "auxiliary_loss_mlp": 0.01035501,
      "balance_loss_clip": 1.05281317,
      "balance_loss_mlp": 1.02634621,
      "epoch": 0.46101124271027477,
      "flos": 19099234362240.0,
      "grad_norm": 2.1106582955599618,
      "language_loss": 0.79175138,
      "learning_rate": 2.347978234344483e-06,
      "loss": 0.81378388,
      "num_input_tokens_seen": 82467010,
      "step": 3834,
      "time_per_iteration": 2.658766269683838
    },
    {
      "auxiliary_loss_clip": 0.01196655,
      "auxiliary_loss_mlp": 0.01036425,
      "balance_loss_clip": 1.05874515,
      "balance_loss_mlp": 1.0270133,
      "epoch": 0.4611314856009138,
      "flos": 39347931853440.0,
      "grad_norm": 1.857904962343266,
      "language_loss": 0.69344491,
      "learning_rate": 2.347211117528935e-06,
      "loss": 0.71577573,
      "num_input_tokens_seen": 82489310,
      "step": 3835,
      "time_per_iteration": 2.81950306892395
    },
    {
      "auxiliary_loss_clip": 0.01199564,
      "auxiliary_loss_mlp": 0.01033256,
      "balance_loss_clip": 1.06408691,
      "balance_loss_mlp": 1.0241189,
      "epoch": 0.46125172849155294,
      "flos": 20810772489600.0,
      "grad_norm": 1.6934189729900302,
      "language_loss": 0.71836257,
      "learning_rate": 2.3464439480417374e-06,
      "loss": 0.74069071,
      "num_input_tokens_seen": 82508830,
      "step": 3836,
      "time_per_iteration": 2.6473453044891357
    },
    {
      "auxiliary_loss_clip": 0.01185996,
      "auxiliary_loss_mlp": 0.01032066,
      "balance_loss_clip": 1.0571233,
      "balance_loss_mlp": 1.02229047,
      "epoch": 0.46137197138219205,
      "flos": 17930808852480.0,
      "grad_norm": 2.9678872885636434,
      "language_loss": 0.76973021,
      "learning_rate": 2.3456767259992676e-06,
      "loss": 0.79191083,
      "num_input_tokens_seen": 82526475,
      "step": 3837,
      "time_per_iteration": 2.6544148921966553
    },
    {
      "auxiliary_loss_clip": 0.01183327,
      "auxiliary_loss_mlp": 0.01039392,
      "balance_loss_clip": 1.05473077,
      "balance_loss_mlp": 1.03042722,
      "epoch": 0.4614922142728311,
      "flos": 16836610798080.0,
      "grad_norm": 3.5060988667380344,
      "language_loss": 0.88839734,
      "learning_rate": 2.3449094515179135e-06,
      "loss": 0.91062462,
      "num_input_tokens_seen": 82543935,
      "step": 3838,
      "time_per_iteration": 2.729320764541626
    },
    {
      "auxiliary_loss_clip": 0.01169359,
      "auxiliary_loss_mlp": 0.0103276,
      "balance_loss_clip": 1.05319738,
      "balance_loss_mlp": 1.02346134,
      "epoch": 0.4616124571634702,
      "flos": 26614906427520.0,
      "grad_norm": 2.311619782064255,
      "language_loss": 0.81745005,
      "learning_rate": 2.34414212471407e-06,
      "loss": 0.83947122,
      "num_input_tokens_seen": 82563730,
      "step": 3839,
      "time_per_iteration": 2.725602388381958
    },
    {
      "auxiliary_loss_clip": 0.01183399,
      "auxiliary_loss_mlp": 0.01033013,
      "balance_loss_clip": 1.05479813,
      "balance_loss_mlp": 1.02329111,
      "epoch": 0.4617327000541093,
      "flos": 20340127560960.0,
      "grad_norm": 2.8750964967660417,
      "language_loss": 0.7304163,
      "learning_rate": 2.3433747457041394e-06,
      "loss": 0.7525804,
      "num_input_tokens_seen": 82582435,
      "step": 3840,
      "time_per_iteration": 3.653162717819214
    },
    {
      "auxiliary_loss_clip": 0.0119632,
      "auxiliary_loss_mlp": 0.01030868,
      "balance_loss_clip": 1.05875564,
      "balance_loss_mlp": 1.0212239,
      "epoch": 0.4618529429447484,
      "flos": 29570749545600.0,
      "grad_norm": 1.8058722311208029,
      "language_loss": 0.85050702,
      "learning_rate": 2.342607314604533e-06,
      "loss": 0.87277889,
      "num_input_tokens_seen": 82602185,
      "step": 3841,
      "time_per_iteration": 2.678453207015991
    },
    {
      "auxiliary_loss_clip": 0.01195751,
      "auxiliary_loss_mlp": 0.01030299,
      "balance_loss_clip": 1.05899858,
      "balance_loss_mlp": 1.02094126,
      "epoch": 0.4619731858353875,
      "flos": 19787030962560.0,
      "grad_norm": 1.9714323431580545,
      "language_loss": 0.84288204,
      "learning_rate": 2.3418398315316694e-06,
      "loss": 0.86514258,
      "num_input_tokens_seen": 82620005,
      "step": 3842,
      "time_per_iteration": 2.6998772621154785
    },
    {
      "auxiliary_loss_clip": 0.01158508,
      "auxiliary_loss_mlp": 0.01033224,
      "balance_loss_clip": 1.05206966,
      "balance_loss_mlp": 1.02359176,
      "epoch": 0.4620934287260266,
      "flos": 18951138587520.0,
      "grad_norm": 3.220539445260332,
      "language_loss": 0.78712642,
      "learning_rate": 2.3410722966019755e-06,
      "loss": 0.80904377,
      "num_input_tokens_seen": 82635120,
      "step": 3843,
      "time_per_iteration": 2.5984737873077393
    },
    {
      "auxiliary_loss_clip": 0.01191732,
      "auxiliary_loss_mlp": 0.01037067,
      "balance_loss_clip": 1.05584955,
      "balance_loss_mlp": 1.02729726,
      "epoch": 0.46221367161666566,
      "flos": 37341674634240.0,
      "grad_norm": 1.886476869771166,
      "language_loss": 0.65605485,
      "learning_rate": 2.3403047099318848e-06,
      "loss": 0.67834282,
      "num_input_tokens_seen": 82659190,
      "step": 3844,
      "time_per_iteration": 2.7648227214813232
    },
    {
      "auxiliary_loss_clip": 0.0119143,
      "auxiliary_loss_mlp": 0.01034924,
      "balance_loss_clip": 1.05603313,
      "balance_loss_mlp": 1.02612638,
      "epoch": 0.46233391450730477,
      "flos": 14428549065600.0,
      "grad_norm": 3.5551041886444255,
      "language_loss": 0.75798386,
      "learning_rate": 2.3395370716378405e-06,
      "loss": 0.78024745,
      "num_input_tokens_seen": 82676635,
      "step": 3845,
      "time_per_iteration": 2.620084285736084
    },
    {
      "auxiliary_loss_clip": 0.01192906,
      "auxiliary_loss_mlp": 0.00762857,
      "balance_loss_clip": 1.05722523,
      "balance_loss_mlp": 1.00097919,
      "epoch": 0.4624541573979438,
      "flos": 22493044010880.0,
      "grad_norm": 2.9061215965146836,
      "language_loss": 0.72212124,
      "learning_rate": 2.338769381836292e-06,
      "loss": 0.74167883,
      "num_input_tokens_seen": 82696245,
      "step": 3846,
      "time_per_iteration": 2.6809535026550293
    },
    {
      "auxiliary_loss_clip": 0.01138589,
      "auxiliary_loss_mlp": 0.01037845,
      "balance_loss_clip": 1.05032539,
      "balance_loss_mlp": 1.02852321,
      "epoch": 0.46257440028858293,
      "flos": 14465070218880.0,
      "grad_norm": 2.099778770817763,
      "language_loss": 0.73696458,
      "learning_rate": 2.3380016406436984e-06,
      "loss": 0.75872886,
      "num_input_tokens_seen": 82713725,
      "step": 3847,
      "time_per_iteration": 2.694362163543701
    },
    {
      "auxiliary_loss_clip": 0.01196693,
      "auxiliary_loss_mlp": 0.01031348,
      "balance_loss_clip": 1.05924416,
      "balance_loss_mlp": 1.02283704,
      "epoch": 0.46269464317922204,
      "flos": 23332204523520.0,
      "grad_norm": 4.5883870177941795,
      "language_loss": 0.81286955,
      "learning_rate": 2.337233848176524e-06,
      "loss": 0.83514994,
      "num_input_tokens_seen": 82731495,
      "step": 3848,
      "time_per_iteration": 2.7270750999450684
    },
    {
      "auxiliary_loss_clip": 0.01186382,
      "auxiliary_loss_mlp": 0.01027919,
      "balance_loss_clip": 1.0583241,
      "balance_loss_mlp": 1.01826894,
      "epoch": 0.4628148860698611,
      "flos": 18552027594240.0,
      "grad_norm": 1.8518170195828798,
      "language_loss": 0.83126146,
      "learning_rate": 2.3364660045512435e-06,
      "loss": 0.85340446,
      "num_input_tokens_seen": 82750255,
      "step": 3849,
      "time_per_iteration": 2.689089059829712
    },
    {
      "auxiliary_loss_clip": 0.01076839,
      "auxiliary_loss_mlp": 0.01002108,
      "balance_loss_clip": 1.01969171,
      "balance_loss_mlp": 1.00071907,
      "epoch": 0.4629351289605002,
      "flos": 70667569670400.0,
      "grad_norm": 0.7547193282110121,
      "language_loss": 0.5819205,
      "learning_rate": 2.335698109884337e-06,
      "loss": 0.60271001,
      "num_input_tokens_seen": 82815460,
      "step": 3850,
      "time_per_iteration": 3.3963096141815186
    },
    {
      "auxiliary_loss_clip": 0.01084438,
      "auxiliary_loss_mlp": 0.0101192,
      "balance_loss_clip": 1.03229308,
      "balance_loss_mlp": 1.01068592,
      "epoch": 0.4630553718511393,
      "flos": 59687200465920.0,
      "grad_norm": 0.7914204556068111,
      "language_loss": 0.59926116,
      "learning_rate": 2.334930164292294e-06,
      "loss": 0.62022471,
      "num_input_tokens_seen": 82878010,
      "step": 3851,
      "time_per_iteration": 3.3577651977539062
    },
    {
      "auxiliary_loss_clip": 0.01170838,
      "auxiliary_loss_mlp": 0.010335,
      "balance_loss_clip": 1.0554409,
      "balance_loss_mlp": 1.02485776,
      "epoch": 0.4631756147417784,
      "flos": 15960605909760.0,
      "grad_norm": 1.8841058962699426,
      "language_loss": 0.7960428,
      "learning_rate": 2.334162167891612e-06,
      "loss": 0.81808615,
      "num_input_tokens_seen": 82895275,
      "step": 3852,
      "time_per_iteration": 2.664501667022705
    },
    {
      "auxiliary_loss_clip": 0.01151792,
      "auxiliary_loss_mlp": 0.01033742,
      "balance_loss_clip": 1.04950786,
      "balance_loss_mlp": 1.02396691,
      "epoch": 0.4632958576324175,
      "flos": 16472907636480.0,
      "grad_norm": 2.028429069034793,
      "language_loss": 0.75347674,
      "learning_rate": 2.333394120798795e-06,
      "loss": 0.77533209,
      "num_input_tokens_seen": 82914010,
      "step": 3853,
      "time_per_iteration": 2.7479350566864014
    },
    {
      "auxiliary_loss_clip": 0.01190739,
      "auxiliary_loss_mlp": 0.01027348,
      "balance_loss_clip": 1.05622864,
      "balance_loss_mlp": 1.01851439,
      "epoch": 0.4634161005230566,
      "flos": 22346492520960.0,
      "grad_norm": 2.9968320798286947,
      "language_loss": 0.7247479,
      "learning_rate": 2.3326260231303545e-06,
      "loss": 0.74692881,
      "num_input_tokens_seen": 82932610,
      "step": 3854,
      "time_per_iteration": 2.657308578491211
    },
    {
      "auxiliary_loss_clip": 0.01173933,
      "auxiliary_loss_mlp": 0.01027989,
      "balance_loss_clip": 1.05689085,
      "balance_loss_mlp": 1.01951885,
      "epoch": 0.46353634341369565,
      "flos": 15742233175680.0,
      "grad_norm": 1.9171676013046137,
      "language_loss": 0.87002599,
      "learning_rate": 2.331857875002811e-06,
      "loss": 0.89204514,
      "num_input_tokens_seen": 82951210,
      "step": 3855,
      "time_per_iteration": 2.6692872047424316
    },
    {
      "auxiliary_loss_clip": 0.01194185,
      "auxiliary_loss_mlp": 0.01033462,
      "balance_loss_clip": 1.06050158,
      "balance_loss_mlp": 1.02421689,
      "epoch": 0.46365658630433476,
      "flos": 28329820433280.0,
      "grad_norm": 1.6160517682556679,
      "language_loss": 0.76136571,
      "learning_rate": 2.3310896765326916e-06,
      "loss": 0.78364217,
      "num_input_tokens_seen": 82972210,
      "step": 3856,
      "time_per_iteration": 2.6825122833251953
    },
    {
      "auxiliary_loss_clip": 0.01182458,
      "auxiliary_loss_mlp": 0.01033837,
      "balance_loss_clip": 1.05674255,
      "balance_loss_mlp": 1.0248313,
      "epoch": 0.46377682919497387,
      "flos": 24608074590720.0,
      "grad_norm": 1.743492085417856,
      "language_loss": 0.84424233,
      "learning_rate": 2.330321427836531e-06,
      "loss": 0.86640525,
      "num_input_tokens_seen": 82994080,
      "step": 3857,
      "time_per_iteration": 2.732937812805176
    },
    {
      "auxiliary_loss_clip": 0.01171337,
      "auxiliary_loss_mlp": 0.01032921,
      "balance_loss_clip": 1.05834067,
      "balance_loss_mlp": 1.02434993,
      "epoch": 0.4638970720856129,
      "flos": 19060953442560.0,
      "grad_norm": 1.8590644072105238,
      "language_loss": 0.82963133,
      "learning_rate": 2.3295531290308733e-06,
      "loss": 0.85167396,
      "num_input_tokens_seen": 83012230,
      "step": 3858,
      "time_per_iteration": 3.6780779361724854
    },
    {
      "auxiliary_loss_clip": 0.01199875,
      "auxiliary_loss_mlp": 0.01037414,
      "balance_loss_clip": 1.06165886,
      "balance_loss_mlp": 1.02770424,
      "epoch": 0.46401731497625204,
      "flos": 18471012468480.0,
      "grad_norm": 2.9132660210782886,
      "language_loss": 0.75687635,
      "learning_rate": 2.3287847802322678e-06,
      "loss": 0.77924931,
      "num_input_tokens_seen": 83027800,
      "step": 3859,
      "time_per_iteration": 3.502215623855591
    },
    {
      "auxiliary_loss_clip": 0.01199602,
      "auxiliary_loss_mlp": 0.01035206,
      "balance_loss_clip": 1.05990732,
      "balance_loss_mlp": 1.02547288,
      "epoch": 0.4641375578668911,
      "flos": 26067053214720.0,
      "grad_norm": 4.984118385733637,
      "language_loss": 0.84043729,
      "learning_rate": 2.3280163815572723e-06,
      "loss": 0.86278546,
      "num_input_tokens_seen": 83048395,
      "step": 3860,
      "time_per_iteration": 2.735954999923706
    },
    {
      "auxiliary_loss_clip": 0.01175047,
      "auxiliary_loss_mlp": 0.01029992,
      "balance_loss_clip": 1.05585003,
      "balance_loss_mlp": 1.02115238,
      "epoch": 0.4642578007575302,
      "flos": 19570382081280.0,
      "grad_norm": 3.1208377838778967,
      "language_loss": 0.77397019,
      "learning_rate": 2.3272479331224522e-06,
      "loss": 0.79602057,
      "num_input_tokens_seen": 83065825,
      "step": 3861,
      "time_per_iteration": 2.60920786857605
    },
    {
      "auxiliary_loss_clip": 0.01192973,
      "auxiliary_loss_mlp": 0.01026295,
      "balance_loss_clip": 1.05644751,
      "balance_loss_mlp": 1.0177536,
      "epoch": 0.4643780436481693,
      "flos": 28186249772160.0,
      "grad_norm": 1.9426178011076223,
      "language_loss": 0.78079391,
      "learning_rate": 2.3264794350443817e-06,
      "loss": 0.80298662,
      "num_input_tokens_seen": 83087920,
      "step": 3862,
      "time_per_iteration": 2.6828441619873047
    },
    {
      "auxiliary_loss_clip": 0.0117955,
      "auxiliary_loss_mlp": 0.01034534,
      "balance_loss_clip": 1.05312979,
      "balance_loss_mlp": 1.02555192,
      "epoch": 0.46449828653880837,
      "flos": 25375270204800.0,
      "grad_norm": 3.328722406841546,
      "language_loss": 0.79024053,
      "learning_rate": 2.3257108874396396e-06,
      "loss": 0.81238139,
      "num_input_tokens_seen": 83109015,
      "step": 3863,
      "time_per_iteration": 2.716987133026123
    },
    {
      "auxiliary_loss_clip": 0.011659,
      "auxiliary_loss_mlp": 0.00763841,
      "balance_loss_clip": 1.05299056,
      "balance_loss_mlp": 1.00087082,
      "epoch": 0.4646185294294475,
      "flos": 16034330574720.0,
      "grad_norm": 2.2869494478603998,
      "language_loss": 0.73583972,
      "learning_rate": 2.3249422904248152e-06,
      "loss": 0.75513709,
      "num_input_tokens_seen": 83127450,
      "step": 3864,
      "time_per_iteration": 2.701594352722168
    },
    {
      "auxiliary_loss_clip": 0.01192106,
      "auxiliary_loss_mlp": 0.01031767,
      "balance_loss_clip": 1.05742311,
      "balance_loss_mlp": 1.02252269,
      "epoch": 0.4647387723200866,
      "flos": 26363101109760.0,
      "grad_norm": 2.20878228316623,
      "language_loss": 0.87364382,
      "learning_rate": 2.324173644116504e-06,
      "loss": 0.89588261,
      "num_input_tokens_seen": 83150300,
      "step": 3865,
      "time_per_iteration": 2.739708662033081
    },
    {
      "auxiliary_loss_clip": 0.01179016,
      "auxiliary_loss_mlp": 0.0103259,
      "balance_loss_clip": 1.05424047,
      "balance_loss_mlp": 1.02343464,
      "epoch": 0.46485901521072565,
      "flos": 27160209774720.0,
      "grad_norm": 5.735387073876956,
      "language_loss": 0.81727219,
      "learning_rate": 2.3234049486313087e-06,
      "loss": 0.83938819,
      "num_input_tokens_seen": 83171750,
      "step": 3866,
      "time_per_iteration": 3.6542751789093018
    },
    {
      "auxiliary_loss_clip": 0.01188334,
      "auxiliary_loss_mlp": 0.01032343,
      "balance_loss_clip": 1.05622816,
      "balance_loss_mlp": 1.0244453,
      "epoch": 0.46497925810136476,
      "flos": 24279851088000.0,
      "grad_norm": 1.7818885940593303,
      "language_loss": 0.76133251,
      "learning_rate": 2.322636204085839e-06,
      "loss": 0.7835393,
      "num_input_tokens_seen": 83191820,
      "step": 3867,
      "time_per_iteration": 2.743474245071411
    },
    {
      "auxiliary_loss_clip": 0.01166443,
      "auxiliary_loss_mlp": 0.01034155,
      "balance_loss_clip": 1.05513597,
      "balance_loss_mlp": 1.02537525,
      "epoch": 0.46509950099200387,
      "flos": 16253134272000.0,
      "grad_norm": 2.3672718024368105,
      "language_loss": 0.79419011,
      "learning_rate": 2.3218674105967143e-06,
      "loss": 0.81619608,
      "num_input_tokens_seen": 83210085,
      "step": 3868,
      "time_per_iteration": 2.677513360977173
    },
    {
      "auxiliary_loss_clip": 0.01175538,
      "auxiliary_loss_mlp": 0.01029614,
      "balance_loss_clip": 1.05343902,
      "balance_loss_mlp": 1.02098334,
      "epoch": 0.4652197438826429,
      "flos": 23442270773760.0,
      "grad_norm": 3.707276052821359,
      "language_loss": 0.8369205,
      "learning_rate": 2.3210985682805593e-06,
      "loss": 0.85897201,
      "num_input_tokens_seen": 83231865,
      "step": 3869,
      "time_per_iteration": 2.7175307273864746
    },
    {
      "auxiliary_loss_clip": 0.0118523,
      "auxiliary_loss_mlp": 0.01032539,
      "balance_loss_clip": 1.05876911,
      "balance_loss_mlp": 1.02314556,
      "epoch": 0.46533998677328203,
      "flos": 16216397637120.0,
      "grad_norm": 2.239515958516299,
      "language_loss": 0.68239367,
      "learning_rate": 2.320329677254007e-06,
      "loss": 0.70457137,
      "num_input_tokens_seen": 83249195,
      "step": 3870,
      "time_per_iteration": 2.6951816082000732
    },
    {
      "auxiliary_loss_clip": 0.01152534,
      "auxiliary_loss_mlp": 0.01029638,
      "balance_loss_clip": 1.05268717,
      "balance_loss_mlp": 1.02113235,
      "epoch": 0.46546022966392114,
      "flos": 21141869080320.0,
      "grad_norm": 3.4226434457828327,
      "language_loss": 0.72723693,
      "learning_rate": 2.319560737633697e-06,
      "loss": 0.7490586,
      "num_input_tokens_seen": 83267915,
      "step": 3871,
      "time_per_iteration": 2.6919538974761963
    },
    {
      "auxiliary_loss_clip": 0.01183457,
      "auxiliary_loss_mlp": 0.01033342,
      "balance_loss_clip": 1.05691314,
      "balance_loss_mlp": 1.02399051,
      "epoch": 0.4655804725545602,
      "flos": 41171942442240.0,
      "grad_norm": 1.653675002063202,
      "language_loss": 0.68448418,
      "learning_rate": 2.3187917495362775e-06,
      "loss": 0.70665216,
      "num_input_tokens_seen": 83292325,
      "step": 3872,
      "time_per_iteration": 2.8762764930725098
    },
    {
      "auxiliary_loss_clip": 0.01177446,
      "auxiliary_loss_mlp": 0.01033149,
      "balance_loss_clip": 1.05899715,
      "balance_loss_mlp": 1.0246253,
      "epoch": 0.4657007154451993,
      "flos": 19570956698880.0,
      "grad_norm": 3.484443547564193,
      "language_loss": 0.76798522,
      "learning_rate": 2.318022713078403e-06,
      "loss": 0.79009116,
      "num_input_tokens_seen": 83306905,
      "step": 3873,
      "time_per_iteration": 2.6355881690979004
    },
    {
      "auxiliary_loss_clip": 0.01189233,
      "auxiliary_loss_mlp": 0.01037425,
      "balance_loss_clip": 1.05469728,
      "balance_loss_mlp": 1.02900243,
      "epoch": 0.4658209583358384,
      "flos": 15517826956800.0,
      "grad_norm": 2.57195691690964,
      "language_loss": 0.85315895,
      "learning_rate": 2.3172536283767354e-06,
      "loss": 0.87542552,
      "num_input_tokens_seen": 83320665,
      "step": 3874,
      "time_per_iteration": 2.668882369995117
    },
    {
      "auxiliary_loss_clip": 0.01194443,
      "auxiliary_loss_mlp": 0.01031561,
      "balance_loss_clip": 1.05985689,
      "balance_loss_mlp": 1.02263248,
      "epoch": 0.4659412012264775,
      "flos": 14903180403840.0,
      "grad_norm": 7.176804823338437,
      "language_loss": 0.81535208,
      "learning_rate": 2.3164844955479447e-06,
      "loss": 0.83761215,
      "num_input_tokens_seen": 83336475,
      "step": 3875,
      "time_per_iteration": 2.626044750213623
    },
    {
      "auxiliary_loss_clip": 0.01182444,
      "auxiliary_loss_mlp": 0.01032593,
      "balance_loss_clip": 1.0585103,
      "balance_loss_mlp": 1.02334201,
      "epoch": 0.4660614441171166,
      "flos": 24425612478720.0,
      "grad_norm": 1.7575030458392307,
      "language_loss": 0.70902812,
      "learning_rate": 2.3157153147087082e-06,
      "loss": 0.73117852,
      "num_input_tokens_seen": 83358365,
      "step": 3876,
      "time_per_iteration": 2.7362215518951416
    },
    {
      "auxiliary_loss_clip": 0.01195093,
      "auxiliary_loss_mlp": 0.01032087,
      "balance_loss_clip": 1.05933261,
      "balance_loss_mlp": 1.02295256,
      "epoch": 0.46618168700775564,
      "flos": 22091095843200.0,
      "grad_norm": 1.8427017963696661,
      "language_loss": 0.83458376,
      "learning_rate": 2.314946085975709e-06,
      "loss": 0.85685557,
      "num_input_tokens_seen": 83377345,
      "step": 3877,
      "time_per_iteration": 2.627753734588623
    },
    {
      "auxiliary_loss_clip": 0.01193022,
      "auxiliary_loss_mlp": 0.01035182,
      "balance_loss_clip": 1.0603869,
      "balance_loss_mlp": 1.02681971,
      "epoch": 0.46630192989839475,
      "flos": 26176975810560.0,
      "grad_norm": 2.1911422388556985,
      "language_loss": 0.82411659,
      "learning_rate": 2.3141768094656393e-06,
      "loss": 0.84639871,
      "num_input_tokens_seen": 83395920,
      "step": 3878,
      "time_per_iteration": 2.8406412601470947
    },
    {
      "auxiliary_loss_clip": 0.01165287,
      "auxiliary_loss_mlp": 0.01035884,
      "balance_loss_clip": 1.05484772,
      "balance_loss_mlp": 1.0277245,
      "epoch": 0.46642217278903386,
      "flos": 11509622150400.0,
      "grad_norm": 3.02934265731231,
      "language_loss": 0.83586562,
      "learning_rate": 2.3134074852951966e-06,
      "loss": 0.85787737,
      "num_input_tokens_seen": 83412510,
      "step": 3879,
      "time_per_iteration": 2.66739559173584
    },
    {
      "auxiliary_loss_clip": 0.01192434,
      "auxiliary_loss_mlp": 0.01036043,
      "balance_loss_clip": 1.06021237,
      "balance_loss_mlp": 1.02715635,
      "epoch": 0.4665424156796729,
      "flos": 32306819299200.0,
      "grad_norm": 1.8016464927918752,
      "language_loss": 0.78005415,
      "learning_rate": 2.312638113581088e-06,
      "loss": 0.8023389,
      "num_input_tokens_seen": 83432995,
      "step": 3880,
      "time_per_iteration": 2.7618587017059326
    },
    {
      "auxiliary_loss_clip": 0.01189784,
      "auxiliary_loss_mlp": 0.01028551,
      "balance_loss_clip": 1.05424309,
      "balance_loss_mlp": 1.01992631,
      "epoch": 0.46666265857031203,
      "flos": 18436179254400.0,
      "grad_norm": 3.7501700118317243,
      "language_loss": 0.78703398,
      "learning_rate": 2.311868694440027e-06,
      "loss": 0.80921733,
      "num_input_tokens_seen": 83447415,
      "step": 3881,
      "time_per_iteration": 2.6323018074035645
    },
    {
      "auxiliary_loss_clip": 0.01071607,
      "auxiliary_loss_mlp": 0.01001537,
      "balance_loss_clip": 1.01781368,
      "balance_loss_mlp": 1.0001061,
      "epoch": 0.46678290146095114,
      "flos": 68438989221120.0,
      "grad_norm": 0.7366682111852433,
      "language_loss": 0.62386823,
      "learning_rate": 2.3110992279887323e-06,
      "loss": 0.64459968,
      "num_input_tokens_seen": 83519340,
      "step": 3882,
      "time_per_iteration": 3.337390661239624
    },
    {
      "auxiliary_loss_clip": 0.01192821,
      "auxiliary_loss_mlp": 0.01041373,
      "balance_loss_clip": 1.05563307,
      "balance_loss_mlp": 1.03206277,
      "epoch": 0.4669031443515902,
      "flos": 17712507945600.0,
      "grad_norm": 3.044077083912898,
      "language_loss": 0.84913182,
      "learning_rate": 2.310329714343932e-06,
      "loss": 0.87147379,
      "num_input_tokens_seen": 83535490,
      "step": 3883,
      "time_per_iteration": 2.686206340789795
    },
    {
      "auxiliary_loss_clip": 0.01193742,
      "auxiliary_loss_mlp": 0.00762877,
      "balance_loss_clip": 1.05920994,
      "balance_loss_mlp": 1.00097871,
      "epoch": 0.4670233872422293,
      "flos": 23947748916480.0,
      "grad_norm": 1.985141794904474,
      "language_loss": 0.8237865,
      "learning_rate": 2.309560153622361e-06,
      "loss": 0.84335268,
      "num_input_tokens_seen": 83552400,
      "step": 3884,
      "time_per_iteration": 3.6281239986419678
    },
    {
      "auxiliary_loss_clip": 0.01158885,
      "auxiliary_loss_mlp": 0.01032467,
      "balance_loss_clip": 1.05175793,
      "balance_loss_mlp": 1.02219081,
      "epoch": 0.4671436301328684,
      "flos": 28111268131200.0,
      "grad_norm": 3.3638656073309847,
      "language_loss": 0.74921811,
      "learning_rate": 2.3087905459407602e-06,
      "loss": 0.77113163,
      "num_input_tokens_seen": 83571340,
      "step": 3885,
      "time_per_iteration": 3.7310268878936768
    },
    {
      "auxiliary_loss_clip": 0.01071477,
      "auxiliary_loss_mlp": 0.01000761,
      "balance_loss_clip": 1.01406944,
      "balance_loss_mlp": 0.99927729,
      "epoch": 0.46726387302350747,
      "flos": 69369684566400.0,
      "grad_norm": 0.7905337697756966,
      "language_loss": 0.62842429,
      "learning_rate": 2.3080208914158795e-06,
      "loss": 0.64914668,
      "num_input_tokens_seen": 83634340,
      "step": 3886,
      "time_per_iteration": 3.289712905883789
    },
    {
      "auxiliary_loss_clip": 0.01182468,
      "auxiliary_loss_mlp": 0.01030993,
      "balance_loss_clip": 1.05691242,
      "balance_loss_mlp": 1.02220714,
      "epoch": 0.4673841159141466,
      "flos": 25519666878720.0,
      "grad_norm": 4.198519840972746,
      "language_loss": 0.72654986,
      "learning_rate": 2.3072511901644753e-06,
      "loss": 0.74868447,
      "num_input_tokens_seen": 83653410,
      "step": 3887,
      "time_per_iteration": 2.6499297618865967
    },
    {
      "auxiliary_loss_clip": 0.01176253,
      "auxiliary_loss_mlp": 0.01034679,
      "balance_loss_clip": 1.05736852,
      "balance_loss_mlp": 1.02604222,
      "epoch": 0.4675043588047857,
      "flos": 24499265316480.0,
      "grad_norm": 1.869197711238631,
      "language_loss": 0.80965596,
      "learning_rate": 2.306481442303309e-06,
      "loss": 0.83176529,
      "num_input_tokens_seen": 83672985,
      "step": 3888,
      "time_per_iteration": 2.751302480697632
    },
    {
      "auxiliary_loss_clip": 0.01191656,
      "auxiliary_loss_mlp": 0.01033483,
      "balance_loss_clip": 1.05563021,
      "balance_loss_mlp": 1.02436388,
      "epoch": 0.46762460169542475,
      "flos": 20960771685120.0,
      "grad_norm": 1.8383868783206336,
      "language_loss": 0.73484886,
      "learning_rate": 2.3057116479491515e-06,
      "loss": 0.75710028,
      "num_input_tokens_seen": 83692395,
      "step": 3889,
      "time_per_iteration": 2.623382806777954
    },
    {
      "auxiliary_loss_clip": 0.0119067,
      "auxiliary_loss_mlp": 0.01025774,
      "balance_loss_clip": 1.05605471,
      "balance_loss_mlp": 1.01728058,
      "epoch": 0.46774484458606386,
      "flos": 19171666137600.0,
      "grad_norm": 2.0290296917733914,
      "language_loss": 0.76240152,
      "learning_rate": 2.30494180721878e-06,
      "loss": 0.78456593,
      "num_input_tokens_seen": 83709735,
      "step": 3890,
      "time_per_iteration": 2.6648826599121094
    },
    {
      "auxiliary_loss_clip": 0.01163006,
      "auxiliary_loss_mlp": 0.0103164,
      "balance_loss_clip": 1.05633557,
      "balance_loss_mlp": 1.02352762,
      "epoch": 0.4678650874767029,
      "flos": 17967689141760.0,
      "grad_norm": 2.1255080438248224,
      "language_loss": 0.89648759,
      "learning_rate": 2.3041719202289794e-06,
      "loss": 0.91843414,
      "num_input_tokens_seen": 83725910,
      "step": 3891,
      "time_per_iteration": 2.6946041584014893
    },
    {
      "auxiliary_loss_clip": 0.01191078,
      "auxiliary_loss_mlp": 0.01027144,
      "balance_loss_clip": 1.05941677,
      "balance_loss_mlp": 1.01810837,
      "epoch": 0.467985330367342,
      "flos": 21360816432000.0,
      "grad_norm": 1.9624470643357856,
      "language_loss": 0.80697417,
      "learning_rate": 2.30340198709654e-06,
      "loss": 0.8291564,
      "num_input_tokens_seen": 83745745,
      "step": 3892,
      "time_per_iteration": 2.594919204711914
    },
    {
      "auxiliary_loss_clip": 0.01190421,
      "auxiliary_loss_mlp": 0.01029838,
      "balance_loss_clip": 1.05780268,
      "balance_loss_mlp": 1.02161872,
      "epoch": 0.46810557325798113,
      "flos": 20521835487360.0,
      "grad_norm": 2.437128685858197,
      "language_loss": 0.74158633,
      "learning_rate": 2.3026320079382605e-06,
      "loss": 0.76378894,
      "num_input_tokens_seen": 83762680,
      "step": 3893,
      "time_per_iteration": 3.5985867977142334
    },
    {
      "auxiliary_loss_clip": 0.01171398,
      "auxiliary_loss_mlp": 0.0103173,
      "balance_loss_clip": 1.05592024,
      "balance_loss_mlp": 1.02417779,
      "epoch": 0.4682258161486202,
      "flos": 30117848572800.0,
      "grad_norm": 2.2911492933996276,
      "language_loss": 0.76930982,
      "learning_rate": 2.3018619828709454e-06,
      "loss": 0.79134107,
      "num_input_tokens_seen": 83784220,
      "step": 3894,
      "time_per_iteration": 2.735471725463867
    },
    {
      "auxiliary_loss_clip": 0.01172948,
      "auxiliary_loss_mlp": 0.01033796,
      "balance_loss_clip": 1.056301,
      "balance_loss_mlp": 1.02593434,
      "epoch": 0.4683460590392593,
      "flos": 25293357239040.0,
      "grad_norm": 8.597169559800442,
      "language_loss": 0.82096303,
      "learning_rate": 2.3010919120114084e-06,
      "loss": 0.84303045,
      "num_input_tokens_seen": 83800750,
      "step": 3895,
      "time_per_iteration": 2.6498894691467285
    },
    {
      "auxiliary_loss_clip": 0.01173484,
      "auxiliary_loss_mlp": 0.01029187,
      "balance_loss_clip": 1.05397415,
      "balance_loss_mlp": 1.02041912,
      "epoch": 0.4684663019298984,
      "flos": 15368330551680.0,
      "grad_norm": 2.285412663054142,
      "language_loss": 0.66325295,
      "learning_rate": 2.3003217954764672e-06,
      "loss": 0.68527967,
      "num_input_tokens_seen": 83815455,
      "step": 3896,
      "time_per_iteration": 2.5837490558624268
    },
    {
      "auxiliary_loss_clip": 0.01180845,
      "auxiliary_loss_mlp": 0.0102656,
      "balance_loss_clip": 1.05333614,
      "balance_loss_mlp": 1.01721406,
      "epoch": 0.46858654482053747,
      "flos": 27778842737280.0,
      "grad_norm": 1.9226771231900015,
      "language_loss": 0.79202557,
      "learning_rate": 2.299551633382949e-06,
      "loss": 0.81409967,
      "num_input_tokens_seen": 83835765,
      "step": 3897,
      "time_per_iteration": 2.6995387077331543
    },
    {
      "auxiliary_loss_clip": 0.01175396,
      "auxiliary_loss_mlp": 0.01037823,
      "balance_loss_clip": 1.0561918,
      "balance_loss_mlp": 1.02854836,
      "epoch": 0.4687067877111766,
      "flos": 18040623707520.0,
      "grad_norm": 1.9060381248459564,
      "language_loss": 0.85686219,
      "learning_rate": 2.2987814258476854e-06,
      "loss": 0.87899435,
      "num_input_tokens_seen": 83853565,
      "step": 3898,
      "time_per_iteration": 2.6350014209747314
    },
    {
      "auxiliary_loss_clip": 0.01196197,
      "auxiliary_loss_mlp": 0.01032284,
      "balance_loss_clip": 1.05717969,
      "balance_loss_mlp": 1.02246165,
      "epoch": 0.4688270306018157,
      "flos": 16977380198400.0,
      "grad_norm": 3.047671544197785,
      "language_loss": 0.67980242,
      "learning_rate": 2.2980111729875177e-06,
      "loss": 0.70208728,
      "num_input_tokens_seen": 83869815,
      "step": 3899,
      "time_per_iteration": 2.6048195362091064
    },
    {
      "auxiliary_loss_clip": 0.01191567,
      "auxiliary_loss_mlp": 0.01034831,
      "balance_loss_clip": 1.05693579,
      "balance_loss_mlp": 1.02646267,
      "epoch": 0.46894727349245474,
      "flos": 17821640442240.0,
      "grad_norm": 1.6520230192495298,
      "language_loss": 0.83084089,
      "learning_rate": 2.2972408749192917e-06,
      "loss": 0.85310483,
      "num_input_tokens_seen": 83887545,
      "step": 3900,
      "time_per_iteration": 2.594728708267212
    },
    {
      "auxiliary_loss_clip": 0.01174523,
      "auxiliary_loss_mlp": 0.01034709,
      "balance_loss_clip": 1.05680823,
      "balance_loss_mlp": 1.0267036,
      "epoch": 0.46906751638309385,
      "flos": 21471349559040.0,
      "grad_norm": 2.9864390647842223,
      "language_loss": 0.66852421,
      "learning_rate": 2.296470531759861e-06,
      "loss": 0.69061649,
      "num_input_tokens_seen": 83905645,
      "step": 3901,
      "time_per_iteration": 2.66251802444458
    },
    {
      "auxiliary_loss_clip": 0.01175632,
      "auxiliary_loss_mlp": 0.01027703,
      "balance_loss_clip": 1.05408144,
      "balance_loss_mlp": 1.01895332,
      "epoch": 0.46918775927373296,
      "flos": 20337829090560.0,
      "grad_norm": 1.9766334578276366,
      "language_loss": 0.79463542,
      "learning_rate": 2.2957001436260866e-06,
      "loss": 0.81666875,
      "num_input_tokens_seen": 83922705,
      "step": 3902,
      "time_per_iteration": 2.6820483207702637
    },
    {
      "auxiliary_loss_clip": 0.01179341,
      "auxiliary_loss_mlp": 0.01032581,
      "balance_loss_clip": 1.05466938,
      "balance_loss_mlp": 1.02465701,
      "epoch": 0.469308002164372,
      "flos": 18403249461120.0,
      "grad_norm": 2.5613706347042693,
      "language_loss": 0.72974265,
      "learning_rate": 2.294929710634836e-06,
      "loss": 0.75186181,
      "num_input_tokens_seen": 83940795,
      "step": 3903,
      "time_per_iteration": 2.644685983657837
    },
    {
      "auxiliary_loss_clip": 0.01172931,
      "auxiliary_loss_mlp": 0.01034149,
      "balance_loss_clip": 1.0544138,
      "balance_loss_mlp": 1.02545881,
      "epoch": 0.46942824505501113,
      "flos": 37962067363200.0,
      "grad_norm": 1.9652710520932846,
      "language_loss": 0.61680734,
      "learning_rate": 2.2941592329029823e-06,
      "loss": 0.63887817,
      "num_input_tokens_seen": 83961900,
      "step": 3904,
      "time_per_iteration": 2.8140623569488525
    },
    {
      "auxiliary_loss_clip": 0.01191006,
      "auxiliary_loss_mlp": 0.01032378,
      "balance_loss_clip": 1.05682302,
      "balance_loss_mlp": 1.02307415,
      "epoch": 0.46954848794565024,
      "flos": 21872507627520.0,
      "grad_norm": 2.130942867825928,
      "language_loss": 0.79230356,
      "learning_rate": 2.2933887105474067e-06,
      "loss": 0.81453741,
      "num_input_tokens_seen": 83980075,
      "step": 3905,
      "time_per_iteration": 2.5627033710479736
    },
    {
      "auxiliary_loss_clip": 0.0116525,
      "auxiliary_loss_mlp": 0.01030739,
      "balance_loss_clip": 1.05301297,
      "balance_loss_mlp": 1.02273405,
      "epoch": 0.4696687308362893,
      "flos": 22016545165440.0,
      "grad_norm": 1.993498222512435,
      "language_loss": 0.81787598,
      "learning_rate": 2.2926181436849974e-06,
      "loss": 0.83983588,
      "num_input_tokens_seen": 83999430,
      "step": 3906,
      "time_per_iteration": 2.7127504348754883
    },
    {
      "auxiliary_loss_clip": 0.01174123,
      "auxiliary_loss_mlp": 0.00762588,
      "balance_loss_clip": 1.05404472,
      "balance_loss_mlp": 1.00091338,
      "epoch": 0.4697889737269284,
      "flos": 21613663244160.0,
      "grad_norm": 2.4963969327174937,
      "language_loss": 0.72751731,
      "learning_rate": 2.2918475324326478e-06,
      "loss": 0.74688447,
      "num_input_tokens_seen": 84019150,
      "step": 3907,
      "time_per_iteration": 2.6922028064727783
    },
    {
      "auxiliary_loss_clip": 0.01196082,
      "auxiliary_loss_mlp": 0.01032962,
      "balance_loss_clip": 1.05823565,
      "balance_loss_mlp": 1.02319264,
      "epoch": 0.46990921661756746,
      "flos": 25228323665280.0,
      "grad_norm": 2.617235357860012,
      "language_loss": 0.91598988,
      "learning_rate": 2.2910768769072603e-06,
      "loss": 0.93828034,
      "num_input_tokens_seen": 84037930,
      "step": 3908,
      "time_per_iteration": 2.6617860794067383
    },
    {
      "auxiliary_loss_clip": 0.01183058,
      "auxiliary_loss_mlp": 0.01036529,
      "balance_loss_clip": 1.0573647,
      "balance_loss_mlp": 1.02808881,
      "epoch": 0.47002945950820657,
      "flos": 13844031045120.0,
      "grad_norm": 2.243391352899384,
      "language_loss": 0.76409537,
      "learning_rate": 2.2903061772257417e-06,
      "loss": 0.78629124,
      "num_input_tokens_seen": 84055915,
      "step": 3909,
      "time_per_iteration": 2.662581205368042
    },
    {
      "auxiliary_loss_clip": 0.01191909,
      "auxiliary_loss_mlp": 0.01033982,
      "balance_loss_clip": 1.05778611,
      "balance_loss_mlp": 1.02469528,
      "epoch": 0.4701497023988457,
      "flos": 26247001374720.0,
      "grad_norm": 1.6807922951204237,
      "language_loss": 0.78848207,
      "learning_rate": 2.289535433505007e-06,
      "loss": 0.81074101,
      "num_input_tokens_seen": 84077270,
      "step": 3910,
      "time_per_iteration": 3.571462869644165
    },
    {
      "auxiliary_loss_clip": 0.01142973,
      "auxiliary_loss_mlp": 0.010298,
      "balance_loss_clip": 1.05173457,
      "balance_loss_mlp": 1.02119303,
      "epoch": 0.47026994528948474,
      "flos": 25629517647360.0,
      "grad_norm": 2.4870850688821875,
      "language_loss": 0.63897371,
      "learning_rate": 2.2887646458619767e-06,
      "loss": 0.66070139,
      "num_input_tokens_seen": 84098635,
      "step": 3911,
      "time_per_iteration": 3.542363166809082
    },
    {
      "auxiliary_loss_clip": 0.01176572,
      "auxiliary_loss_mlp": 0.01029899,
      "balance_loss_clip": 1.05681193,
      "balance_loss_mlp": 1.02058899,
      "epoch": 0.47039018818012385,
      "flos": 20554406144640.0,
      "grad_norm": 2.018332117772234,
      "language_loss": 0.76720071,
      "learning_rate": 2.2879938144135797e-06,
      "loss": 0.78926539,
      "num_input_tokens_seen": 84114740,
      "step": 3912,
      "time_per_iteration": 3.3928329944610596
    },
    {
      "auxiliary_loss_clip": 0.01137847,
      "auxiliary_loss_mlp": 0.01028525,
      "balance_loss_clip": 1.05118382,
      "balance_loss_mlp": 1.02064526,
      "epoch": 0.47051043107076296,
      "flos": 21577249831680.0,
      "grad_norm": 1.6693026127387398,
      "language_loss": 0.75289524,
      "learning_rate": 2.2872229392767496e-06,
      "loss": 0.77455902,
      "num_input_tokens_seen": 84134845,
      "step": 3913,
      "time_per_iteration": 2.633286476135254
    },
    {
      "auxiliary_loss_clip": 0.0117738,
      "auxiliary_loss_mlp": 0.0103117,
      "balance_loss_clip": 1.05393839,
      "balance_loss_mlp": 1.02298677,
      "epoch": 0.470630673961402,
      "flos": 18953185662720.0,
      "grad_norm": 1.9060690804702245,
      "language_loss": 0.74940431,
      "learning_rate": 2.286452020568428e-06,
      "loss": 0.7714898,
      "num_input_tokens_seen": 84152920,
      "step": 3914,
      "time_per_iteration": 2.6111631393432617
    },
    {
      "auxiliary_loss_clip": 0.01179363,
      "auxiliary_loss_mlp": 0.01037614,
      "balance_loss_clip": 1.05680084,
      "balance_loss_mlp": 1.02776706,
      "epoch": 0.4707509168520411,
      "flos": 19938969492480.0,
      "grad_norm": 2.3998284334260296,
      "language_loss": 0.73539972,
      "learning_rate": 2.2856810584055637e-06,
      "loss": 0.75756955,
      "num_input_tokens_seen": 84170455,
      "step": 3915,
      "time_per_iteration": 2.6294257640838623
    },
    {
      "auxiliary_loss_clip": 0.01191745,
      "auxiliary_loss_mlp": 0.01020282,
      "balance_loss_clip": 1.05740452,
      "balance_loss_mlp": 1.01170516,
      "epoch": 0.47087115974268023,
      "flos": 40118754741120.0,
      "grad_norm": 1.5454424033146346,
      "language_loss": 0.67841065,
      "learning_rate": 2.2849100529051085e-06,
      "loss": 0.70053089,
      "num_input_tokens_seen": 84197390,
      "step": 3916,
      "time_per_iteration": 2.8294360637664795
    },
    {
      "auxiliary_loss_clip": 0.01168215,
      "auxiliary_loss_mlp": 0.01028053,
      "balance_loss_clip": 1.05525541,
      "balance_loss_mlp": 1.0195533,
      "epoch": 0.4709914026333193,
      "flos": 13552723745280.0,
      "grad_norm": 4.830609669592732,
      "language_loss": 0.80158925,
      "learning_rate": 2.284139004184026e-06,
      "loss": 0.82355189,
      "num_input_tokens_seen": 84214620,
      "step": 3917,
      "time_per_iteration": 2.6893343925476074
    },
    {
      "auxiliary_loss_clip": 0.01193382,
      "auxiliary_loss_mlp": 0.01031819,
      "balance_loss_clip": 1.05795956,
      "balance_loss_mlp": 1.02305126,
      "epoch": 0.4711116455239584,
      "flos": 19974628719360.0,
      "grad_norm": 5.077404618114846,
      "language_loss": 0.74336231,
      "learning_rate": 2.2833679123592814e-06,
      "loss": 0.76561427,
      "num_input_tokens_seen": 84231880,
      "step": 3918,
      "time_per_iteration": 2.6866137981414795
    },
    {
      "auxiliary_loss_clip": 0.01193946,
      "auxiliary_loss_mlp": 0.0103182,
      "balance_loss_clip": 1.05809748,
      "balance_loss_mlp": 1.02284932,
      "epoch": 0.4712318884145975,
      "flos": 32124824064000.0,
      "grad_norm": 1.8013612203643938,
      "language_loss": 0.63485765,
      "learning_rate": 2.2825967775478508e-06,
      "loss": 0.65711522,
      "num_input_tokens_seen": 84252980,
      "step": 3919,
      "time_per_iteration": 3.574089765548706
    },
    {
      "auxiliary_loss_clip": 0.01176092,
      "auxiliary_loss_mlp": 0.00763125,
      "balance_loss_clip": 1.05630589,
      "balance_loss_mlp": 1.00100935,
      "epoch": 0.47135213130523657,
      "flos": 20047850593920.0,
      "grad_norm": 3.346170905541716,
      "language_loss": 0.8387773,
      "learning_rate": 2.2818255998667135e-06,
      "loss": 0.8581695,
      "num_input_tokens_seen": 84271490,
      "step": 3920,
      "time_per_iteration": 2.7192978858947754
    },
    {
      "auxiliary_loss_clip": 0.0119273,
      "auxiliary_loss_mlp": 0.01032308,
      "balance_loss_clip": 1.05771184,
      "balance_loss_mlp": 1.02330172,
      "epoch": 0.4714723741958757,
      "flos": 19426990988160.0,
      "grad_norm": 3.616964419846331,
      "language_loss": 0.7939533,
      "learning_rate": 2.2810543794328566e-06,
      "loss": 0.81620365,
      "num_input_tokens_seen": 84290525,
      "step": 3921,
      "time_per_iteration": 2.664156675338745
    },
    {
      "auxiliary_loss_clip": 0.01192531,
      "auxiliary_loss_mlp": 0.01027609,
      "balance_loss_clip": 1.05907822,
      "balance_loss_mlp": 1.01850128,
      "epoch": 0.4715926170865148,
      "flos": 20373883367040.0,
      "grad_norm": 3.1772388431493406,
      "language_loss": 0.8263166,
      "learning_rate": 2.2802831163632735e-06,
      "loss": 0.84851795,
      "num_input_tokens_seen": 84309245,
      "step": 3922,
      "time_per_iteration": 2.648434638977051
    },
    {
      "auxiliary_loss_clip": 0.01180193,
      "auxiliary_loss_mlp": 0.0103147,
      "balance_loss_clip": 1.05372691,
      "balance_loss_mlp": 1.02180195,
      "epoch": 0.47171285997715384,
      "flos": 22672884430080.0,
      "grad_norm": 2.0597510395399574,
      "language_loss": 0.74109024,
      "learning_rate": 2.279511810774965e-06,
      "loss": 0.76320684,
      "num_input_tokens_seen": 84330775,
      "step": 3923,
      "time_per_iteration": 2.6854002475738525
    },
    {
      "auxiliary_loss_clip": 0.01136145,
      "auxiliary_loss_mlp": 0.01035039,
      "balance_loss_clip": 1.04680657,
      "balance_loss_mlp": 1.02636075,
      "epoch": 0.47183310286779295,
      "flos": 21105419754240.0,
      "grad_norm": 1.9142644927232766,
      "language_loss": 0.71610558,
      "learning_rate": 2.2787404627849364e-06,
      "loss": 0.73781741,
      "num_input_tokens_seen": 84349985,
      "step": 3924,
      "time_per_iteration": 2.709580421447754
    },
    {
      "auxiliary_loss_clip": 0.01160765,
      "auxiliary_loss_mlp": 0.01032431,
      "balance_loss_clip": 1.05512261,
      "balance_loss_mlp": 1.02387178,
      "epoch": 0.471953345758432,
      "flos": 21726566668800.0,
      "grad_norm": 1.9043675790707237,
      "language_loss": 0.79543769,
      "learning_rate": 2.277969072510202e-06,
      "loss": 0.8173697,
      "num_input_tokens_seen": 84368965,
      "step": 3925,
      "time_per_iteration": 2.6771347522735596
    },
    {
      "auxiliary_loss_clip": 0.01168152,
      "auxiliary_loss_mlp": 0.01027206,
      "balance_loss_clip": 1.05356765,
      "balance_loss_mlp": 1.01873016,
      "epoch": 0.4720735886490711,
      "flos": 19861078849920.0,
      "grad_norm": 48.4264355685719,
      "language_loss": 0.81445247,
      "learning_rate": 2.2771976400677803e-06,
      "loss": 0.83640599,
      "num_input_tokens_seen": 84387795,
      "step": 3926,
      "time_per_iteration": 2.617427110671997
    },
    {
      "auxiliary_loss_clip": 0.01135682,
      "auxiliary_loss_mlp": 0.01028684,
      "balance_loss_clip": 1.04776883,
      "balance_loss_mlp": 1.02032781,
      "epoch": 0.47219383153971023,
      "flos": 19171809792000.0,
      "grad_norm": 1.8574524820840852,
      "language_loss": 0.79175436,
      "learning_rate": 2.2764261655746965e-06,
      "loss": 0.813398,
      "num_input_tokens_seen": 84405290,
      "step": 3927,
      "time_per_iteration": 2.7604119777679443
    },
    {
      "auxiliary_loss_clip": 0.01160208,
      "auxiliary_loss_mlp": 0.01035313,
      "balance_loss_clip": 1.0534265,
      "balance_loss_mlp": 1.02655149,
      "epoch": 0.4723140744303493,
      "flos": 23224005780480.0,
      "grad_norm": 1.7618606654842448,
      "language_loss": 0.75940883,
      "learning_rate": 2.2756546491479832e-06,
      "loss": 0.78136402,
      "num_input_tokens_seen": 84426205,
      "step": 3928,
      "time_per_iteration": 2.66534686088562
    },
    {
      "auxiliary_loss_clip": 0.01142453,
      "auxiliary_loss_mlp": 0.01031041,
      "balance_loss_clip": 1.05105186,
      "balance_loss_mlp": 1.02232099,
      "epoch": 0.4724343173209884,
      "flos": 18223265387520.0,
      "grad_norm": 5.167822078023388,
      "language_loss": 0.80461979,
      "learning_rate": 2.274883090904679e-06,
      "loss": 0.82635474,
      "num_input_tokens_seen": 84443970,
      "step": 3929,
      "time_per_iteration": 2.6870546340942383
    },
    {
      "auxiliary_loss_clip": 0.0116424,
      "auxiliary_loss_mlp": 0.01035178,
      "balance_loss_clip": 1.05770993,
      "balance_loss_mlp": 1.02573037,
      "epoch": 0.4725545602116275,
      "flos": 21251037490560.0,
      "grad_norm": 2.5235278714175338,
      "language_loss": 0.68155074,
      "learning_rate": 2.2741114909618283e-06,
      "loss": 0.70354491,
      "num_input_tokens_seen": 84459865,
      "step": 3930,
      "time_per_iteration": 2.7017223834991455
    },
    {
      "auxiliary_loss_clip": 0.01189549,
      "auxiliary_loss_mlp": 0.01028134,
      "balance_loss_clip": 1.05672061,
      "balance_loss_mlp": 1.01923501,
      "epoch": 0.47267480310226656,
      "flos": 21434002392960.0,
      "grad_norm": 2.0819761950247377,
      "language_loss": 0.72015166,
      "learning_rate": 2.2733398494364828e-06,
      "loss": 0.74232846,
      "num_input_tokens_seen": 84479110,
      "step": 3931,
      "time_per_iteration": 2.6568315029144287
    },
    {
      "auxiliary_loss_clip": 0.01172263,
      "auxiliary_loss_mlp": 0.01031637,
      "balance_loss_clip": 1.05309963,
      "balance_loss_mlp": 1.02272618,
      "epoch": 0.47279504599290567,
      "flos": 18770508069120.0,
      "grad_norm": 2.0260292550013825,
      "language_loss": 0.85011053,
      "learning_rate": 2.272568166445699e-06,
      "loss": 0.87214953,
      "num_input_tokens_seen": 84497675,
      "step": 3932,
      "time_per_iteration": 2.6425507068634033
    },
    {
      "auxiliary_loss_clip": 0.01191594,
      "auxiliary_loss_mlp": 0.01029394,
      "balance_loss_clip": 1.05731726,
      "balance_loss_mlp": 1.02014935,
      "epoch": 0.4729152888835448,
      "flos": 21105742976640.0,
      "grad_norm": 2.3503909244638463,
      "language_loss": 0.6504308,
      "learning_rate": 2.271796442106541e-06,
      "loss": 0.67264068,
      "num_input_tokens_seen": 84517030,
      "step": 3933,
      "time_per_iteration": 2.6494338512420654
    },
    {
      "auxiliary_loss_clip": 0.01086484,
      "auxiliary_loss_mlp": 0.01004896,
      "balance_loss_clip": 1.01948929,
      "balance_loss_mlp": 1.00360286,
      "epoch": 0.47303553177418384,
      "flos": 70201877840640.0,
      "grad_norm": 0.7984780824437253,
      "language_loss": 0.56470621,
      "learning_rate": 2.271024676536079e-06,
      "loss": 0.58562005,
      "num_input_tokens_seen": 84577290,
      "step": 3934,
      "time_per_iteration": 3.1738369464874268
    },
    {
      "auxiliary_loss_clip": 0.01197183,
      "auxiliary_loss_mlp": 0.01028969,
      "balance_loss_clip": 1.06026101,
      "balance_loss_mlp": 1.01987958,
      "epoch": 0.47315577466482295,
      "flos": 22455122227200.0,
      "grad_norm": 6.455475618423606,
      "language_loss": 0.73855257,
      "learning_rate": 2.2702528698513894e-06,
      "loss": 0.76081407,
      "num_input_tokens_seen": 84598415,
      "step": 3935,
      "time_per_iteration": 2.643977165222168
    },
    {
      "auxiliary_loss_clip": 0.01196505,
      "auxiliary_loss_mlp": 0.01033025,
      "balance_loss_clip": 1.06059718,
      "balance_loss_mlp": 1.02337539,
      "epoch": 0.47327601755546206,
      "flos": 24352857480960.0,
      "grad_norm": 2.473321791840773,
      "language_loss": 0.78707767,
      "learning_rate": 2.269481022169554e-06,
      "loss": 0.80937296,
      "num_input_tokens_seen": 84617010,
      "step": 3936,
      "time_per_iteration": 3.511284589767456
    },
    {
      "auxiliary_loss_clip": 0.01177052,
      "auxiliary_loss_mlp": 0.01034976,
      "balance_loss_clip": 1.05797338,
      "balance_loss_mlp": 1.02532613,
      "epoch": 0.4733962604461011,
      "flos": 22926772736640.0,
      "grad_norm": 1.9103732143519463,
      "language_loss": 0.80814075,
      "learning_rate": 2.2687091336076614e-06,
      "loss": 0.83026099,
      "num_input_tokens_seen": 84636350,
      "step": 3937,
      "time_per_iteration": 3.7592709064483643
    },
    {
      "auxiliary_loss_clip": 0.01164161,
      "auxiliary_loss_mlp": 0.01034159,
      "balance_loss_clip": 1.05451894,
      "balance_loss_mlp": 1.02539706,
      "epoch": 0.4735165033367402,
      "flos": 18327369980160.0,
      "grad_norm": 2.0175124725263562,
      "language_loss": 0.80155158,
      "learning_rate": 2.267937204282807e-06,
      "loss": 0.82353479,
      "num_input_tokens_seen": 84653490,
      "step": 3938,
      "time_per_iteration": 3.6283130645751953
    },
    {
      "auxiliary_loss_clip": 0.01200756,
      "auxiliary_loss_mlp": 0.01032451,
      "balance_loss_clip": 1.06118405,
      "balance_loss_mlp": 1.02265751,
      "epoch": 0.4736367462273793,
      "flos": 23037018554880.0,
      "grad_norm": 2.263809049742704,
      "language_loss": 0.79317808,
      "learning_rate": 2.2671652343120926e-06,
      "loss": 0.81551015,
      "num_input_tokens_seen": 84673965,
      "step": 3939,
      "time_per_iteration": 2.6498074531555176
    },
    {
      "auxiliary_loss_clip": 0.0116447,
      "auxiliary_loss_mlp": 0.0103147,
      "balance_loss_clip": 1.05827773,
      "balance_loss_mlp": 1.02266657,
      "epoch": 0.4737569891180184,
      "flos": 25374336451200.0,
      "grad_norm": 5.358214334052023,
      "language_loss": 0.8044548,
      "learning_rate": 2.2663932238126236e-06,
      "loss": 0.82641417,
      "num_input_tokens_seen": 84692525,
      "step": 3940,
      "time_per_iteration": 2.7156012058258057
    },
    {
      "auxiliary_loss_clip": 0.01171247,
      "auxiliary_loss_mlp": 0.01033673,
      "balance_loss_clip": 1.05421555,
      "balance_loss_mlp": 1.02494693,
      "epoch": 0.4738772320086575,
      "flos": 25849326925440.0,
      "grad_norm": 1.6038480372973498,
      "language_loss": 0.80612463,
      "learning_rate": 2.265621172901515e-06,
      "loss": 0.82817382,
      "num_input_tokens_seen": 84715640,
      "step": 3941,
      "time_per_iteration": 2.7144150733947754
    },
    {
      "auxiliary_loss_clip": 0.01185784,
      "auxiliary_loss_mlp": 0.01031572,
      "balance_loss_clip": 1.06176209,
      "balance_loss_mlp": 1.02278662,
      "epoch": 0.47399747489929656,
      "flos": 27564420499200.0,
      "grad_norm": 2.972976191242072,
      "language_loss": 0.71680939,
      "learning_rate": 2.2648490816958854e-06,
      "loss": 0.73898292,
      "num_input_tokens_seen": 84736635,
      "step": 3942,
      "time_per_iteration": 2.700252056121826
    },
    {
      "auxiliary_loss_clip": 0.01194305,
      "auxiliary_loss_mlp": 0.01033575,
      "balance_loss_clip": 1.05755639,
      "balance_loss_mlp": 1.02458692,
      "epoch": 0.47411771778993567,
      "flos": 24863650836480.0,
      "grad_norm": 3.0394820539437757,
      "language_loss": 0.7348628,
      "learning_rate": 2.264076950312861e-06,
      "loss": 0.75714165,
      "num_input_tokens_seen": 84755445,
      "step": 3943,
      "time_per_iteration": 2.7097361087799072
    },
    {
      "auxiliary_loss_clip": 0.01195211,
      "auxiliary_loss_mlp": 0.0102753,
      "balance_loss_clip": 1.05805731,
      "balance_loss_mlp": 1.01801741,
      "epoch": 0.4742379606805748,
      "flos": 22748009725440.0,
      "grad_norm": 1.9091109851134782,
      "language_loss": 0.82278281,
      "learning_rate": 2.2633047788695727e-06,
      "loss": 0.84501028,
      "num_input_tokens_seen": 84775750,
      "step": 3944,
      "time_per_iteration": 2.6280174255371094
    },
    {
      "auxiliary_loss_clip": 0.01194102,
      "auxiliary_loss_mlp": 0.01034763,
      "balance_loss_clip": 1.06037402,
      "balance_loss_mlp": 1.025769,
      "epoch": 0.47435820357121383,
      "flos": 19681130689920.0,
      "grad_norm": 1.8245195124378337,
      "language_loss": 0.64205664,
      "learning_rate": 2.262532567483159e-06,
      "loss": 0.66434538,
      "num_input_tokens_seen": 84794310,
      "step": 3945,
      "time_per_iteration": 4.236657381057739
    },
    {
      "auxiliary_loss_clip": 0.01196378,
      "auxiliary_loss_mlp": 0.01032141,
      "balance_loss_clip": 1.0605458,
      "balance_loss_mlp": 1.02272332,
      "epoch": 0.47447844646185294,
      "flos": 25228718714880.0,
      "grad_norm": 2.3360356818298422,
      "language_loss": 0.8052969,
      "learning_rate": 2.2617603162707635e-06,
      "loss": 0.82758212,
      "num_input_tokens_seen": 84814720,
      "step": 3946,
      "time_per_iteration": 2.6707122325897217
    },
    {
      "auxiliary_loss_clip": 0.01180115,
      "auxiliary_loss_mlp": 0.01039802,
      "balance_loss_clip": 1.06140113,
      "balance_loss_mlp": 1.03030658,
      "epoch": 0.47459868935249205,
      "flos": 24570619683840.0,
      "grad_norm": 2.0120887413608024,
      "language_loss": 0.82722282,
      "learning_rate": 2.2609880253495363e-06,
      "loss": 0.84942198,
      "num_input_tokens_seen": 84834355,
      "step": 3947,
      "time_per_iteration": 2.704730987548828
    },
    {
      "auxiliary_loss_clip": 0.01167781,
      "auxiliary_loss_mlp": 0.01036626,
      "balance_loss_clip": 1.0579977,
      "balance_loss_mlp": 1.02705932,
      "epoch": 0.4747189322431311,
      "flos": 20558500295040.0,
      "grad_norm": 1.943887697011717,
      "language_loss": 0.86416948,
      "learning_rate": 2.260215694836633e-06,
      "loss": 0.8862136,
      "num_input_tokens_seen": 84853530,
      "step": 3948,
      "time_per_iteration": 2.8582346439361572
    },
    {
      "auxiliary_loss_clip": 0.01196044,
      "auxiliary_loss_mlp": 0.01034082,
      "balance_loss_clip": 1.06176198,
      "balance_loss_mlp": 1.02506375,
      "epoch": 0.4748391751337702,
      "flos": 25995231970560.0,
      "grad_norm": 2.2889926492706727,
      "language_loss": 0.64922333,
      "learning_rate": 2.2594433248492157e-06,
      "loss": 0.67152458,
      "num_input_tokens_seen": 84872505,
      "step": 3949,
      "time_per_iteration": 2.7553939819335938
    },
    {
      "auxiliary_loss_clip": 0.01161624,
      "auxiliary_loss_mlp": 0.0103254,
      "balance_loss_clip": 1.05095637,
      "balance_loss_mlp": 1.02327156,
      "epoch": 0.47495941802440933,
      "flos": 22821052032000.0,
      "grad_norm": 2.604894113662823,
      "language_loss": 0.80211347,
      "learning_rate": 2.2586709155044527e-06,
      "loss": 0.82405508,
      "num_input_tokens_seen": 84893105,
      "step": 3950,
      "time_per_iteration": 2.681229591369629
    },
    {
      "auxiliary_loss_clip": 0.01194023,
      "auxiliary_loss_mlp": 0.01031977,
      "balance_loss_clip": 1.05841255,
      "balance_loss_mlp": 1.02286303,
      "epoch": 0.4750796609150484,
      "flos": 27891782075520.0,
      "grad_norm": 1.7202456983085122,
      "language_loss": 0.76327366,
      "learning_rate": 2.2578984669195167e-06,
      "loss": 0.78553367,
      "num_input_tokens_seen": 84914070,
      "step": 3951,
      "time_per_iteration": 2.713944673538208
    },
    {
      "auxiliary_loss_clip": 0.01171804,
      "auxiliary_loss_mlp": 0.01031023,
      "balance_loss_clip": 1.05545712,
      "balance_loss_mlp": 1.02211237,
      "epoch": 0.4751999038056875,
      "flos": 35660085471360.0,
      "grad_norm": 1.9785489479909166,
      "language_loss": 0.67888057,
      "learning_rate": 2.2571259792115887e-06,
      "loss": 0.70090878,
      "num_input_tokens_seen": 84935290,
      "step": 3952,
      "time_per_iteration": 2.8179566860198975
    },
    {
      "auxiliary_loss_clip": 0.01194744,
      "auxiliary_loss_mlp": 0.01030109,
      "balance_loss_clip": 1.06101227,
      "balance_loss_mlp": 1.02165687,
      "epoch": 0.4753201466963266,
      "flos": 22090880361600.0,
      "grad_norm": 1.9131153487663217,
      "language_loss": 0.79599625,
      "learning_rate": 2.2563534524978544e-06,
      "loss": 0.81824476,
      "num_input_tokens_seen": 84952760,
      "step": 3953,
      "time_per_iteration": 2.6447741985321045
    },
    {
      "auxiliary_loss_clip": 0.01178517,
      "auxiliary_loss_mlp": 0.01032178,
      "balance_loss_clip": 1.05947375,
      "balance_loss_mlp": 1.02320111,
      "epoch": 0.47544038958696566,
      "flos": 30190854965760.0,
      "grad_norm": 2.1122028297761193,
      "language_loss": 0.70834535,
      "learning_rate": 2.2555808868955052e-06,
      "loss": 0.7304523,
      "num_input_tokens_seen": 84974890,
      "step": 3954,
      "time_per_iteration": 2.6664116382598877
    },
    {
      "auxiliary_loss_clip": 0.01195418,
      "auxiliary_loss_mlp": 0.0103232,
      "balance_loss_clip": 1.05784106,
      "balance_loss_mlp": 1.02248478,
      "epoch": 0.47556063247760477,
      "flos": 23472219738240.0,
      "grad_norm": 2.529835307848566,
      "language_loss": 0.73940945,
      "learning_rate": 2.254808282521738e-06,
      "loss": 0.7616868,
      "num_input_tokens_seen": 84993640,
      "step": 3955,
      "time_per_iteration": 2.6751112937927246
    },
    {
      "auxiliary_loss_clip": 0.01176778,
      "auxiliary_loss_mlp": 0.01037524,
      "balance_loss_clip": 1.05638635,
      "balance_loss_mlp": 1.02897632,
      "epoch": 0.4756808753682438,
      "flos": 25155209531520.0,
      "grad_norm": 1.875111192339821,
      "language_loss": 0.81387323,
      "learning_rate": 2.2540356394937573e-06,
      "loss": 0.8360163,
      "num_input_tokens_seen": 85012340,
      "step": 3956,
      "time_per_iteration": 2.6477346420288086
    },
    {
      "auxiliary_loss_clip": 0.01172892,
      "auxiliary_loss_mlp": 0.01031077,
      "balance_loss_clip": 1.05476785,
      "balance_loss_mlp": 1.02146888,
      "epoch": 0.47580111825888294,
      "flos": 15669729573120.0,
      "grad_norm": 2.848705068025989,
      "language_loss": 0.84270203,
      "learning_rate": 2.253262957928772e-06,
      "loss": 0.86474168,
      "num_input_tokens_seen": 85029225,
      "step": 3957,
      "time_per_iteration": 2.743210792541504
    },
    {
      "auxiliary_loss_clip": 0.0117767,
      "auxiliary_loss_mlp": 0.01032122,
      "balance_loss_clip": 1.05703712,
      "balance_loss_mlp": 1.0231216,
      "epoch": 0.47592136114952205,
      "flos": 17636556637440.0,
      "grad_norm": 1.8540339560778192,
      "language_loss": 0.72133893,
      "learning_rate": 2.2524902379439976e-06,
      "loss": 0.74343681,
      "num_input_tokens_seen": 85047895,
      "step": 3958,
      "time_per_iteration": 2.7445905208587646
    },
    {
      "auxiliary_loss_clip": 0.01049316,
      "auxiliary_loss_mlp": 0.01007017,
      "balance_loss_clip": 1.01765585,
      "balance_loss_mlp": 1.00568795,
      "epoch": 0.4760416040401611,
      "flos": 61417159292160.0,
      "grad_norm": 0.7435748016330811,
      "language_loss": 0.63713503,
      "learning_rate": 2.251717479656655e-06,
      "loss": 0.65769833,
      "num_input_tokens_seen": 85112690,
      "step": 3959,
      "time_per_iteration": 3.3405911922454834
    },
    {
      "auxiliary_loss_clip": 0.01166702,
      "auxiliary_loss_mlp": 0.01034724,
      "balance_loss_clip": 1.05746877,
      "balance_loss_mlp": 1.02563453,
      "epoch": 0.4761618469308002,
      "flos": 18405871153920.0,
      "grad_norm": 2.2934091453121543,
      "language_loss": 0.76768672,
      "learning_rate": 2.2509446831839704e-06,
      "loss": 0.78970093,
      "num_input_tokens_seen": 85132130,
      "step": 3960,
      "time_per_iteration": 2.738924264907837
    },
    {
      "auxiliary_loss_clip": 0.01166077,
      "auxiliary_loss_mlp": 0.01037097,
      "balance_loss_clip": 1.05546975,
      "balance_loss_mlp": 1.02845418,
      "epoch": 0.4762820898214393,
      "flos": 18040911016320.0,
      "grad_norm": 3.133736573932086,
      "language_loss": 0.82168406,
      "learning_rate": 2.250171848643177e-06,
      "loss": 0.84371573,
      "num_input_tokens_seen": 85149420,
      "step": 3961,
      "time_per_iteration": 2.6874277591705322
    },
    {
      "auxiliary_loss_clip": 0.01181451,
      "auxiliary_loss_mlp": 0.0103582,
      "balance_loss_clip": 1.05801558,
      "balance_loss_mlp": 1.02695692,
      "epoch": 0.4764023327120784,
      "flos": 19318253541120.0,
      "grad_norm": 2.6241915077506235,
      "language_loss": 0.86470121,
      "learning_rate": 2.249398976151513e-06,
      "loss": 0.88687396,
      "num_input_tokens_seen": 85166970,
      "step": 3962,
      "time_per_iteration": 3.537783622741699
    },
    {
      "auxiliary_loss_clip": 0.01167884,
      "auxiliary_loss_mlp": 0.0103264,
      "balance_loss_clip": 1.05515063,
      "balance_loss_mlp": 1.02386618,
      "epoch": 0.4765225756027175,
      "flos": 22747255539840.0,
      "grad_norm": 2.4758534930277762,
      "language_loss": 0.78664505,
      "learning_rate": 2.248626065826223e-06,
      "loss": 0.80865026,
      "num_input_tokens_seen": 85185175,
      "step": 3963,
      "time_per_iteration": 3.667377233505249
    },
    {
      "auxiliary_loss_clip": 0.01084596,
      "auxiliary_loss_mlp": 0.01002403,
      "balance_loss_clip": 1.0179745,
      "balance_loss_mlp": 1.00124097,
      "epoch": 0.4766428184933566,
      "flos": 65933392106880.0,
      "grad_norm": 0.7690101453973526,
      "language_loss": 0.6253407,
      "learning_rate": 2.2478531177845564e-06,
      "loss": 0.64621067,
      "num_input_tokens_seen": 85246170,
      "step": 3964,
      "time_per_iteration": 4.184909820556641
    },
    {
      "auxiliary_loss_clip": 0.01169217,
      "auxiliary_loss_mlp": 0.01038457,
      "balance_loss_clip": 1.05713952,
      "balance_loss_mlp": 1.02943933,
      "epoch": 0.47676306138399566,
      "flos": 24136495908480.0,
      "grad_norm": 1.807458561353767,
      "language_loss": 0.85254836,
      "learning_rate": 2.247080132143769e-06,
      "loss": 0.87462515,
      "num_input_tokens_seen": 85268525,
      "step": 3965,
      "time_per_iteration": 2.7817628383636475
    },
    {
      "auxiliary_loss_clip": 0.01177274,
      "auxiliary_loss_mlp": 0.01033634,
      "balance_loss_clip": 1.05446231,
      "balance_loss_mlp": 1.0233283,
      "epoch": 0.47688330427463477,
      "flos": 12604322995200.0,
      "grad_norm": 2.6589736672738096,
      "language_loss": 0.69369566,
      "learning_rate": 2.246307109021121e-06,
      "loss": 0.7158047,
      "num_input_tokens_seen": 85285930,
      "step": 3966,
      "time_per_iteration": 2.7175986766815186
    },
    {
      "auxiliary_loss_clip": 0.01194745,
      "auxiliary_loss_mlp": 0.01034624,
      "balance_loss_clip": 1.06087601,
      "balance_loss_mlp": 1.02621329,
      "epoch": 0.4770035471652739,
      "flos": 21390585828480.0,
      "grad_norm": 2.1355314541772037,
      "language_loss": 0.82608807,
      "learning_rate": 2.2455340485338817e-06,
      "loss": 0.8483817,
      "num_input_tokens_seen": 85303565,
      "step": 3967,
      "time_per_iteration": 2.6735310554504395
    },
    {
      "auxiliary_loss_clip": 0.01178621,
      "auxiliary_loss_mlp": 0.01032372,
      "balance_loss_clip": 1.05533981,
      "balance_loss_mlp": 1.02347255,
      "epoch": 0.47712379005591293,
      "flos": 25156251025920.0,
      "grad_norm": 2.481149899949319,
      "language_loss": 0.67964041,
      "learning_rate": 2.244760950799322e-06,
      "loss": 0.70175028,
      "num_input_tokens_seen": 85321835,
      "step": 3968,
      "time_per_iteration": 2.703688859939575
    },
    {
      "auxiliary_loss_clip": 0.01177888,
      "auxiliary_loss_mlp": 0.01029588,
      "balance_loss_clip": 1.0553695,
      "balance_loss_mlp": 1.02159786,
      "epoch": 0.47724403294655204,
      "flos": 22054323294720.0,
      "grad_norm": 1.8488770500757097,
      "language_loss": 0.72493756,
      "learning_rate": 2.2439878159347203e-06,
      "loss": 0.74701238,
      "num_input_tokens_seen": 85341260,
      "step": 3969,
      "time_per_iteration": 2.628864288330078
    },
    {
      "auxiliary_loss_clip": 0.01083027,
      "auxiliary_loss_mlp": 0.01000648,
      "balance_loss_clip": 1.01648259,
      "balance_loss_mlp": 0.99940854,
      "epoch": 0.4773642758371911,
      "flos": 70229387658240.0,
      "grad_norm": 0.73111210616157,
      "language_loss": 0.55277634,
      "learning_rate": 2.2432146440573616e-06,
      "loss": 0.57361311,
      "num_input_tokens_seen": 85407220,
      "step": 3970,
      "time_per_iteration": 3.340772867202759
    },
    {
      "auxiliary_loss_clip": 0.01180955,
      "auxiliary_loss_mlp": 0.01030207,
      "balance_loss_clip": 1.05667806,
      "balance_loss_mlp": 1.02090847,
      "epoch": 0.4774845187278302,
      "flos": 23548602009600.0,
      "grad_norm": 2.0209505766220817,
      "language_loss": 0.66648364,
      "learning_rate": 2.242441435284534e-06,
      "loss": 0.68859529,
      "num_input_tokens_seen": 85426095,
      "step": 3971,
      "time_per_iteration": 3.567690372467041
    },
    {
      "auxiliary_loss_clip": 0.01183951,
      "auxiliary_loss_mlp": 0.01035506,
      "balance_loss_clip": 1.0577091,
      "balance_loss_mlp": 1.0260464,
      "epoch": 0.4776047616184693,
      "flos": 23075371301760.0,
      "grad_norm": 2.052563240258214,
      "language_loss": 0.85169888,
      "learning_rate": 2.2416681897335337e-06,
      "loss": 0.87389338,
      "num_input_tokens_seen": 85444245,
      "step": 3972,
      "time_per_iteration": 2.6731648445129395
    },
    {
      "auxiliary_loss_clip": 0.01180472,
      "auxiliary_loss_mlp": 0.0103229,
      "balance_loss_clip": 1.0575403,
      "balance_loss_mlp": 1.02349269,
      "epoch": 0.4777250045091084,
      "flos": 31898119374720.0,
      "grad_norm": 1.7944037764977867,
      "language_loss": 0.67354333,
      "learning_rate": 2.240894907521661e-06,
      "loss": 0.69567096,
      "num_input_tokens_seen": 85463325,
      "step": 3973,
      "time_per_iteration": 2.732475757598877
    },
    {
      "auxiliary_loss_clip": 0.01175877,
      "auxiliary_loss_mlp": 0.01035218,
      "balance_loss_clip": 1.0543313,
      "balance_loss_mlp": 1.02637911,
      "epoch": 0.4778452473997475,
      "flos": 24278163148800.0,
      "grad_norm": 2.0129448811713613,
      "language_loss": 0.64095223,
      "learning_rate": 2.240121588766223e-06,
      "loss": 0.66306317,
      "num_input_tokens_seen": 85483375,
      "step": 3974,
      "time_per_iteration": 2.672583818435669
    },
    {
      "auxiliary_loss_clip": 0.01171937,
      "auxiliary_loss_mlp": 0.01030515,
      "balance_loss_clip": 1.05615163,
      "balance_loss_mlp": 1.02219403,
      "epoch": 0.4779654902903866,
      "flos": 31575031516800.0,
      "grad_norm": 2.1858990065714368,
      "language_loss": 0.71845579,
      "learning_rate": 2.239348233584531e-06,
      "loss": 0.7404803,
      "num_input_tokens_seen": 85504230,
      "step": 3975,
      "time_per_iteration": 2.723923921585083
    },
    {
      "auxiliary_loss_clip": 0.01164653,
      "auxiliary_loss_mlp": 0.01027597,
      "balance_loss_clip": 1.05300307,
      "balance_loss_mlp": 1.01829815,
      "epoch": 0.47808573318102565,
      "flos": 19500428344320.0,
      "grad_norm": 2.265113351392501,
      "language_loss": 0.81500262,
      "learning_rate": 2.2385748420939013e-06,
      "loss": 0.83692509,
      "num_input_tokens_seen": 85523425,
      "step": 3976,
      "time_per_iteration": 2.7101213932037354
    },
    {
      "auxiliary_loss_clip": 0.01184283,
      "auxiliary_loss_mlp": 0.0103163,
      "balance_loss_clip": 1.05990791,
      "balance_loss_mlp": 1.0228796,
      "epoch": 0.47820597607166476,
      "flos": 22601135013120.0,
      "grad_norm": 4.102238016999529,
      "language_loss": 0.72655302,
      "learning_rate": 2.2378014144116583e-06,
      "loss": 0.74871212,
      "num_input_tokens_seen": 85542235,
      "step": 3977,
      "time_per_iteration": 2.719405174255371
    },
    {
      "auxiliary_loss_clip": 0.01194367,
      "auxiliary_loss_mlp": 0.01034119,
      "balance_loss_clip": 1.05903625,
      "balance_loss_mlp": 1.02547073,
      "epoch": 0.4783262189623039,
      "flos": 23003011353600.0,
      "grad_norm": 3.0424625144095945,
      "language_loss": 0.79999715,
      "learning_rate": 2.23702795065513e-06,
      "loss": 0.82228202,
      "num_input_tokens_seen": 85561815,
      "step": 3978,
      "time_per_iteration": 2.626932144165039
    },
    {
      "auxiliary_loss_clip": 0.01080896,
      "auxiliary_loss_mlp": 0.01001167,
      "balance_loss_clip": 1.01485538,
      "balance_loss_mlp": 0.9999572,
      "epoch": 0.47844646185294293,
      "flos": 49772801226240.0,
      "grad_norm": 0.98033782211854,
      "language_loss": 0.67408782,
      "learning_rate": 2.2362544509416493e-06,
      "loss": 0.69490844,
      "num_input_tokens_seen": 85613930,
      "step": 3979,
      "time_per_iteration": 3.0830817222595215
    },
    {
      "auxiliary_loss_clip": 0.011883,
      "auxiliary_loss_mlp": 0.01029856,
      "balance_loss_clip": 1.05644977,
      "balance_loss_mlp": 1.02126682,
      "epoch": 0.47856670474358204,
      "flos": 20229558520320.0,
      "grad_norm": 2.923925603912282,
      "language_loss": 0.82967925,
      "learning_rate": 2.2354809153885572e-06,
      "loss": 0.8518607,
      "num_input_tokens_seen": 85631000,
      "step": 3980,
      "time_per_iteration": 2.6217355728149414
    },
    {
      "auxiliary_loss_clip": 0.01162303,
      "auxiliary_loss_mlp": 0.01034817,
      "balance_loss_clip": 1.05359817,
      "balance_loss_mlp": 1.02542925,
      "epoch": 0.47868694763422115,
      "flos": 20990936131200.0,
      "grad_norm": 2.9252988968166633,
      "language_loss": 0.83035362,
      "learning_rate": 2.234707344113197e-06,
      "loss": 0.85232478,
      "num_input_tokens_seen": 85649095,
      "step": 3981,
      "time_per_iteration": 2.7099218368530273
    },
    {
      "auxiliary_loss_clip": 0.01172615,
      "auxiliary_loss_mlp": 0.01028498,
      "balance_loss_clip": 1.05622363,
      "balance_loss_mlp": 1.01986098,
      "epoch": 0.4788071905248602,
      "flos": 19026551191680.0,
      "grad_norm": 2.0063299129004397,
      "language_loss": 0.7786209,
      "learning_rate": 2.233933737232919e-06,
      "loss": 0.80063206,
      "num_input_tokens_seen": 85666875,
      "step": 3982,
      "time_per_iteration": 2.6996402740478516
    },
    {
      "auxiliary_loss_clip": 0.0116853,
      "auxiliary_loss_mlp": 0.01033416,
      "balance_loss_clip": 1.05508804,
      "balance_loss_mlp": 1.02502406,
      "epoch": 0.4789274334154993,
      "flos": 23002221254400.0,
      "grad_norm": 3.7754689088789974,
      "language_loss": 0.78478926,
      "learning_rate": 2.2331600948650793e-06,
      "loss": 0.80680871,
      "num_input_tokens_seen": 85687020,
      "step": 3983,
      "time_per_iteration": 2.643388032913208
    },
    {
      "auxiliary_loss_clip": 0.01178239,
      "auxiliary_loss_mlp": 0.0103514,
      "balance_loss_clip": 1.05688095,
      "balance_loss_mlp": 1.02553177,
      "epoch": 0.4790476763061384,
      "flos": 23075586783360.0,
      "grad_norm": 1.7466090997310488,
      "language_loss": 0.80450189,
      "learning_rate": 2.2323864171270386e-06,
      "loss": 0.82663566,
      "num_input_tokens_seen": 85708290,
      "step": 3984,
      "time_per_iteration": 2.827057123184204
    },
    {
      "auxiliary_loss_clip": 0.01191648,
      "auxiliary_loss_mlp": 0.01030446,
      "balance_loss_clip": 1.05610156,
      "balance_loss_mlp": 1.02164841,
      "epoch": 0.4791679191967775,
      "flos": 21179288073600.0,
      "grad_norm": 2.093696786351175,
      "language_loss": 0.73350251,
      "learning_rate": 2.231612704136164e-06,
      "loss": 0.75572342,
      "num_input_tokens_seen": 85728660,
      "step": 3985,
      "time_per_iteration": 2.6865291595458984
    },
    {
      "auxiliary_loss_clip": 0.01196159,
      "auxiliary_loss_mlp": 0.01034346,
      "balance_loss_clip": 1.05971575,
      "balance_loss_mlp": 1.02512538,
      "epoch": 0.4792881620874166,
      "flos": 22301495758080.0,
      "grad_norm": 4.90892348126209,
      "language_loss": 0.75069726,
      "learning_rate": 2.2308389560098253e-06,
      "loss": 0.77300233,
      "num_input_tokens_seen": 85745035,
      "step": 3986,
      "time_per_iteration": 2.625974655151367
    },
    {
      "auxiliary_loss_clip": 0.01162013,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.04956055,
      "balance_loss_mlp": 1.01899385,
      "epoch": 0.47940840497805565,
      "flos": 17420877423360.0,
      "grad_norm": 2.1919610384453345,
      "language_loss": 0.76982331,
      "learning_rate": 2.2300651728654008e-06,
      "loss": 0.79172564,
      "num_input_tokens_seen": 85760295,
      "step": 3987,
      "time_per_iteration": 2.676466703414917
    },
    {
      "auxiliary_loss_clip": 0.01080515,
      "auxiliary_loss_mlp": 0.01002339,
      "balance_loss_clip": 1.01454496,
      "balance_loss_mlp": 1.00117719,
      "epoch": 0.47952864786869476,
      "flos": 65358175708800.0,
      "grad_norm": 0.7256845966598332,
      "language_loss": 0.60142738,
      "learning_rate": 2.229291354820272e-06,
      "loss": 0.62225592,
      "num_input_tokens_seen": 85821305,
      "step": 3988,
      "time_per_iteration": 4.100970983505249
    },
    {
      "auxiliary_loss_clip": 0.01170695,
      "auxiliary_loss_mlp": 0.01028812,
      "balance_loss_clip": 1.0557735,
      "balance_loss_mlp": 1.01914406,
      "epoch": 0.47964889075933387,
      "flos": 16799802336000.0,
      "grad_norm": 1.9319384900545042,
      "language_loss": 0.76000726,
      "learning_rate": 2.228517501991828e-06,
      "loss": 0.78200233,
      "num_input_tokens_seen": 85840105,
      "step": 3989,
      "time_per_iteration": 2.6534793376922607
    },
    {
      "auxiliary_loss_clip": 0.01059233,
      "auxiliary_loss_mlp": 0.01002277,
      "balance_loss_clip": 1.01620579,
      "balance_loss_mlp": 1.00093031,
      "epoch": 0.4797691336499729,
      "flos": 70079244808320.0,
      "grad_norm": 0.815728201674796,
      "language_loss": 0.61039263,
      "learning_rate": 2.22774361449746e-06,
      "loss": 0.63100773,
      "num_input_tokens_seen": 85896585,
      "step": 3990,
      "time_per_iteration": 5.164991617202759
    },
    {
      "auxiliary_loss_clip": 0.01162603,
      "auxiliary_loss_mlp": 0.01035635,
      "balance_loss_clip": 1.05233598,
      "balance_loss_mlp": 1.02632451,
      "epoch": 0.47988937654061203,
      "flos": 18953329317120.0,
      "grad_norm": 7.151985316103263,
      "language_loss": 0.70530522,
      "learning_rate": 2.2269696924545668e-06,
      "loss": 0.72728753,
      "num_input_tokens_seen": 85914415,
      "step": 3991,
      "time_per_iteration": 2.6908254623413086
    },
    {
      "auxiliary_loss_clip": 0.01160358,
      "auxiliary_loss_mlp": 0.01040921,
      "balance_loss_clip": 1.05898094,
      "balance_loss_mlp": 1.03251696,
      "epoch": 0.48000961943125114,
      "flos": 14461981649280.0,
      "grad_norm": 3.048417184369726,
      "language_loss": 0.7879836,
      "learning_rate": 2.2261957359805523e-06,
      "loss": 0.80999637,
      "num_input_tokens_seen": 85931650,
      "step": 3992,
      "time_per_iteration": 2.8073384761810303
    },
    {
      "auxiliary_loss_clip": 0.01178282,
      "auxiliary_loss_mlp": 0.01031386,
      "balance_loss_clip": 1.05636013,
      "balance_loss_mlp": 1.02217758,
      "epoch": 0.4801298623218902,
      "flos": 27051149105280.0,
      "grad_norm": 2.212761765534598,
      "language_loss": 0.74337608,
      "learning_rate": 2.225421745192823e-06,
      "loss": 0.76547277,
      "num_input_tokens_seen": 85951805,
      "step": 3993,
      "time_per_iteration": 2.7566044330596924
    },
    {
      "auxiliary_loss_clip": 0.01178368,
      "auxiliary_loss_mlp": 0.01030187,
      "balance_loss_clip": 1.05693698,
      "balance_loss_mlp": 1.02076364,
      "epoch": 0.4802501052125293,
      "flos": 26355236031360.0,
      "grad_norm": 2.808021134608799,
      "language_loss": 0.7851758,
      "learning_rate": 2.2246477202087955e-06,
      "loss": 0.80726135,
      "num_input_tokens_seen": 85972485,
      "step": 3994,
      "time_per_iteration": 2.70806884765625
    },
    {
      "auxiliary_loss_clip": 0.01191419,
      "auxiliary_loss_mlp": 0.01036641,
      "balance_loss_clip": 1.05788291,
      "balance_loss_mlp": 1.02793288,
      "epoch": 0.4803703481031684,
      "flos": 20993916960000.0,
      "grad_norm": 2.5362244667485685,
      "language_loss": 0.83196568,
      "learning_rate": 2.223873661145887e-06,
      "loss": 0.85424626,
      "num_input_tokens_seen": 85992540,
      "step": 3995,
      "time_per_iteration": 2.602586507797241
    },
    {
      "auxiliary_loss_clip": 0.01169284,
      "auxiliary_loss_mlp": 0.01034698,
      "balance_loss_clip": 1.05411148,
      "balance_loss_mlp": 1.02609062,
      "epoch": 0.4804905909938075,
      "flos": 20703722981760.0,
      "grad_norm": 1.620417459670771,
      "language_loss": 0.71552205,
      "learning_rate": 2.2230995681215226e-06,
      "loss": 0.73756188,
      "num_input_tokens_seen": 86012065,
      "step": 3996,
      "time_per_iteration": 2.6613903045654297
    },
    {
      "auxiliary_loss_clip": 0.01181474,
      "auxiliary_loss_mlp": 0.01029362,
      "balance_loss_clip": 1.06042087,
      "balance_loss_mlp": 1.02017689,
      "epoch": 0.4806108338844466,
      "flos": 16654831044480.0,
      "grad_norm": 1.8477558238057339,
      "language_loss": 0.7826103,
      "learning_rate": 2.2223254412531305e-06,
      "loss": 0.80471867,
      "num_input_tokens_seen": 86029435,
      "step": 3997,
      "time_per_iteration": 3.5807299613952637
    },
    {
      "auxiliary_loss_clip": 0.01172244,
      "auxiliary_loss_mlp": 0.01028281,
      "balance_loss_clip": 1.05122876,
      "balance_loss_mlp": 1.02018654,
      "epoch": 0.4807310767750857,
      "flos": 20011329440640.0,
      "grad_norm": 2.4341187447680466,
      "language_loss": 0.82391191,
      "learning_rate": 2.221551280658146e-06,
      "loss": 0.84591717,
      "num_input_tokens_seen": 86048495,
      "step": 3998,
      "time_per_iteration": 2.662184238433838
    },
    {
      "auxiliary_loss_clip": 0.01176182,
      "auxiliary_loss_mlp": 0.01025843,
      "balance_loss_clip": 1.05908537,
      "balance_loss_mlp": 1.0171411,
      "epoch": 0.48085131966572475,
      "flos": 23185257984000.0,
      "grad_norm": 2.1367221467041007,
      "language_loss": 0.74188566,
      "learning_rate": 2.2207770864540085e-06,
      "loss": 0.76390588,
      "num_input_tokens_seen": 86067470,
      "step": 3999,
      "time_per_iteration": 2.683330774307251
    },
    {
      "auxiliary_loss_clip": 0.01180431,
      "auxiliary_loss_mlp": 0.0102869,
      "balance_loss_clip": 1.05730593,
      "balance_loss_mlp": 1.02041078,
      "epoch": 0.48097156255636386,
      "flos": 20558643949440.0,
      "grad_norm": 2.1283486648684375,
      "language_loss": 0.73086041,
      "learning_rate": 2.220002858758162e-06,
      "loss": 0.75295162,
      "num_input_tokens_seen": 86085460,
      "step": 4000,
      "time_per_iteration": 2.6034247875213623
    },
    {
      "auxiliary_loss_clip": 0.01066902,
      "auxiliary_loss_mlp": 0.01001096,
      "balance_loss_clip": 1.01582038,
      "balance_loss_mlp": 0.99979109,
      "epoch": 0.481091805447003,
      "flos": 70511608817280.0,
      "grad_norm": 0.8918196299549029,
      "language_loss": 0.60815084,
      "learning_rate": 2.2192285976880573e-06,
      "loss": 0.62883079,
      "num_input_tokens_seen": 86149715,
      "step": 4001,
      "time_per_iteration": 3.2375435829162598
    },
    {
      "auxiliary_loss_clip": 0.01169435,
      "auxiliary_loss_mlp": 0.01030977,
      "balance_loss_clip": 1.05417943,
      "balance_loss_mlp": 1.0229125,
      "epoch": 0.48121204833764203,
      "flos": 36428214839040.0,
      "grad_norm": 1.9917937415633589,
      "language_loss": 0.80943376,
      "learning_rate": 2.2184543033611485e-06,
      "loss": 0.83143789,
      "num_input_tokens_seen": 86170795,
      "step": 4002,
      "time_per_iteration": 2.733590602874756
    },
    {
      "auxiliary_loss_clip": 0.01190163,
      "auxiliary_loss_mlp": 0.01028085,
      "balance_loss_clip": 1.05647898,
      "balance_loss_mlp": 1.01897717,
      "epoch": 0.48133229122828114,
      "flos": 27490264871040.0,
      "grad_norm": 2.8821329711452988,
      "language_loss": 0.82419348,
      "learning_rate": 2.2176799758948957e-06,
      "loss": 0.84637594,
      "num_input_tokens_seen": 86190955,
      "step": 4003,
      "time_per_iteration": 2.720069408416748
    },
    {
      "auxiliary_loss_clip": 0.0115349,
      "auxiliary_loss_mlp": 0.01036431,
      "balance_loss_clip": 1.05548036,
      "balance_loss_mlp": 1.02721584,
      "epoch": 0.4814525341189202,
      "flos": 43072802179200.0,
      "grad_norm": 2.451102139197596,
      "language_loss": 0.73656857,
      "learning_rate": 2.2169056154067635e-06,
      "loss": 0.75846779,
      "num_input_tokens_seen": 86214875,
      "step": 4004,
      "time_per_iteration": 2.939652681350708
    },
    {
      "auxiliary_loss_clip": 0.01155593,
      "auxiliary_loss_mlp": 0.01034033,
      "balance_loss_clip": 1.04998899,
      "balance_loss_mlp": 1.02503312,
      "epoch": 0.4815727770095593,
      "flos": 24236901400320.0,
      "grad_norm": 5.642972563938197,
      "language_loss": 0.82494026,
      "learning_rate": 2.216131222014222e-06,
      "loss": 0.84683651,
      "num_input_tokens_seen": 86232950,
      "step": 4005,
      "time_per_iteration": 2.7104766368865967
    },
    {
      "auxiliary_loss_clip": 0.01195699,
      "auxiliary_loss_mlp": 0.01035836,
      "balance_loss_clip": 1.06149268,
      "balance_loss_mlp": 1.02657914,
      "epoch": 0.4816930199001984,
      "flos": 18113630100480.0,
      "grad_norm": 2.9054857075182,
      "language_loss": 0.80894017,
      "learning_rate": 2.2153567958347455e-06,
      "loss": 0.83125556,
      "num_input_tokens_seen": 86249160,
      "step": 4006,
      "time_per_iteration": 2.6123952865600586
    },
    {
      "auxiliary_loss_clip": 0.01167628,
      "auxiliary_loss_mlp": 0.01035926,
      "balance_loss_clip": 1.05784512,
      "balance_loss_mlp": 1.02739072,
      "epoch": 0.48181326279083747,
      "flos": 17274720983040.0,
      "grad_norm": 2.394779764630656,
      "language_loss": 0.79980654,
      "learning_rate": 2.214582336985815e-06,
      "loss": 0.82184207,
      "num_input_tokens_seen": 86267060,
      "step": 4007,
      "time_per_iteration": 2.6846673488616943
    },
    {
      "auxiliary_loss_clip": 0.01168909,
      "auxiliary_loss_mlp": 0.01039717,
      "balance_loss_clip": 1.05338705,
      "balance_loss_mlp": 1.02981663,
      "epoch": 0.4819335056814766,
      "flos": 14903252231040.0,
      "grad_norm": 2.3518286972597964,
      "language_loss": 0.66422951,
      "learning_rate": 2.2138078455849142e-06,
      "loss": 0.68631577,
      "num_input_tokens_seen": 86285055,
      "step": 4008,
      "time_per_iteration": 2.695948600769043
    },
    {
      "auxiliary_loss_clip": 0.01172072,
      "auxiliary_loss_mlp": 0.01036687,
      "balance_loss_clip": 1.05321169,
      "balance_loss_mlp": 1.02767444,
      "epoch": 0.4820537485721157,
      "flos": 19244888012160.0,
      "grad_norm": 7.259090656419329,
      "language_loss": 0.79217088,
      "learning_rate": 2.2130333217495334e-06,
      "loss": 0.81425846,
      "num_input_tokens_seen": 86304225,
      "step": 4009,
      "time_per_iteration": 2.7164366245269775
    },
    {
      "auxiliary_loss_clip": 0.01174199,
      "auxiliary_loss_mlp": 0.01028989,
      "balance_loss_clip": 1.0547421,
      "balance_loss_mlp": 1.01957178,
      "epoch": 0.48217399146275475,
      "flos": 16033791870720.0,
      "grad_norm": 2.5712567259335217,
      "language_loss": 0.68380618,
      "learning_rate": 2.2122587655971665e-06,
      "loss": 0.70583802,
      "num_input_tokens_seen": 86319170,
      "step": 4010,
      "time_per_iteration": 2.676635980606079
    },
    {
      "auxiliary_loss_clip": 0.01191284,
      "auxiliary_loss_mlp": 0.01028391,
      "balance_loss_clip": 1.05833972,
      "balance_loss_mlp": 1.01942647,
      "epoch": 0.48229423435339386,
      "flos": 24134197438080.0,
      "grad_norm": 1.6571251838008878,
      "language_loss": 0.64075673,
      "learning_rate": 2.211484177245314e-06,
      "loss": 0.6629535,
      "num_input_tokens_seen": 86338760,
      "step": 4011,
      "time_per_iteration": 2.609992742538452
    },
    {
      "auxiliary_loss_clip": 0.01126143,
      "auxiliary_loss_mlp": 0.01032517,
      "balance_loss_clip": 1.05057704,
      "balance_loss_mlp": 1.02371979,
      "epoch": 0.48241447724403297,
      "flos": 23805435231360.0,
      "grad_norm": 2.319901886944808,
      "language_loss": 0.72630632,
      "learning_rate": 2.21070955681148e-06,
      "loss": 0.74789292,
      "num_input_tokens_seen": 86357865,
      "step": 4012,
      "time_per_iteration": 2.90079402923584
    },
    {
      "auxiliary_loss_clip": 0.01150232,
      "auxiliary_loss_mlp": 0.01032256,
      "balance_loss_clip": 1.05232906,
      "balance_loss_mlp": 1.02333331,
      "epoch": 0.482534720134672,
      "flos": 23110312256640.0,
      "grad_norm": 2.0560232463067707,
      "language_loss": 0.78534389,
      "learning_rate": 2.209934904413174e-06,
      "loss": 0.80716872,
      "num_input_tokens_seen": 86379470,
      "step": 4013,
      "time_per_iteration": 2.756913423538208
    },
    {
      "auxiliary_loss_clip": 0.0118182,
      "auxiliary_loss_mlp": 0.0102925,
      "balance_loss_clip": 1.05570626,
      "balance_loss_mlp": 1.02001691,
      "epoch": 0.48265496302531113,
      "flos": 20923819568640.0,
      "grad_norm": 3.7492992608278293,
      "language_loss": 0.71186125,
      "learning_rate": 2.2091602201679095e-06,
      "loss": 0.73397195,
      "num_input_tokens_seen": 86399080,
      "step": 4014,
      "time_per_iteration": 3.5853965282440186
    },
    {
      "auxiliary_loss_clip": 0.01192623,
      "auxiliary_loss_mlp": 0.01030925,
      "balance_loss_clip": 1.05917335,
      "balance_loss_mlp": 1.02292562,
      "epoch": 0.48277520591595025,
      "flos": 15231152511360.0,
      "grad_norm": 2.2664167667197628,
      "language_loss": 0.83124232,
      "learning_rate": 2.208385504193206e-06,
      "loss": 0.85347784,
      "num_input_tokens_seen": 86416580,
      "step": 4015,
      "time_per_iteration": 2.6414577960968018
    },
    {
      "auxiliary_loss_clip": 0.01162462,
      "auxiliary_loss_mlp": 0.01030358,
      "balance_loss_clip": 1.05178249,
      "balance_loss_mlp": 1.02199507,
      "epoch": 0.4828954488065893,
      "flos": 17858664385920.0,
      "grad_norm": 2.2239341654220097,
      "language_loss": 0.81799734,
      "learning_rate": 2.2076107566065873e-06,
      "loss": 0.83992553,
      "num_input_tokens_seen": 86434365,
      "step": 4016,
      "time_per_iteration": 4.590765714645386
    },
    {
      "auxiliary_loss_clip": 0.0119195,
      "auxiliary_loss_mlp": 0.01029685,
      "balance_loss_clip": 1.06000209,
      "balance_loss_mlp": 1.02114332,
      "epoch": 0.4830156916972284,
      "flos": 32087405070720.0,
      "grad_norm": 2.3224300633919266,
      "language_loss": 0.76017666,
      "learning_rate": 2.2068359775255816e-06,
      "loss": 0.78239298,
      "num_input_tokens_seen": 86452675,
      "step": 4017,
      "time_per_iteration": 2.722554922103882
    },
    {
      "auxiliary_loss_clip": 0.01178845,
      "auxiliary_loss_mlp": 0.01029876,
      "balance_loss_clip": 1.05638838,
      "balance_loss_mlp": 1.02160335,
      "epoch": 0.48313593458786747,
      "flos": 21871717528320.0,
      "grad_norm": 3.318850883762124,
      "language_loss": 0.78893411,
      "learning_rate": 2.206061167067723e-06,
      "loss": 0.81102133,
      "num_input_tokens_seen": 86470785,
      "step": 4018,
      "time_per_iteration": 2.6793484687805176
    },
    {
      "auxiliary_loss_clip": 0.01175405,
      "auxiliary_loss_mlp": 0.01036831,
      "balance_loss_clip": 1.05672705,
      "balance_loss_mlp": 1.02688313,
      "epoch": 0.4832561774785066,
      "flos": 22601206840320.0,
      "grad_norm": 2.7306723546191125,
      "language_loss": 0.79923391,
      "learning_rate": 2.205286325350549e-06,
      "loss": 0.8213563,
      "num_input_tokens_seen": 86489850,
      "step": 4019,
      "time_per_iteration": 2.7332749366760254
    },
    {
      "auxiliary_loss_clip": 0.01190746,
      "auxiliary_loss_mlp": 0.01035184,
      "balance_loss_clip": 1.05714834,
      "balance_loss_mlp": 1.02618408,
      "epoch": 0.4833764203691457,
      "flos": 13437342282240.0,
      "grad_norm": 2.2276157573242297,
      "language_loss": 0.72536868,
      "learning_rate": 2.204511452491603e-06,
      "loss": 0.74762797,
      "num_input_tokens_seen": 86506475,
      "step": 4020,
      "time_per_iteration": 2.6228368282318115
    },
    {
      "auxiliary_loss_clip": 0.0115735,
      "auxiliary_loss_mlp": 0.01029362,
      "balance_loss_clip": 1.05347013,
      "balance_loss_mlp": 1.02094018,
      "epoch": 0.48349666325978474,
      "flos": 44128036955520.0,
      "grad_norm": 4.0353805556493665,
      "language_loss": 0.74480873,
      "learning_rate": 2.2037365486084316e-06,
      "loss": 0.76667583,
      "num_input_tokens_seen": 86529715,
      "step": 4021,
      "time_per_iteration": 2.8505706787109375
    },
    {
      "auxiliary_loss_clip": 0.01191856,
      "auxiliary_loss_mlp": 0.01033494,
      "balance_loss_clip": 1.05672991,
      "balance_loss_mlp": 1.02482712,
      "epoch": 0.48361690615042385,
      "flos": 26028377245440.0,
      "grad_norm": 1.9211805154812829,
      "language_loss": 0.77955586,
      "learning_rate": 2.2029616138185886e-06,
      "loss": 0.80180931,
      "num_input_tokens_seen": 86548715,
      "step": 4022,
      "time_per_iteration": 2.741821050643921
    },
    {
      "auxiliary_loss_clip": 0.01191773,
      "auxiliary_loss_mlp": 0.01040836,
      "balance_loss_clip": 1.05879629,
      "balance_loss_mlp": 1.03277111,
      "epoch": 0.48373714904106296,
      "flos": 22273306560000.0,
      "grad_norm": 2.191752435574354,
      "language_loss": 0.83065253,
      "learning_rate": 2.202186648239629e-06,
      "loss": 0.85297865,
      "num_input_tokens_seen": 86568650,
      "step": 4023,
      "time_per_iteration": 3.545156955718994
    },
    {
      "auxiliary_loss_clip": 0.01139934,
      "auxiliary_loss_mlp": 0.01027393,
      "balance_loss_clip": 1.04682326,
      "balance_loss_mlp": 1.01928687,
      "epoch": 0.483857391931702,
      "flos": 28292293699200.0,
      "grad_norm": 1.9373833320300642,
      "language_loss": 0.71818793,
      "learning_rate": 2.201411651989117e-06,
      "loss": 0.73986125,
      "num_input_tokens_seen": 86590630,
      "step": 4024,
      "time_per_iteration": 2.774613618850708
    },
    {
      "auxiliary_loss_clip": 0.01192318,
      "auxiliary_loss_mlp": 0.01028617,
      "balance_loss_clip": 1.05815411,
      "balance_loss_mlp": 1.02024889,
      "epoch": 0.48397763482234113,
      "flos": 27418048577280.0,
      "grad_norm": 2.1197349183534726,
      "language_loss": 0.77963412,
      "learning_rate": 2.2006366251846167e-06,
      "loss": 0.8018434,
      "num_input_tokens_seen": 86611270,
      "step": 4025,
      "time_per_iteration": 2.7034919261932373
    },
    {
      "auxiliary_loss_clip": 0.01189215,
      "auxiliary_loss_mlp": 0.0103054,
      "balance_loss_clip": 1.05809879,
      "balance_loss_mlp": 1.02254725,
      "epoch": 0.48409787771298024,
      "flos": 16797252470400.0,
      "grad_norm": 2.5229431149895394,
      "language_loss": 0.75315166,
      "learning_rate": 2.1998615679436997e-06,
      "loss": 0.7753492,
      "num_input_tokens_seen": 86628810,
      "step": 4026,
      "time_per_iteration": 2.6037039756774902
    },
    {
      "auxiliary_loss_clip": 0.01139608,
      "auxiliary_loss_mlp": 0.01030393,
      "balance_loss_clip": 1.05014133,
      "balance_loss_mlp": 1.02141666,
      "epoch": 0.4842181206036193,
      "flos": 25083496028160.0,
      "grad_norm": 5.457806823742608,
      "language_loss": 0.77429003,
      "learning_rate": 2.199086480383942e-06,
      "loss": 0.79598999,
      "num_input_tokens_seen": 86648185,
      "step": 4027,
      "time_per_iteration": 2.768707513809204
    },
    {
      "auxiliary_loss_clip": 0.01156524,
      "auxiliary_loss_mlp": 0.01038151,
      "balance_loss_clip": 1.05425119,
      "balance_loss_mlp": 1.02789867,
      "epoch": 0.4843383634942584,
      "flos": 30372311496960.0,
      "grad_norm": 2.9328991101526714,
      "language_loss": 0.67827445,
      "learning_rate": 2.1983113626229234e-06,
      "loss": 0.70022118,
      "num_input_tokens_seen": 86667435,
      "step": 4028,
      "time_per_iteration": 2.7375705242156982
    },
    {
      "auxiliary_loss_clip": 0.01178961,
      "auxiliary_loss_mlp": 0.01036811,
      "balance_loss_clip": 1.05513859,
      "balance_loss_mlp": 1.02819872,
      "epoch": 0.4844586063848975,
      "flos": 20413564917120.0,
      "grad_norm": 19.416338579075045,
      "language_loss": 0.78611231,
      "learning_rate": 2.1975362147782293e-06,
      "loss": 0.80827004,
      "num_input_tokens_seen": 86686630,
      "step": 4029,
      "time_per_iteration": 2.675021171569824
    },
    {
      "auxiliary_loss_clip": 0.01071605,
      "auxiliary_loss_mlp": 0.01001429,
      "balance_loss_clip": 1.01480913,
      "balance_loss_mlp": 1.00029016,
      "epoch": 0.48457884927553657,
      "flos": 70303722854400.0,
      "grad_norm": 0.6920904846394319,
      "language_loss": 0.54093367,
      "learning_rate": 2.196761036967448e-06,
      "loss": 0.56166404,
      "num_input_tokens_seen": 86754595,
      "step": 4030,
      "time_per_iteration": 3.3537445068359375
    },
    {
      "auxiliary_loss_clip": 0.01186195,
      "auxiliary_loss_mlp": 0.01030445,
      "balance_loss_clip": 1.05577946,
      "balance_loss_mlp": 1.02294707,
      "epoch": 0.4846990921661757,
      "flos": 19934516206080.0,
      "grad_norm": 1.9580395673669366,
      "language_loss": 0.77717519,
      "learning_rate": 2.1959858293081743e-06,
      "loss": 0.79934156,
      "num_input_tokens_seen": 86773730,
      "step": 4031,
      "time_per_iteration": 2.595853328704834
    },
    {
      "auxiliary_loss_clip": 0.0118052,
      "auxiliary_loss_mlp": 0.01031667,
      "balance_loss_clip": 1.05894709,
      "balance_loss_mlp": 1.02315581,
      "epoch": 0.4848193350568148,
      "flos": 23075945919360.0,
      "grad_norm": 1.6600813142074382,
      "language_loss": 0.76204801,
      "learning_rate": 2.1952105919180056e-06,
      "loss": 0.78416991,
      "num_input_tokens_seen": 86792985,
      "step": 4032,
      "time_per_iteration": 2.6460769176483154
    },
    {
      "auxiliary_loss_clip": 0.0118296,
      "auxiliary_loss_mlp": 0.01028874,
      "balance_loss_clip": 1.05653512,
      "balance_loss_mlp": 1.0199753,
      "epoch": 0.48493957794745385,
      "flos": 22455481363200.0,
      "grad_norm": 2.2822359078099264,
      "language_loss": 0.68190914,
      "learning_rate": 2.1944353249145456e-06,
      "loss": 0.70402747,
      "num_input_tokens_seen": 86812095,
      "step": 4033,
      "time_per_iteration": 2.6872141361236572
    },
    {
      "auxiliary_loss_clip": 0.01176638,
      "auxiliary_loss_mlp": 0.01033335,
      "balance_loss_clip": 1.05673182,
      "balance_loss_mlp": 1.02539539,
      "epoch": 0.48505982083809296,
      "flos": 25046112948480.0,
      "grad_norm": 1.9452276902199037,
      "language_loss": 0.75198013,
      "learning_rate": 2.193660028415401e-06,
      "loss": 0.77407986,
      "num_input_tokens_seen": 86832875,
      "step": 4034,
      "time_per_iteration": 2.776327610015869
    },
    {
      "auxiliary_loss_clip": 0.01176221,
      "auxiliary_loss_mlp": 0.01032367,
      "balance_loss_clip": 1.05514908,
      "balance_loss_mlp": 1.02391481,
      "epoch": 0.485180063728732,
      "flos": 26761386090240.0,
      "grad_norm": 2.1577367504949376,
      "language_loss": 0.82141995,
      "learning_rate": 2.1928847025381852e-06,
      "loss": 0.8435058,
      "num_input_tokens_seen": 86853480,
      "step": 4035,
      "time_per_iteration": 2.719308614730835
    },
    {
      "auxiliary_loss_clip": 0.01190265,
      "auxiliary_loss_mlp": 0.01029638,
      "balance_loss_clip": 1.05448437,
      "balance_loss_mlp": 1.02054262,
      "epoch": 0.4853003066193711,
      "flos": 24059143969920.0,
      "grad_norm": 1.8061236637681324,
      "language_loss": 0.84225559,
      "learning_rate": 2.192109347400512e-06,
      "loss": 0.86445463,
      "num_input_tokens_seen": 86873695,
      "step": 4036,
      "time_per_iteration": 2.812817096710205
    },
    {
      "auxiliary_loss_clip": 0.01183713,
      "auxiliary_loss_mlp": 0.01028867,
      "balance_loss_clip": 1.05547369,
      "balance_loss_mlp": 1.01953244,
      "epoch": 0.48542054951001024,
      "flos": 23076376882560.0,
      "grad_norm": 2.4029288574135124,
      "language_loss": 0.78981948,
      "learning_rate": 2.191333963120004e-06,
      "loss": 0.8119452,
      "num_input_tokens_seen": 86892675,
      "step": 4037,
      "time_per_iteration": 2.673276424407959
    },
    {
      "auxiliary_loss_clip": 0.01191177,
      "auxiliary_loss_mlp": 0.01026603,
      "balance_loss_clip": 1.05622005,
      "balance_loss_mlp": 1.01739991,
      "epoch": 0.4855407924006493,
      "flos": 25664889565440.0,
      "grad_norm": 3.208345295254102,
      "language_loss": 0.70469654,
      "learning_rate": 2.190558549814286e-06,
      "loss": 0.72687435,
      "num_input_tokens_seen": 86912835,
      "step": 4038,
      "time_per_iteration": 2.634516716003418
    },
    {
      "auxiliary_loss_clip": 0.01192713,
      "auxiliary_loss_mlp": 0.01034531,
      "balance_loss_clip": 1.0588479,
      "balance_loss_mlp": 1.02623987,
      "epoch": 0.4856610352912884,
      "flos": 23987933256960.0,
      "grad_norm": 3.66244871572608,
      "language_loss": 0.79521859,
      "learning_rate": 2.1897831076009872e-06,
      "loss": 0.81749105,
      "num_input_tokens_seen": 86932475,
      "step": 4039,
      "time_per_iteration": 2.593172311782837
    },
    {
      "auxiliary_loss_clip": 0.01144672,
      "auxiliary_loss_mlp": 0.01032041,
      "balance_loss_clip": 1.05298352,
      "balance_loss_mlp": 1.02320766,
      "epoch": 0.4857812781819275,
      "flos": 24096814358400.0,
      "grad_norm": 1.712930735201016,
      "language_loss": 0.79975855,
      "learning_rate": 2.1890076365977426e-06,
      "loss": 0.82152563,
      "num_input_tokens_seen": 86952300,
      "step": 4040,
      "time_per_iteration": 3.7179207801818848
    },
    {
      "auxiliary_loss_clip": 0.01070132,
      "auxiliary_loss_mlp": 0.01000582,
      "balance_loss_clip": 1.01582909,
      "balance_loss_mlp": 0.99940789,
      "epoch": 0.48590152107256657,
      "flos": 56266635185280.0,
      "grad_norm": 0.8528373086694535,
      "language_loss": 0.52744132,
      "learning_rate": 2.188232136922189e-06,
      "loss": 0.54814845,
      "num_input_tokens_seen": 87010420,
      "step": 4041,
      "time_per_iteration": 3.164889335632324
    },
    {
      "auxiliary_loss_clip": 0.01169866,
      "auxiliary_loss_mlp": 0.01034166,
      "balance_loss_clip": 1.05295396,
      "balance_loss_mlp": 1.02480793,
      "epoch": 0.4860217639632057,
      "flos": 20046988667520.0,
      "grad_norm": 2.032047299254846,
      "language_loss": 0.76257133,
      "learning_rate": 2.187456608691971e-06,
      "loss": 0.7846117,
      "num_input_tokens_seen": 87029295,
      "step": 4042,
      "time_per_iteration": 3.6691622734069824
    },
    {
      "auxiliary_loss_clip": 0.01192643,
      "auxiliary_loss_mlp": 0.01031508,
      "balance_loss_clip": 1.05852926,
      "balance_loss_mlp": 1.02256727,
      "epoch": 0.4861420068538448,
      "flos": 17822143232640.0,
      "grad_norm": 2.233389774747314,
      "language_loss": 0.87514007,
      "learning_rate": 2.1866810520247334e-06,
      "loss": 0.89738154,
      "num_input_tokens_seen": 87048165,
      "step": 4043,
      "time_per_iteration": 2.6702327728271484
    },
    {
      "auxiliary_loss_clip": 0.01162388,
      "auxiliary_loss_mlp": 0.01026707,
      "balance_loss_clip": 1.05572581,
      "balance_loss_mlp": 1.0174619,
      "epoch": 0.48626224974448384,
      "flos": 26250125857920.0,
      "grad_norm": 2.5572587108498013,
      "language_loss": 0.65217662,
      "learning_rate": 2.185905467038129e-06,
      "loss": 0.67406756,
      "num_input_tokens_seen": 87067070,
      "step": 4044,
      "time_per_iteration": 2.733731508255005
    },
    {
      "auxiliary_loss_clip": 0.01178702,
      "auxiliary_loss_mlp": 0.00763196,
      "balance_loss_clip": 1.05811608,
      "balance_loss_mlp": 1.00123608,
      "epoch": 0.48638249263512295,
      "flos": 22054502862720.0,
      "grad_norm": 1.6894632645973895,
      "language_loss": 0.7781837,
      "learning_rate": 2.1851298538498127e-06,
      "loss": 0.79760271,
      "num_input_tokens_seen": 87086785,
      "step": 4045,
      "time_per_iteration": 2.782163143157959
    },
    {
      "auxiliary_loss_clip": 0.01198944,
      "auxiliary_loss_mlp": 0.01029832,
      "balance_loss_clip": 1.06149995,
      "balance_loss_mlp": 1.01978898,
      "epoch": 0.48650273552576206,
      "flos": 25119945354240.0,
      "grad_norm": 2.2419770150612965,
      "language_loss": 0.80053967,
      "learning_rate": 2.184354212577446e-06,
      "loss": 0.82282746,
      "num_input_tokens_seen": 87107090,
      "step": 4046,
      "time_per_iteration": 2.6951780319213867
    },
    {
      "auxiliary_loss_clip": 0.01160433,
      "auxiliary_loss_mlp": 0.01031012,
      "balance_loss_clip": 1.05350542,
      "balance_loss_mlp": 1.02186298,
      "epoch": 0.4866229784164011,
      "flos": 17456931699840.0,
      "grad_norm": 4.389890103727963,
      "language_loss": 0.62947989,
      "learning_rate": 2.1835785433386907e-06,
      "loss": 0.65139437,
      "num_input_tokens_seen": 87125905,
      "step": 4047,
      "time_per_iteration": 2.659602165222168
    },
    {
      "auxiliary_loss_clip": 0.0119421,
      "auxiliary_loss_mlp": 0.01033109,
      "balance_loss_clip": 1.05899549,
      "balance_loss_mlp": 1.02383471,
      "epoch": 0.48674322130704023,
      "flos": 23331127115520.0,
      "grad_norm": 2.0880113857557565,
      "language_loss": 0.66159916,
      "learning_rate": 2.182802846251216e-06,
      "loss": 0.68387234,
      "num_input_tokens_seen": 87146175,
      "step": 4048,
      "time_per_iteration": 2.7185354232788086
    },
    {
      "auxiliary_loss_clip": 0.0116042,
      "auxiliary_loss_mlp": 0.01028329,
      "balance_loss_clip": 1.0525949,
      "balance_loss_mlp": 1.01941204,
      "epoch": 0.4868634641976793,
      "flos": 28804344030720.0,
      "grad_norm": 1.7890649458627383,
      "language_loss": 0.72494066,
      "learning_rate": 2.182027121432696e-06,
      "loss": 0.74682814,
      "num_input_tokens_seen": 87166800,
      "step": 4049,
      "time_per_iteration": 3.7137632369995117
    },
    {
      "auxiliary_loss_clip": 0.01167156,
      "auxiliary_loss_mlp": 0.01034765,
      "balance_loss_clip": 1.0550338,
      "balance_loss_mlp": 1.0254848,
      "epoch": 0.4869837070883184,
      "flos": 19025976574080.0,
      "grad_norm": 2.582855225520839,
      "language_loss": 0.82304275,
      "learning_rate": 2.1812513690008054e-06,
      "loss": 0.84506196,
      "num_input_tokens_seen": 87185920,
      "step": 4050,
      "time_per_iteration": 2.691098690032959
    },
    {
      "auxiliary_loss_clip": 0.01155619,
      "auxiliary_loss_mlp": 0.01032426,
      "balance_loss_clip": 1.05330324,
      "balance_loss_mlp": 1.02327073,
      "epoch": 0.4871039499789575,
      "flos": 15121409483520.0,
      "grad_norm": 2.322537785895201,
      "language_loss": 0.79564154,
      "learning_rate": 2.180475589073227e-06,
      "loss": 0.81752193,
      "num_input_tokens_seen": 87203620,
      "step": 4051,
      "time_per_iteration": 2.6592180728912354
    },
    {
      "auxiliary_loss_clip": 0.01158669,
      "auxiliary_loss_mlp": 0.01026645,
      "balance_loss_clip": 1.0506115,
      "balance_loss_mlp": 1.01799083,
      "epoch": 0.48722419286959656,
      "flos": 26174066808960.0,
      "grad_norm": 1.9680424236579184,
      "language_loss": 0.7381056,
      "learning_rate": 2.1796997817676456e-06,
      "loss": 0.75995874,
      "num_input_tokens_seen": 87224630,
      "step": 4052,
      "time_per_iteration": 2.713355779647827
    },
    {
      "auxiliary_loss_clip": 0.01131425,
      "auxiliary_loss_mlp": 0.01025689,
      "balance_loss_clip": 1.04629111,
      "balance_loss_mlp": 1.01668239,
      "epoch": 0.4873444357602357,
      "flos": 24026142349440.0,
      "grad_norm": 2.295700611804765,
      "language_loss": 0.67701364,
      "learning_rate": 2.1789239472017494e-06,
      "loss": 0.69858479,
      "num_input_tokens_seen": 87246280,
      "step": 4053,
      "time_per_iteration": 2.7832446098327637
    },
    {
      "auxiliary_loss_clip": 0.01177286,
      "auxiliary_loss_mlp": 0.01030879,
      "balance_loss_clip": 1.05780804,
      "balance_loss_mlp": 1.02223659,
      "epoch": 0.4874646786508748,
      "flos": 22820441500800.0,
      "grad_norm": 3.7965327528399824,
      "language_loss": 0.73318064,
      "learning_rate": 2.1781480854932326e-06,
      "loss": 0.75526226,
      "num_input_tokens_seen": 87266045,
      "step": 4054,
      "time_per_iteration": 2.684431552886963
    },
    {
      "auxiliary_loss_clip": 0.01159616,
      "auxiliary_loss_mlp": 0.0103282,
      "balance_loss_clip": 1.05414891,
      "balance_loss_mlp": 1.02454638,
      "epoch": 0.48758492154151384,
      "flos": 21287594557440.0,
      "grad_norm": 1.9172477931594611,
      "language_loss": 0.79130077,
      "learning_rate": 2.1773721967597933e-06,
      "loss": 0.81322503,
      "num_input_tokens_seen": 87284495,
      "step": 4055,
      "time_per_iteration": 2.69551157951355
    },
    {
      "auxiliary_loss_clip": 0.01063371,
      "auxiliary_loss_mlp": 0.01001245,
      "balance_loss_clip": 1.01266885,
      "balance_loss_mlp": 1.00008261,
      "epoch": 0.48770516443215295,
      "flos": 62244109180800.0,
      "grad_norm": 0.8476156805217356,
      "language_loss": 0.57351553,
      "learning_rate": 2.1765962811191322e-06,
      "loss": 0.59416175,
      "num_input_tokens_seen": 87338960,
      "step": 4056,
      "time_per_iteration": 3.181565046310425
    },
    {
      "auxiliary_loss_clip": 0.01072294,
      "auxiliary_loss_mlp": 0.01000767,
      "balance_loss_clip": 1.01552606,
      "balance_loss_mlp": 0.99956352,
      "epoch": 0.48782540732279206,
      "flos": 66133451882880.0,
      "grad_norm": 0.8246497029316533,
      "language_loss": 0.61951989,
      "learning_rate": 2.1758203386889566e-06,
      "loss": 0.6402505,
      "num_input_tokens_seen": 87401730,
      "step": 4057,
      "time_per_iteration": 3.2545979022979736
    },
    {
      "auxiliary_loss_clip": 0.0117846,
      "auxiliary_loss_mlp": 0.01037627,
      "balance_loss_clip": 1.05830836,
      "balance_loss_mlp": 1.02826953,
      "epoch": 0.4879456502134311,
      "flos": 14607922608000.0,
      "grad_norm": 2.2278667034407458,
      "language_loss": 0.84319311,
      "learning_rate": 2.1750443695869746e-06,
      "loss": 0.865354,
      "num_input_tokens_seen": 87417300,
      "step": 4058,
      "time_per_iteration": 2.6282265186309814
    },
    {
      "auxiliary_loss_clip": 0.01191931,
      "auxiliary_loss_mlp": 0.01036237,
      "balance_loss_clip": 1.05650067,
      "balance_loss_mlp": 1.02739143,
      "epoch": 0.4880658931040702,
      "flos": 19500464257920.0,
      "grad_norm": 1.9933174507672855,
      "language_loss": 0.85848331,
      "learning_rate": 2.174268373930901e-06,
      "loss": 0.88076496,
      "num_input_tokens_seen": 87434815,
      "step": 4059,
      "time_per_iteration": 2.656975746154785
    },
    {
      "auxiliary_loss_clip": 0.01174385,
      "auxiliary_loss_mlp": 0.01032235,
      "balance_loss_clip": 1.05683029,
      "balance_loss_mlp": 1.02329445,
      "epoch": 0.48818613599470934,
      "flos": 16723060928640.0,
      "grad_norm": 2.5202944822275635,
      "language_loss": 0.80106747,
      "learning_rate": 2.1734923518384537e-06,
      "loss": 0.82313365,
      "num_input_tokens_seen": 87451420,
      "step": 4060,
      "time_per_iteration": 2.616678237915039
    },
    {
      "auxiliary_loss_clip": 0.0116675,
      "auxiliary_loss_mlp": 0.01030806,
      "balance_loss_clip": 1.05256939,
      "balance_loss_mlp": 1.02203274,
      "epoch": 0.4883063788853484,
      "flos": 26756932803840.0,
      "grad_norm": 2.27002691312737,
      "language_loss": 0.82241881,
      "learning_rate": 2.1727163034273547e-06,
      "loss": 0.84439433,
      "num_input_tokens_seen": 87469585,
      "step": 4061,
      "time_per_iteration": 2.701932668685913
    },
    {
      "auxiliary_loss_clip": 0.01174048,
      "auxiliary_loss_mlp": 0.01031331,
      "balance_loss_clip": 1.05585563,
      "balance_loss_mlp": 1.02233696,
      "epoch": 0.4884266217759875,
      "flos": 16763388923520.0,
      "grad_norm": 2.4421484535529547,
      "language_loss": 0.78720713,
      "learning_rate": 2.17194022881533e-06,
      "loss": 0.8092609,
      "num_input_tokens_seen": 87485675,
      "step": 4062,
      "time_per_iteration": 2.6434171199798584
    },
    {
      "auxiliary_loss_clip": 0.01159589,
      "auxiliary_loss_mlp": 0.01031119,
      "balance_loss_clip": 1.05224991,
      "balance_loss_mlp": 1.02164149,
      "epoch": 0.4885468646666266,
      "flos": 24207132003840.0,
      "grad_norm": 1.9848091578361045,
      "language_loss": 0.67732024,
      "learning_rate": 2.1711641281201092e-06,
      "loss": 0.69922733,
      "num_input_tokens_seen": 87505605,
      "step": 4063,
      "time_per_iteration": 2.673769950866699
    },
    {
      "auxiliary_loss_clip": 0.01190873,
      "auxiliary_loss_mlp": 0.00763345,
      "balance_loss_clip": 1.05795443,
      "balance_loss_mlp": 1.00107479,
      "epoch": 0.48866710755726567,
      "flos": 14610795696000.0,
      "grad_norm": 2.480354542276921,
      "language_loss": 0.79825383,
      "learning_rate": 2.1703880014594264e-06,
      "loss": 0.81779605,
      "num_input_tokens_seen": 87523195,
      "step": 4064,
      "time_per_iteration": 2.63419508934021
    },
    {
      "auxiliary_loss_clip": 0.0119403,
      "auxiliary_loss_mlp": 0.0103117,
      "balance_loss_clip": 1.06047428,
      "balance_loss_mlp": 1.02243209,
      "epoch": 0.4887873504479048,
      "flos": 28804451771520.0,
      "grad_norm": 1.8836498624981202,
      "language_loss": 0.73669285,
      "learning_rate": 2.1696118489510182e-06,
      "loss": 0.75894487,
      "num_input_tokens_seen": 87544125,
      "step": 4065,
      "time_per_iteration": 2.72647762298584
    },
    {
      "auxiliary_loss_clip": 0.0119598,
      "auxiliary_loss_mlp": 0.01033573,
      "balance_loss_clip": 1.06083965,
      "balance_loss_mlp": 1.02476299,
      "epoch": 0.48890759333854383,
      "flos": 22784387224320.0,
      "grad_norm": 2.8458967346894757,
      "language_loss": 0.72630858,
      "learning_rate": 2.1688356707126286e-06,
      "loss": 0.74860406,
      "num_input_tokens_seen": 87563745,
      "step": 4066,
      "time_per_iteration": 3.5461723804473877
    },
    {
      "auxiliary_loss_clip": 0.01170431,
      "auxiliary_loss_mlp": 0.01035515,
      "balance_loss_clip": 1.0561502,
      "balance_loss_mlp": 1.02630019,
      "epoch": 0.48902783622918294,
      "flos": 17786088956160.0,
      "grad_norm": 2.454537606476401,
      "language_loss": 0.70140159,
      "learning_rate": 2.168059466862001e-06,
      "loss": 0.72346097,
      "num_input_tokens_seen": 87581895,
      "step": 4067,
      "time_per_iteration": 2.6583621501922607
    },
    {
      "auxiliary_loss_clip": 0.01158265,
      "auxiliary_loss_mlp": 0.01037887,
      "balance_loss_clip": 1.05100632,
      "balance_loss_mlp": 1.02908921,
      "epoch": 0.48914807911982205,
      "flos": 22310294590080.0,
      "grad_norm": 2.2562216665185586,
      "language_loss": 0.82086396,
      "learning_rate": 2.167283237516887e-06,
      "loss": 0.84282547,
      "num_input_tokens_seen": 87600170,
      "step": 4068,
      "time_per_iteration": 3.6160128116607666
    },
    {
      "auxiliary_loss_clip": 0.01167736,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.05263948,
      "balance_loss_mlp": 1.02167964,
      "epoch": 0.4892683220104611,
      "flos": 16363020954240.0,
      "grad_norm": 1.8908051258313423,
      "language_loss": 0.74960434,
      "learning_rate": 2.1665069827950383e-06,
      "loss": 0.77159673,
      "num_input_tokens_seen": 87617455,
      "step": 4069,
      "time_per_iteration": 2.6196818351745605
    },
    {
      "auxiliary_loss_clip": 0.01171913,
      "auxiliary_loss_mlp": 0.01032757,
      "balance_loss_clip": 1.05338478,
      "balance_loss_mlp": 1.0246985,
      "epoch": 0.4893885649011002,
      "flos": 15739144606080.0,
      "grad_norm": 2.6754770708030984,
      "language_loss": 0.87074029,
      "learning_rate": 2.1657307028142126e-06,
      "loss": 0.89278698,
      "num_input_tokens_seen": 87634995,
      "step": 4070,
      "time_per_iteration": 2.6241981983184814
    },
    {
      "auxiliary_loss_clip": 0.01176015,
      "auxiliary_loss_mlp": 0.01031953,
      "balance_loss_clip": 1.05490994,
      "balance_loss_mlp": 1.02263117,
      "epoch": 0.48950880779173933,
      "flos": 28581984887040.0,
      "grad_norm": 2.9309161814844664,
      "language_loss": 0.67474866,
      "learning_rate": 2.164954397692171e-06,
      "loss": 0.69682831,
      "num_input_tokens_seen": 87654420,
      "step": 4071,
      "time_per_iteration": 2.7474043369293213
    },
    {
      "auxiliary_loss_clip": 0.01079579,
      "auxiliary_loss_mlp": 0.01008116,
      "balance_loss_clip": 1.01373863,
      "balance_loss_mlp": 1.00692415,
      "epoch": 0.4896290506823784,
      "flos": 66186310746240.0,
      "grad_norm": 1.094011214278641,
      "language_loss": 0.77236336,
      "learning_rate": 2.164178067546678e-06,
      "loss": 0.79324037,
      "num_input_tokens_seen": 87713585,
      "step": 4072,
      "time_per_iteration": 3.222982883453369
    },
    {
      "auxiliary_loss_clip": 0.01176049,
      "auxiliary_loss_mlp": 0.01031248,
      "balance_loss_clip": 1.05158019,
      "balance_loss_mlp": 1.02256989,
      "epoch": 0.4897492935730175,
      "flos": 12531065207040.0,
      "grad_norm": 1.7406650591428803,
      "language_loss": 0.91314465,
      "learning_rate": 2.163401712495504e-06,
      "loss": 0.93521762,
      "num_input_tokens_seen": 87731280,
      "step": 4073,
      "time_per_iteration": 2.674539804458618
    },
    {
      "auxiliary_loss_clip": 0.01191198,
      "auxiliary_loss_mlp": 0.01029563,
      "balance_loss_clip": 1.05625534,
      "balance_loss_mlp": 1.02110505,
      "epoch": 0.4898695364636566,
      "flos": 23476816679040.0,
      "grad_norm": 2.680389112824541,
      "language_loss": 0.79402924,
      "learning_rate": 2.1626253326564194e-06,
      "loss": 0.81623691,
      "num_input_tokens_seen": 87750230,
      "step": 4074,
      "time_per_iteration": 2.7791383266448975
    },
    {
      "auxiliary_loss_clip": 0.0114068,
      "auxiliary_loss_mlp": 0.01034373,
      "balance_loss_clip": 1.04964566,
      "balance_loss_mlp": 1.02472925,
      "epoch": 0.48998977935429566,
      "flos": 27160209774720.0,
      "grad_norm": 2.0088709989533355,
      "language_loss": 0.77128971,
      "learning_rate": 2.161848928147201e-06,
      "loss": 0.79304022,
      "num_input_tokens_seen": 87770500,
      "step": 4075,
      "time_per_iteration": 3.7063629627227783
    },
    {
      "auxiliary_loss_clip": 0.01194223,
      "auxiliary_loss_mlp": 0.01033,
      "balance_loss_clip": 1.05830097,
      "balance_loss_mlp": 1.02318931,
      "epoch": 0.4901100222449348,
      "flos": 20339588856960.0,
      "grad_norm": 2.0280692446432536,
      "language_loss": 0.80703771,
      "learning_rate": 2.161072499085629e-06,
      "loss": 0.82930994,
      "num_input_tokens_seen": 87789495,
      "step": 4076,
      "time_per_iteration": 2.633770704269409
    },
    {
      "auxiliary_loss_clip": 0.01162798,
      "auxiliary_loss_mlp": 0.01035526,
      "balance_loss_clip": 1.0583756,
      "balance_loss_mlp": 1.02676392,
      "epoch": 0.4902302651355739,
      "flos": 30446359384320.0,
      "grad_norm": 1.5667613451892473,
      "language_loss": 0.831738,
      "learning_rate": 2.160296045589487e-06,
      "loss": 0.8537212,
      "num_input_tokens_seen": 87812955,
      "step": 4077,
      "time_per_iteration": 2.7627410888671875
    },
    {
      "auxiliary_loss_clip": 0.01192746,
      "auxiliary_loss_mlp": 0.0102775,
      "balance_loss_clip": 1.05745196,
      "balance_loss_mlp": 1.01840985,
      "epoch": 0.49035050802621294,
      "flos": 19174180089600.0,
      "grad_norm": 1.7627929328705192,
      "language_loss": 0.70246392,
      "learning_rate": 2.159519567776562e-06,
      "loss": 0.72466886,
      "num_input_tokens_seen": 87832605,
      "step": 4078,
      "time_per_iteration": 2.5906693935394287
    },
    {
      "auxiliary_loss_clip": 0.0119091,
      "auxiliary_loss_mlp": 0.01026379,
      "balance_loss_clip": 1.05578148,
      "balance_loss_mlp": 1.01768303,
      "epoch": 0.49047075091685205,
      "flos": 22228489365120.0,
      "grad_norm": 2.6249476285972135,
      "language_loss": 0.71050459,
      "learning_rate": 2.1587430657646463e-06,
      "loss": 0.73267746,
      "num_input_tokens_seen": 87846040,
      "step": 4079,
      "time_per_iteration": 2.5629935264587402
    },
    {
      "auxiliary_loss_clip": 0.01175461,
      "auxiliary_loss_mlp": 0.01032484,
      "balance_loss_clip": 1.05595112,
      "balance_loss_mlp": 1.023633,
      "epoch": 0.4905909938074911,
      "flos": 20156516213760.0,
      "grad_norm": 2.5866096097589746,
      "language_loss": 0.7790879,
      "learning_rate": 2.157966539671533e-06,
      "loss": 0.80116737,
      "num_input_tokens_seen": 87865680,
      "step": 4080,
      "time_per_iteration": 2.6515655517578125
    },
    {
      "auxiliary_loss_clip": 0.01190126,
      "auxiliary_loss_mlp": 0.00762647,
      "balance_loss_clip": 1.05689311,
      "balance_loss_mlp": 1.00104964,
      "epoch": 0.4907112366981302,
      "flos": 17202217380480.0,
      "grad_norm": 3.8971625319280236,
      "language_loss": 0.67187804,
      "learning_rate": 2.157189989615021e-06,
      "loss": 0.69140577,
      "num_input_tokens_seen": 87884270,
      "step": 4081,
      "time_per_iteration": 2.5685248374938965
    },
    {
      "auxiliary_loss_clip": 0.01170269,
      "auxiliary_loss_mlp": 0.01035786,
      "balance_loss_clip": 1.05960226,
      "balance_loss_mlp": 1.02624941,
      "epoch": 0.4908314795887693,
      "flos": 21688968107520.0,
      "grad_norm": 2.275218507005481,
      "language_loss": 0.74898547,
      "learning_rate": 2.156413415712913e-06,
      "loss": 0.77104598,
      "num_input_tokens_seen": 87906320,
      "step": 4082,
      "time_per_iteration": 2.7588229179382324
    },
    {
      "auxiliary_loss_clip": 0.01144418,
      "auxiliary_loss_mlp": 0.01029002,
      "balance_loss_clip": 1.05477715,
      "balance_loss_mlp": 1.02018106,
      "epoch": 0.4909517224794084,
      "flos": 26213676531840.0,
      "grad_norm": 4.19105253100352,
      "language_loss": 0.78799069,
      "learning_rate": 2.155636818083014e-06,
      "loss": 0.80972481,
      "num_input_tokens_seen": 87927690,
      "step": 4083,
      "time_per_iteration": 2.7043118476867676
    },
    {
      "auxiliary_loss_clip": 0.01192165,
      "auxiliary_loss_mlp": 0.01036989,
      "balance_loss_clip": 1.05930352,
      "balance_loss_mlp": 1.02832246,
      "epoch": 0.4910719653700475,
      "flos": 23148377694720.0,
      "grad_norm": 1.8896012670097704,
      "language_loss": 0.84601772,
      "learning_rate": 2.154860196843134e-06,
      "loss": 0.86830926,
      "num_input_tokens_seen": 87946885,
      "step": 4084,
      "time_per_iteration": 2.67984938621521
    },
    {
      "auxiliary_loss_clip": 0.01179686,
      "auxiliary_loss_mlp": 0.01030228,
      "balance_loss_clip": 1.05768633,
      "balance_loss_mlp": 1.02148962,
      "epoch": 0.4911922082606866,
      "flos": 23331845387520.0,
      "grad_norm": 3.9573934473064467,
      "language_loss": 0.77106279,
      "learning_rate": 2.154083552111085e-06,
      "loss": 0.79316187,
      "num_input_tokens_seen": 87966055,
      "step": 4085,
      "time_per_iteration": 2.6803781986236572
    },
    {
      "auxiliary_loss_clip": 0.01165435,
      "auxiliary_loss_mlp": 0.01030739,
      "balance_loss_clip": 1.05392361,
      "balance_loss_mlp": 1.02170861,
      "epoch": 0.49131245115132566,
      "flos": 29203239542400.0,
      "grad_norm": 2.108692868421273,
      "language_loss": 0.82108676,
      "learning_rate": 2.1533068840046834e-06,
      "loss": 0.84304851,
      "num_input_tokens_seen": 87986320,
      "step": 4086,
      "time_per_iteration": 2.679006814956665
    },
    {
      "auxiliary_loss_clip": 0.01194445,
      "auxiliary_loss_mlp": 0.01030921,
      "balance_loss_clip": 1.05924845,
      "balance_loss_mlp": 1.02182531,
      "epoch": 0.49143269404196477,
      "flos": 20147465986560.0,
      "grad_norm": 3.600743924830791,
      "language_loss": 0.62157351,
      "learning_rate": 2.152530192641749e-06,
      "loss": 0.6438272,
      "num_input_tokens_seen": 88001230,
      "step": 4087,
      "time_per_iteration": 2.595346689224243
    },
    {
      "auxiliary_loss_clip": 0.01191602,
      "auxiliary_loss_mlp": 0.01042938,
      "balance_loss_clip": 1.05854821,
      "balance_loss_mlp": 1.03392005,
      "epoch": 0.4915529369326039,
      "flos": 24389809597440.0,
      "grad_norm": 2.7582326240960815,
      "language_loss": 0.72784078,
      "learning_rate": 2.1517534781401068e-06,
      "loss": 0.7501862,
      "num_input_tokens_seen": 88019110,
      "step": 4088,
      "time_per_iteration": 2.614706516265869
    },
    {
      "auxiliary_loss_clip": 0.01192233,
      "auxiliary_loss_mlp": 0.01028686,
      "balance_loss_clip": 1.05837214,
      "balance_loss_mlp": 1.01950705,
      "epoch": 0.49167317982324293,
      "flos": 10524305197440.0,
      "grad_norm": 8.38546814405424,
      "language_loss": 0.69775486,
      "learning_rate": 2.150976740617581e-06,
      "loss": 0.71996409,
      "num_input_tokens_seen": 88035670,
      "step": 4089,
      "time_per_iteration": 2.7011559009552
    },
    {
      "auxiliary_loss_clip": 0.01176006,
      "auxiliary_loss_mlp": 0.01031232,
      "balance_loss_clip": 1.05345404,
      "balance_loss_mlp": 1.02247047,
      "epoch": 0.49179342271388204,
      "flos": 25593427457280.0,
      "grad_norm": 2.065985655200042,
      "language_loss": 0.71958524,
      "learning_rate": 2.150199980192006e-06,
      "loss": 0.74165761,
      "num_input_tokens_seen": 88054790,
      "step": 4090,
      "time_per_iteration": 2.6528127193450928
    },
    {
      "auxiliary_loss_clip": 0.01191851,
      "auxiliary_loss_mlp": 0.01029672,
      "balance_loss_clip": 1.05851328,
      "balance_loss_mlp": 1.02130365,
      "epoch": 0.49191366560452116,
      "flos": 21102043875840.0,
      "grad_norm": 2.4698831470930034,
      "language_loss": 0.80812955,
      "learning_rate": 2.1494231969812114e-06,
      "loss": 0.8303448,
      "num_input_tokens_seen": 88073780,
      "step": 4091,
      "time_per_iteration": 2.6479780673980713
    },
    {
      "auxiliary_loss_clip": 0.01194183,
      "auxiliary_loss_mlp": 0.01044208,
      "balance_loss_clip": 1.05958605,
      "balance_loss_mlp": 1.03530931,
      "epoch": 0.4920339084951602,
      "flos": 26067520091520.0,
      "grad_norm": 2.7427482544158357,
      "language_loss": 0.81500626,
      "learning_rate": 2.1486463911030372e-06,
      "loss": 0.83739018,
      "num_input_tokens_seen": 88094430,
      "step": 4092,
      "time_per_iteration": 3.572554588317871
    },
    {
      "auxiliary_loss_clip": 0.01190403,
      "auxiliary_loss_mlp": 0.01031914,
      "balance_loss_clip": 1.05619597,
      "balance_loss_mlp": 1.02335525,
      "epoch": 0.4921541513857993,
      "flos": 25081269384960.0,
      "grad_norm": 5.184153463691187,
      "language_loss": 0.74426615,
      "learning_rate": 2.147869562675324e-06,
      "loss": 0.76648927,
      "num_input_tokens_seen": 88113400,
      "step": 4093,
      "time_per_iteration": 2.7257184982299805
    },
    {
      "auxiliary_loss_clip": 0.0118033,
      "auxiliary_loss_mlp": 0.01034066,
      "balance_loss_clip": 1.05762935,
      "balance_loss_mlp": 1.02455938,
      "epoch": 0.49227439427643843,
      "flos": 24389809597440.0,
      "grad_norm": 3.138806801182321,
      "language_loss": 0.72522038,
      "learning_rate": 2.147092711815915e-06,
      "loss": 0.7473644,
      "num_input_tokens_seen": 88132750,
      "step": 4094,
      "time_per_iteration": 4.640148878097534
    },
    {
      "auxiliary_loss_clip": 0.01192398,
      "auxiliary_loss_mlp": 0.01031542,
      "balance_loss_clip": 1.05730116,
      "balance_loss_mlp": 1.0222795,
      "epoch": 0.4923946371670775,
      "flos": 11363753018880.0,
      "grad_norm": 2.340384280678344,
      "language_loss": 0.86671209,
      "learning_rate": 2.1463158386426593e-06,
      "loss": 0.88895154,
      "num_input_tokens_seen": 88150560,
      "step": 4095,
      "time_per_iteration": 2.5852346420288086
    },
    {
      "auxiliary_loss_clip": 0.01179738,
      "auxiliary_loss_mlp": 0.01032997,
      "balance_loss_clip": 1.05607009,
      "balance_loss_mlp": 1.02378237,
      "epoch": 0.4925148800577166,
      "flos": 30445964334720.0,
      "grad_norm": 2.324194523577912,
      "language_loss": 0.77251661,
      "learning_rate": 2.145538943273407e-06,
      "loss": 0.79464394,
      "num_input_tokens_seen": 88170835,
      "step": 4096,
      "time_per_iteration": 2.72076153755188
    },
    {
      "auxiliary_loss_clip": 0.0119427,
      "auxiliary_loss_mlp": 0.01031558,
      "balance_loss_clip": 1.05917239,
      "balance_loss_mlp": 1.02276039,
      "epoch": 0.49263512294835565,
      "flos": 20850454039680.0,
      "grad_norm": 1.9242588292907938,
      "language_loss": 0.72135848,
      "learning_rate": 2.144762025826013e-06,
      "loss": 0.74361676,
      "num_input_tokens_seen": 88189925,
      "step": 4097,
      "time_per_iteration": 2.632282257080078
    },
    {
      "auxiliary_loss_clip": 0.01192353,
      "auxiliary_loss_mlp": 0.01034116,
      "balance_loss_clip": 1.05756199,
      "balance_loss_mlp": 1.02434063,
      "epoch": 0.49275536583899476,
      "flos": 23767477534080.0,
      "grad_norm": 3.633039580357819,
      "language_loss": 0.87366414,
      "learning_rate": 2.143985086418334e-06,
      "loss": 0.89592886,
      "num_input_tokens_seen": 88205105,
      "step": 4098,
      "time_per_iteration": 2.5777928829193115
    },
    {
      "auxiliary_loss_clip": 0.01175761,
      "auxiliary_loss_mlp": 0.01031346,
      "balance_loss_clip": 1.05445695,
      "balance_loss_mlp": 1.02286386,
      "epoch": 0.4928756087296339,
      "flos": 22273522041600.0,
      "grad_norm": 1.3625240603336999,
      "language_loss": 0.7673319,
      "learning_rate": 2.1432081251682324e-06,
      "loss": 0.78940296,
      "num_input_tokens_seen": 88225475,
      "step": 4099,
      "time_per_iteration": 2.6984612941741943
    },
    {
      "auxiliary_loss_clip": 0.01176694,
      "auxiliary_loss_mlp": 0.0103156,
      "balance_loss_clip": 1.05642927,
      "balance_loss_mlp": 1.02337027,
      "epoch": 0.49299585162027293,
      "flos": 19645471463040.0,
      "grad_norm": 1.8492846920617036,
      "language_loss": 0.87400973,
      "learning_rate": 2.142431142193572e-06,
      "loss": 0.8960923,
      "num_input_tokens_seen": 88243255,
      "step": 4100,
      "time_per_iteration": 2.6612966060638428
    },
    {
      "auxiliary_loss_clip": 0.01178805,
      "auxiliary_loss_mlp": 0.01028621,
      "balance_loss_clip": 1.05653644,
      "balance_loss_mlp": 1.02003777,
      "epoch": 0.49311609451091204,
      "flos": 38837138497920.0,
      "grad_norm": 2.4656352671515864,
      "language_loss": 0.71644711,
      "learning_rate": 2.1416541376122207e-06,
      "loss": 0.73852146,
      "num_input_tokens_seen": 88263435,
      "step": 4101,
      "time_per_iteration": 2.802058219909668
    },
    {
      "auxiliary_loss_clip": 0.01175124,
      "auxiliary_loss_mlp": 0.01025876,
      "balance_loss_clip": 1.05444717,
      "balance_loss_mlp": 1.01755512,
      "epoch": 0.49323633740155115,
      "flos": 28329102161280.0,
      "grad_norm": 1.8509505475438512,
      "language_loss": 0.73331219,
      "learning_rate": 2.1408771115420496e-06,
      "loss": 0.75532216,
      "num_input_tokens_seen": 88283295,
      "step": 4102,
      "time_per_iteration": 3.633453130722046
    },
    {
      "auxiliary_loss_clip": 0.01176693,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 1.05812621,
      "balance_loss_mlp": 1.01929355,
      "epoch": 0.4933565802921902,
      "flos": 21135584200320.0,
      "grad_norm": 1.898820557800959,
      "language_loss": 0.64834547,
      "learning_rate": 2.140100064100932e-06,
      "loss": 0.6703918,
      "num_input_tokens_seen": 88299270,
      "step": 4103,
      "time_per_iteration": 2.6915595531463623
    },
    {
      "auxiliary_loss_clip": 0.01173866,
      "auxiliary_loss_mlp": 0.0103062,
      "balance_loss_clip": 1.05475092,
      "balance_loss_mlp": 1.02206659,
      "epoch": 0.4934768231828293,
      "flos": 18039007595520.0,
      "grad_norm": 2.024410473055194,
      "language_loss": 0.76165098,
      "learning_rate": 2.139322995406746e-06,
      "loss": 0.78369588,
      "num_input_tokens_seen": 88316905,
      "step": 4104,
      "time_per_iteration": 2.6805336475372314
    },
    {
      "auxiliary_loss_clip": 0.01180916,
      "auxiliary_loss_mlp": 0.01039201,
      "balance_loss_clip": 1.05954492,
      "balance_loss_mlp": 1.0300281,
      "epoch": 0.4935970660734684,
      "flos": 23469957181440.0,
      "grad_norm": 2.1014249109841923,
      "language_loss": 0.7986142,
      "learning_rate": 2.1385459055773727e-06,
      "loss": 0.82081532,
      "num_input_tokens_seen": 88335095,
      "step": 4105,
      "time_per_iteration": 2.7233142852783203
    },
    {
      "auxiliary_loss_clip": 0.01190267,
      "auxiliary_loss_mlp": 0.01031015,
      "balance_loss_clip": 1.05854392,
      "balance_loss_mlp": 1.02291441,
      "epoch": 0.4937173089641075,
      "flos": 64479258840960.0,
      "grad_norm": 2.137418312552909,
      "language_loss": 0.74042624,
      "learning_rate": 2.137768794730696e-06,
      "loss": 0.76263905,
      "num_input_tokens_seen": 88358545,
      "step": 4106,
      "time_per_iteration": 3.0405869483947754
    },
    {
      "auxiliary_loss_clip": 0.01183037,
      "auxiliary_loss_mlp": 0.01034833,
      "balance_loss_clip": 1.05656815,
      "balance_loss_mlp": 1.02539182,
      "epoch": 0.4938375518547466,
      "flos": 22346025644160.0,
      "grad_norm": 2.1062094508515545,
      "language_loss": 0.80910152,
      "learning_rate": 2.1369916629846026e-06,
      "loss": 0.83128023,
      "num_input_tokens_seen": 88378295,
      "step": 4107,
      "time_per_iteration": 2.664571523666382
    },
    {
      "auxiliary_loss_clip": 0.01165136,
      "auxiliary_loss_mlp": 0.01028968,
      "balance_loss_clip": 1.05269265,
      "balance_loss_mlp": 1.02036738,
      "epoch": 0.4939577947453857,
      "flos": 17858700299520.0,
      "grad_norm": 2.450992258678159,
      "language_loss": 0.75299025,
      "learning_rate": 2.136214510456983e-06,
      "loss": 0.77493125,
      "num_input_tokens_seen": 88396750,
      "step": 4108,
      "time_per_iteration": 2.7818474769592285
    },
    {
      "auxiliary_loss_clip": 0.01069103,
      "auxiliary_loss_mlp": 0.0100176,
      "balance_loss_clip": 1.01628327,
      "balance_loss_mlp": 1.00068665,
      "epoch": 0.49407803763602476,
      "flos": 70066746875520.0,
      "grad_norm": 0.8843873484410988,
      "language_loss": 0.6315068,
      "learning_rate": 2.1354373372657296e-06,
      "loss": 0.65221542,
      "num_input_tokens_seen": 88455190,
      "step": 4109,
      "time_per_iteration": 3.2368855476379395
    },
    {
      "auxiliary_loss_clip": 0.01164795,
      "auxiliary_loss_mlp": 0.01030374,
      "balance_loss_clip": 1.0557276,
      "balance_loss_mlp": 1.02104557,
      "epoch": 0.49419828052666387,
      "flos": 24317485562880.0,
      "grad_norm": 1.9501586050823863,
      "language_loss": 0.71169806,
      "learning_rate": 2.1346601435287404e-06,
      "loss": 0.73364973,
      "num_input_tokens_seen": 88477460,
      "step": 4110,
      "time_per_iteration": 2.7555577754974365
    },
    {
      "auxiliary_loss_clip": 0.01149877,
      "auxiliary_loss_mlp": 0.01026894,
      "balance_loss_clip": 1.05499816,
      "balance_loss_mlp": 1.01868033,
      "epoch": 0.494318523417303,
      "flos": 29386060790400.0,
      "grad_norm": 1.892784298301285,
      "language_loss": 0.80746222,
      "learning_rate": 2.1338829293639144e-06,
      "loss": 0.82922995,
      "num_input_tokens_seen": 88497820,
      "step": 4111,
      "time_per_iteration": 2.744398593902588
    },
    {
      "auxiliary_loss_clip": 0.01191199,
      "auxiliary_loss_mlp": 0.01032228,
      "balance_loss_clip": 1.05695963,
      "balance_loss_mlp": 1.02334666,
      "epoch": 0.49443876630794203,
      "flos": 15268284195840.0,
      "grad_norm": 1.9185360885467555,
      "language_loss": 0.83056849,
      "learning_rate": 2.1331056948891547e-06,
      "loss": 0.85280275,
      "num_input_tokens_seen": 88514920,
      "step": 4112,
      "time_per_iteration": 2.6070663928985596
    },
    {
      "auxiliary_loss_clip": 0.01178549,
      "auxiliary_loss_mlp": 0.0103321,
      "balance_loss_clip": 1.05259061,
      "balance_loss_mlp": 1.02409613,
      "epoch": 0.49455900919858115,
      "flos": 12347453859840.0,
      "grad_norm": 2.183207404630501,
      "language_loss": 0.76481605,
      "learning_rate": 2.1323284402223666e-06,
      "loss": 0.7869336,
      "num_input_tokens_seen": 88530910,
      "step": 4113,
      "time_per_iteration": 2.6275291442871094
    },
    {
      "auxiliary_loss_clip": 0.01160974,
      "auxiliary_loss_mlp": 0.01027334,
      "balance_loss_clip": 1.0543921,
      "balance_loss_mlp": 1.01906645,
      "epoch": 0.4946792520892202,
      "flos": 22779610715520.0,
      "grad_norm": 1.8505861315153798,
      "language_loss": 0.88279837,
      "learning_rate": 2.1315511654814597e-06,
      "loss": 0.90468144,
      "num_input_tokens_seen": 88549320,
      "step": 4114,
      "time_per_iteration": 2.698796272277832
    },
    {
      "auxiliary_loss_clip": 0.01162969,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.0515604,
      "balance_loss_mlp": 1.01858807,
      "epoch": 0.4947994949798593,
      "flos": 23148126299520.0,
      "grad_norm": 2.0360824014265804,
      "language_loss": 0.7873112,
      "learning_rate": 2.1307738707843456e-06,
      "loss": 0.80920827,
      "num_input_tokens_seen": 88568985,
      "step": 4115,
      "time_per_iteration": 2.6960103511810303
    },
    {
      "auxiliary_loss_clip": 0.01197536,
      "auxiliary_loss_mlp": 0.01036478,
      "balance_loss_clip": 1.06140828,
      "balance_loss_mlp": 1.02611876,
      "epoch": 0.4949197378704984,
      "flos": 23659997063040.0,
      "grad_norm": 2.1780201589026333,
      "language_loss": 0.69651175,
      "learning_rate": 2.1299965562489385e-06,
      "loss": 0.71885192,
      "num_input_tokens_seen": 88588790,
      "step": 4116,
      "time_per_iteration": 2.6303012371063232
    },
    {
      "auxiliary_loss_clip": 0.01145528,
      "auxiliary_loss_mlp": 0.01034551,
      "balance_loss_clip": 1.05098903,
      "balance_loss_mlp": 1.02525878,
      "epoch": 0.4950399807611375,
      "flos": 26911493026560.0,
      "grad_norm": 1.6376127394821227,
      "language_loss": 0.78746343,
      "learning_rate": 2.129219221993158e-06,
      "loss": 0.80926418,
      "num_input_tokens_seen": 88613575,
      "step": 4117,
      "time_per_iteration": 2.7591121196746826
    },
    {
      "auxiliary_loss_clip": 0.01079041,
      "auxiliary_loss_mlp": 0.01004384,
      "balance_loss_clip": 1.01378155,
      "balance_loss_mlp": 1.00333452,
      "epoch": 0.4951602236517766,
      "flos": 67315270187520.0,
      "grad_norm": 0.7887990182969797,
      "language_loss": 0.59863639,
      "learning_rate": 2.128441868134924e-06,
      "loss": 0.61947072,
      "num_input_tokens_seen": 88675510,
      "step": 4118,
      "time_per_iteration": 3.214782953262329
    },
    {
      "auxiliary_loss_clip": 0.01162094,
      "auxiliary_loss_mlp": 0.01036965,
      "balance_loss_clip": 1.05699813,
      "balance_loss_mlp": 1.02841187,
      "epoch": 0.4952804665424157,
      "flos": 19901442758400.0,
      "grad_norm": 2.969785767052511,
      "language_loss": 0.83206224,
      "learning_rate": 2.1276644947921606e-06,
      "loss": 0.85405278,
      "num_input_tokens_seen": 88694425,
      "step": 4119,
      "time_per_iteration": 3.566817283630371
    },
    {
      "auxiliary_loss_clip": 0.0115934,
      "auxiliary_loss_mlp": 0.01029182,
      "balance_loss_clip": 1.05315256,
      "balance_loss_mlp": 1.02031291,
      "epoch": 0.49540070943305475,
      "flos": 18806813740800.0,
      "grad_norm": 2.1526064456423755,
      "language_loss": 0.82761651,
      "learning_rate": 2.126887102082795e-06,
      "loss": 0.84950173,
      "num_input_tokens_seen": 88714450,
      "step": 4120,
      "time_per_iteration": 3.6059393882751465
    },
    {
      "auxiliary_loss_clip": 0.01190723,
      "auxiliary_loss_mlp": 0.01031633,
      "balance_loss_clip": 1.05776143,
      "balance_loss_mlp": 1.02345562,
      "epoch": 0.49552095232369386,
      "flos": 24934179191040.0,
      "grad_norm": 2.2050761948306548,
      "language_loss": 0.70469499,
      "learning_rate": 2.126109690124757e-06,
      "loss": 0.72691858,
      "num_input_tokens_seen": 88735265,
      "step": 4121,
      "time_per_iteration": 3.6053690910339355
    },
    {
      "auxiliary_loss_clip": 0.0119192,
      "auxiliary_loss_mlp": 0.01034793,
      "balance_loss_clip": 1.0580523,
      "balance_loss_mlp": 1.02599549,
      "epoch": 0.495641195214333,
      "flos": 22857249962880.0,
      "grad_norm": 2.422071379160058,
      "language_loss": 0.71061218,
      "learning_rate": 2.1253322590359786e-06,
      "loss": 0.73287934,
      "num_input_tokens_seen": 88754600,
      "step": 4122,
      "time_per_iteration": 2.6651580333709717
    },
    {
      "auxiliary_loss_clip": 0.01192682,
      "auxiliary_loss_mlp": 0.01031547,
      "balance_loss_clip": 1.05865312,
      "balance_loss_mlp": 1.02240944,
      "epoch": 0.49576143810497203,
      "flos": 25769748343680.0,
      "grad_norm": 2.7740095111561756,
      "language_loss": 0.73658752,
      "learning_rate": 2.124554808934397e-06,
      "loss": 0.75882983,
      "num_input_tokens_seen": 88775180,
      "step": 4123,
      "time_per_iteration": 2.671278476715088
    },
    {
      "auxiliary_loss_clip": 0.01180468,
      "auxiliary_loss_mlp": 0.00763329,
      "balance_loss_clip": 1.05721331,
      "balance_loss_mlp": 1.0012362,
      "epoch": 0.49588168099561114,
      "flos": 22128838058880.0,
      "grad_norm": 1.8189488385571733,
      "language_loss": 0.73188287,
      "learning_rate": 2.1237773399379496e-06,
      "loss": 0.75132084,
      "num_input_tokens_seen": 88796145,
      "step": 4124,
      "time_per_iteration": 2.6684987545013428
    },
    {
      "auxiliary_loss_clip": 0.01192873,
      "auxiliary_loss_mlp": 0.01029146,
      "balance_loss_clip": 1.05764687,
      "balance_loss_mlp": 1.01987123,
      "epoch": 0.49600192388625025,
      "flos": 24387331559040.0,
      "grad_norm": 1.832467613082472,
      "language_loss": 0.86891752,
      "learning_rate": 2.122999852164578e-06,
      "loss": 0.89113772,
      "num_input_tokens_seen": 88816765,
      "step": 4125,
      "time_per_iteration": 2.637770652770996
    },
    {
      "auxiliary_loss_clip": 0.01184799,
      "auxiliary_loss_mlp": 0.01035485,
      "balance_loss_clip": 1.05805302,
      "balance_loss_mlp": 1.02579904,
      "epoch": 0.4961221667768893,
      "flos": 22857429530880.0,
      "grad_norm": 2.4853514392951985,
      "language_loss": 0.58642226,
      "learning_rate": 2.122222345732227e-06,
      "loss": 0.60862511,
      "num_input_tokens_seen": 88836680,
      "step": 4126,
      "time_per_iteration": 2.6891143321990967
    },
    {
      "auxiliary_loss_clip": 0.01112732,
      "auxiliary_loss_mlp": 0.01033976,
      "balance_loss_clip": 1.04336858,
      "balance_loss_mlp": 1.02515435,
      "epoch": 0.4962424096675284,
      "flos": 17858089768320.0,
      "grad_norm": 2.4721699574509213,
      "language_loss": 0.83306223,
      "learning_rate": 2.121444820758843e-06,
      "loss": 0.85452926,
      "num_input_tokens_seen": 88855320,
      "step": 4127,
      "time_per_iteration": 2.7437541484832764
    },
    {
      "auxiliary_loss_clip": 0.0119545,
      "auxiliary_loss_mlp": 0.01031105,
      "balance_loss_clip": 1.05843782,
      "balance_loss_mlp": 1.02145469,
      "epoch": 0.49636265255816747,
      "flos": 21793611404160.0,
      "grad_norm": 3.1887380581140854,
      "language_loss": 0.78560305,
      "learning_rate": 2.120667277362376e-06,
      "loss": 0.8078686,
      "num_input_tokens_seen": 88874035,
      "step": 4128,
      "time_per_iteration": 3.4914731979370117
    },
    {
      "auxiliary_loss_clip": 0.01167203,
      "auxiliary_loss_mlp": 0.01036576,
      "balance_loss_clip": 1.05583715,
      "balance_loss_mlp": 1.02760577,
      "epoch": 0.4964828954488066,
      "flos": 16358603581440.0,
      "grad_norm": 2.4346996431447083,
      "language_loss": 0.85072124,
      "learning_rate": 2.1198897156607796e-06,
      "loss": 0.87275898,
      "num_input_tokens_seen": 88891390,
      "step": 4129,
      "time_per_iteration": 2.6876060962677
    },
    {
      "auxiliary_loss_clip": 0.01192819,
      "auxiliary_loss_mlp": 0.0103707,
      "balance_loss_clip": 1.05709982,
      "balance_loss_mlp": 1.02834368,
      "epoch": 0.4966031383394457,
      "flos": 24711101775360.0,
      "grad_norm": 2.5463482114917215,
      "language_loss": 0.74479651,
      "learning_rate": 2.1191121357720085e-06,
      "loss": 0.76709539,
      "num_input_tokens_seen": 88909450,
      "step": 4130,
      "time_per_iteration": 2.600074529647827
    },
    {
      "auxiliary_loss_clip": 0.01176606,
      "auxiliary_loss_mlp": 0.01030941,
      "balance_loss_clip": 1.05638528,
      "balance_loss_mlp": 1.0220542,
      "epoch": 0.49672338123008475,
      "flos": 22930615491840.0,
      "grad_norm": 2.2034697597503405,
      "language_loss": 0.75276172,
      "learning_rate": 2.1183345378140206e-06,
      "loss": 0.7748372,
      "num_input_tokens_seen": 88929195,
      "step": 4131,
      "time_per_iteration": 2.6820242404937744
    },
    {
      "auxiliary_loss_clip": 0.01070096,
      "auxiliary_loss_mlp": 0.01000581,
      "balance_loss_clip": 1.01383817,
      "balance_loss_mlp": 0.99941915,
      "epoch": 0.49684362412072386,
      "flos": 65976736844160.0,
      "grad_norm": 0.8559206221844136,
      "language_loss": 0.61866874,
      "learning_rate": 2.1175569219047783e-06,
      "loss": 0.63937551,
      "num_input_tokens_seen": 88990635,
      "step": 4132,
      "time_per_iteration": 3.3332359790802
    },
    {
      "auxiliary_loss_clip": 0.01148863,
      "auxiliary_loss_mlp": 0.01034695,
      "balance_loss_clip": 1.05454803,
      "balance_loss_mlp": 1.02611804,
      "epoch": 0.49696386701136297,
      "flos": 19971288754560.0,
      "grad_norm": 1.8064653785483096,
      "language_loss": 0.73745573,
      "learning_rate": 2.1167792881622437e-06,
      "loss": 0.75929135,
      "num_input_tokens_seen": 89009655,
      "step": 4133,
      "time_per_iteration": 2.6461193561553955
    },
    {
      "auxiliary_loss_clip": 0.01193871,
      "auxiliary_loss_mlp": 0.00762968,
      "balance_loss_clip": 1.05917692,
      "balance_loss_mlp": 1.00122285,
      "epoch": 0.497084109902002,
      "flos": 24750819239040.0,
      "grad_norm": 1.6777618709455722,
      "language_loss": 0.8122772,
      "learning_rate": 2.116001636704384e-06,
      "loss": 0.83184564,
      "num_input_tokens_seen": 89030040,
      "step": 4134,
      "time_per_iteration": 2.664649486541748
    },
    {
      "auxiliary_loss_clip": 0.01196572,
      "auxiliary_loss_mlp": 0.00762717,
      "balance_loss_clip": 1.06221509,
      "balance_loss_mlp": 1.00117576,
      "epoch": 0.49720435279264114,
      "flos": 21871825269120.0,
      "grad_norm": 1.8745136757980887,
      "language_loss": 0.80357921,
      "learning_rate": 2.1152239676491685e-06,
      "loss": 0.82317215,
      "num_input_tokens_seen": 89048145,
      "step": 4135,
      "time_per_iteration": 2.64528226852417
    },
    {
      "auxiliary_loss_clip": 0.01176854,
      "auxiliary_loss_mlp": 0.01032207,
      "balance_loss_clip": 1.05697083,
      "balance_loss_mlp": 1.0238626,
      "epoch": 0.49732459568328025,
      "flos": 23805794367360.0,
      "grad_norm": 1.8438847503789952,
      "language_loss": 0.73296893,
      "learning_rate": 2.114446281114569e-06,
      "loss": 0.7550596,
      "num_input_tokens_seen": 89067165,
      "step": 4136,
      "time_per_iteration": 2.7043991088867188
    },
    {
      "auxiliary_loss_clip": 0.01190074,
      "auxiliary_loss_mlp": 0.01042465,
      "balance_loss_clip": 1.05783939,
      "balance_loss_mlp": 1.0333693,
      "epoch": 0.4974448385739193,
      "flos": 20047742853120.0,
      "grad_norm": 2.067265376411905,
      "language_loss": 0.76722962,
      "learning_rate": 2.1136685772185587e-06,
      "loss": 0.78955501,
      "num_input_tokens_seen": 89086190,
      "step": 4137,
      "time_per_iteration": 2.6662509441375732
    },
    {
      "auxiliary_loss_clip": 0.01178926,
      "auxiliary_loss_mlp": 0.01033878,
      "balance_loss_clip": 1.0552423,
      "balance_loss_mlp": 1.02496123,
      "epoch": 0.4975650814645584,
      "flos": 24821347593600.0,
      "grad_norm": 1.8042446978655557,
      "language_loss": 0.78446251,
      "learning_rate": 2.1128908560791163e-06,
      "loss": 0.80659056,
      "num_input_tokens_seen": 89106020,
      "step": 4138,
      "time_per_iteration": 2.717573642730713
    },
    {
      "auxiliary_loss_clip": 0.01194874,
      "auxiliary_loss_mlp": 0.01036299,
      "balance_loss_clip": 1.05981767,
      "balance_loss_mlp": 1.02682209,
      "epoch": 0.4976853243551975,
      "flos": 19829477859840.0,
      "grad_norm": 2.0046250971027115,
      "language_loss": 0.78404808,
      "learning_rate": 2.1121131178142203e-06,
      "loss": 0.80635977,
      "num_input_tokens_seen": 89125385,
      "step": 4139,
      "time_per_iteration": 2.585184097290039
    },
    {
      "auxiliary_loss_clip": 0.01132733,
      "auxiliary_loss_mlp": 0.01027283,
      "balance_loss_clip": 1.05037522,
      "balance_loss_mlp": 1.01895642,
      "epoch": 0.4978055672458366,
      "flos": 23142990654720.0,
      "grad_norm": 3.0288944413610035,
      "language_loss": 0.82454658,
      "learning_rate": 2.1113353625418544e-06,
      "loss": 0.84614664,
      "num_input_tokens_seen": 89143935,
      "step": 4140,
      "time_per_iteration": 2.767960548400879
    },
    {
      "auxiliary_loss_clip": 0.01170811,
      "auxiliary_loss_mlp": 0.01029725,
      "balance_loss_clip": 1.05767643,
      "balance_loss_mlp": 1.02157664,
      "epoch": 0.4979258101364757,
      "flos": 15559914718080.0,
      "grad_norm": 1.6802655407893772,
      "language_loss": 0.79284483,
      "learning_rate": 2.1105575903800017e-06,
      "loss": 0.81485021,
      "num_input_tokens_seen": 89162655,
      "step": 4141,
      "time_per_iteration": 2.598950147628784
    },
    {
      "auxiliary_loss_clip": 0.01158953,
      "auxiliary_loss_mlp": 0.01037883,
      "balance_loss_clip": 1.05531275,
      "balance_loss_mlp": 1.02943075,
      "epoch": 0.4980460530271148,
      "flos": 26356169784960.0,
      "grad_norm": 2.6127325739877048,
      "language_loss": 0.85255706,
      "learning_rate": 2.1097798014466502e-06,
      "loss": 0.87452543,
      "num_input_tokens_seen": 89182255,
      "step": 4142,
      "time_per_iteration": 2.7375314235687256
    },
    {
      "auxiliary_loss_clip": 0.01194045,
      "auxiliary_loss_mlp": 0.01030783,
      "balance_loss_clip": 1.05780041,
      "balance_loss_mlp": 1.02096581,
      "epoch": 0.49816629591775385,
      "flos": 17274541415040.0,
      "grad_norm": 3.233231350790934,
      "language_loss": 0.59100425,
      "learning_rate": 2.109001995859791e-06,
      "loss": 0.61325252,
      "num_input_tokens_seen": 89201155,
      "step": 4143,
      "time_per_iteration": 2.6453449726104736
    },
    {
      "auxiliary_loss_clip": 0.01066866,
      "auxiliary_loss_mlp": 0.01000427,
      "balance_loss_clip": 1.01438618,
      "balance_loss_mlp": 0.99932432,
      "epoch": 0.49828653880839296,
      "flos": 64930947344640.0,
      "grad_norm": 0.7935727005115162,
      "language_loss": 0.60021752,
      "learning_rate": 2.108224173737415e-06,
      "loss": 0.62089044,
      "num_input_tokens_seen": 89264455,
      "step": 4144,
      "time_per_iteration": 3.187645673751831
    },
    {
      "auxiliary_loss_clip": 0.01170155,
      "auxiliary_loss_mlp": 0.01032784,
      "balance_loss_clip": 1.05376768,
      "balance_loss_mlp": 1.02271652,
      "epoch": 0.498406781699032,
      "flos": 27484806003840.0,
      "grad_norm": 1.8235383063923987,
      "language_loss": 0.76114064,
      "learning_rate": 2.1074463351975183e-06,
      "loss": 0.78317004,
      "num_input_tokens_seen": 89283340,
      "step": 4145,
      "time_per_iteration": 3.6463682651519775
    },
    {
      "auxiliary_loss_clip": 0.01189158,
      "auxiliary_loss_mlp": 0.01030563,
      "balance_loss_clip": 1.05690885,
      "balance_loss_mlp": 1.02196252,
      "epoch": 0.49852702458967113,
      "flos": 31499870307840.0,
      "grad_norm": 2.013927821423308,
      "language_loss": 0.71412945,
      "learning_rate": 2.106668480358098e-06,
      "loss": 0.73632669,
      "num_input_tokens_seen": 89303565,
      "step": 4146,
      "time_per_iteration": 2.676730155944824
    },
    {
      "auxiliary_loss_clip": 0.01193223,
      "auxiliary_loss_mlp": 0.01036029,
      "balance_loss_clip": 1.0579741,
      "balance_loss_mlp": 1.02589083,
      "epoch": 0.49864726748031024,
      "flos": 22852868503680.0,
      "grad_norm": 2.1285201251450787,
      "language_loss": 0.71252143,
      "learning_rate": 2.105890609337154e-06,
      "loss": 0.73481393,
      "num_input_tokens_seen": 89322080,
      "step": 4147,
      "time_per_iteration": 3.5866782665252686
    },
    {
      "auxiliary_loss_clip": 0.01069525,
      "auxiliary_loss_mlp": 0.01004214,
      "balance_loss_clip": 1.01339841,
      "balance_loss_mlp": 1.00297987,
      "epoch": 0.4987675103709493,
      "flos": 70405708544640.0,
      "grad_norm": 0.6882927817160689,
      "language_loss": 0.63724631,
      "learning_rate": 2.1051127222526883e-06,
      "loss": 0.65798366,
      "num_input_tokens_seen": 89394195,
      "step": 4148,
      "time_per_iteration": 3.34220027923584
    },
    {
      "auxiliary_loss_clip": 0.01163473,
      "auxiliary_loss_mlp": 0.01034189,
      "balance_loss_clip": 1.0526644,
      "balance_loss_mlp": 1.0259043,
      "epoch": 0.4988877532615884,
      "flos": 28767571482240.0,
      "grad_norm": 1.6455566773616732,
      "language_loss": 0.81171691,
      "learning_rate": 2.1043348192227067e-06,
      "loss": 0.83369356,
      "num_input_tokens_seen": 89414565,
      "step": 4149,
      "time_per_iteration": 2.7642533779144287
    },
    {
      "auxiliary_loss_clip": 0.01189078,
      "auxiliary_loss_mlp": 0.01029267,
      "balance_loss_clip": 1.05891919,
      "balance_loss_mlp": 1.02080941,
      "epoch": 0.4990079961522275,
      "flos": 16872700988160.0,
      "grad_norm": 3.066251921669589,
      "language_loss": 0.61958945,
      "learning_rate": 2.1035569003652156e-06,
      "loss": 0.64177287,
      "num_input_tokens_seen": 89433195,
      "step": 4150,
      "time_per_iteration": 2.698169469833374
    },
    {
      "auxiliary_loss_clip": 0.01166823,
      "auxiliary_loss_mlp": 0.01037625,
      "balance_loss_clip": 1.05531049,
      "balance_loss_mlp": 1.02743208,
      "epoch": 0.4991282390428666,
      "flos": 13291042187520.0,
      "grad_norm": 4.4514034816864605,
      "language_loss": 0.82594317,
      "learning_rate": 2.1027789657982255e-06,
      "loss": 0.84798765,
      "num_input_tokens_seen": 89447410,
      "step": 4151,
      "time_per_iteration": 2.619112968444824
    },
    {
      "auxiliary_loss_clip": 0.0117642,
      "auxiliary_loss_mlp": 0.01033832,
      "balance_loss_clip": 1.05643857,
      "balance_loss_mlp": 1.02559423,
      "epoch": 0.4992484819335057,
      "flos": 21537496454400.0,
      "grad_norm": 2.747051702396346,
      "language_loss": 0.77402902,
      "learning_rate": 2.1020010156397482e-06,
      "loss": 0.79613149,
      "num_input_tokens_seen": 89464630,
      "step": 4152,
      "time_per_iteration": 2.6934616565704346
    },
    {
      "auxiliary_loss_clip": 0.01191801,
      "auxiliary_loss_mlp": 0.01030413,
      "balance_loss_clip": 1.05940628,
      "balance_loss_mlp": 1.02207994,
      "epoch": 0.4993687248241448,
      "flos": 24860095390080.0,
      "grad_norm": 2.426056199211044,
      "language_loss": 0.77561474,
      "learning_rate": 2.101223050007797e-06,
      "loss": 0.7978369,
      "num_input_tokens_seen": 89483180,
      "step": 4153,
      "time_per_iteration": 2.6963367462158203
    },
    {
      "auxiliary_loss_clip": 0.01078907,
      "auxiliary_loss_mlp": 0.01002308,
      "balance_loss_clip": 1.01406753,
      "balance_loss_mlp": 1.00120568,
      "epoch": 0.49948896771478385,
      "flos": 62941602453120.0,
      "grad_norm": 0.8054858306278241,
      "language_loss": 0.53677243,
      "learning_rate": 2.1004450690203904e-06,
      "loss": 0.55758458,
      "num_input_tokens_seen": 89539260,
      "step": 4154,
      "time_per_iteration": 4.118409156799316
    },
    {
      "auxiliary_loss_clip": 0.01047079,
      "auxiliary_loss_mlp": 0.01003194,
      "balance_loss_clip": 1.01268065,
      "balance_loss_mlp": 1.00191271,
      "epoch": 0.49960921060542296,
      "flos": 68284213516800.0,
      "grad_norm": 0.9458735637943643,
      "language_loss": 0.63286114,
      "learning_rate": 2.099667072795546e-06,
      "loss": 0.65336382,
      "num_input_tokens_seen": 89601380,
      "step": 4155,
      "time_per_iteration": 3.331704616546631
    },
    {
      "auxiliary_loss_clip": 0.01159418,
      "auxiliary_loss_mlp": 0.01026427,
      "balance_loss_clip": 1.05446219,
      "balance_loss_mlp": 1.0182488,
      "epoch": 0.49972945349606207,
      "flos": 23659350618240.0,
      "grad_norm": 1.777093205913393,
      "language_loss": 0.80031437,
      "learning_rate": 2.0988890614512864e-06,
      "loss": 0.82217282,
      "num_input_tokens_seen": 89621270,
      "step": 4156,
      "time_per_iteration": 2.707486391067505
    },
    {
      "auxiliary_loss_clip": 0.01177723,
      "auxiliary_loss_mlp": 0.01030366,
      "balance_loss_clip": 1.05528808,
      "balance_loss_mlp": 1.02209234,
      "epoch": 0.4998496963867011,
      "flos": 19755825022080.0,
      "grad_norm": 2.1768058203520133,
      "language_loss": 0.84753704,
      "learning_rate": 2.098111035105635e-06,
      "loss": 0.86961794,
      "num_input_tokens_seen": 89639695,
      "step": 4157,
      "time_per_iteration": 2.6800320148468018
    },
    {
      "auxiliary_loss_clip": 0.01195386,
      "auxiliary_loss_mlp": 0.01029154,
      "balance_loss_clip": 1.06061912,
      "balance_loss_mlp": 1.0198555,
      "epoch": 0.49996993927734024,
      "flos": 22265728790400.0,
      "grad_norm": 1.77076153941669,
      "language_loss": 0.73382258,
      "learning_rate": 2.0973329938766176e-06,
      "loss": 0.75606799,
      "num_input_tokens_seen": 89657125,
      "step": 4158,
      "time_per_iteration": 2.674050807952881
    },
    {
      "auxiliary_loss_clip": 0.01182952,
      "auxiliary_loss_mlp": 0.01036111,
      "balance_loss_clip": 1.05980086,
      "balance_loss_mlp": 1.02735496,
      "epoch": 0.5000901821679793,
      "flos": 23327212533120.0,
      "grad_norm": 9.664132274474838,
      "language_loss": 0.79121935,
      "learning_rate": 2.0965549378822618e-06,
      "loss": 0.81340998,
      "num_input_tokens_seen": 89678415,
      "step": 4159,
      "time_per_iteration": 2.6879539489746094
    },
    {
      "auxiliary_loss_clip": 0.01192199,
      "auxiliary_loss_mlp": 0.01031902,
      "balance_loss_clip": 1.05820584,
      "balance_loss_mlp": 1.02344394,
      "epoch": 0.5002104250586185,
      "flos": 20339014239360.0,
      "grad_norm": 2.370748786799079,
      "language_loss": 0.84345651,
      "learning_rate": 2.095776867240599e-06,
      "loss": 0.8656975,
      "num_input_tokens_seen": 89695405,
      "step": 4160,
      "time_per_iteration": 2.6911094188690186
    },
    {
      "auxiliary_loss_clip": 0.01151323,
      "auxiliary_loss_mlp": 0.01038024,
      "balance_loss_clip": 1.05027151,
      "balance_loss_mlp": 1.02907181,
      "epoch": 0.5003306679492575,
      "flos": 13991372634240.0,
      "grad_norm": 2.149897018950823,
      "language_loss": 0.8248955,
      "learning_rate": 2.094998782069661e-06,
      "loss": 0.84678894,
      "num_input_tokens_seen": 89713110,
      "step": 4161,
      "time_per_iteration": 2.6434261798858643
    },
    {
      "auxiliary_loss_clip": 0.01162562,
      "auxiliary_loss_mlp": 0.0103643,
      "balance_loss_clip": 1.05711424,
      "balance_loss_mlp": 1.02754879,
      "epoch": 0.5004509108398966,
      "flos": 27672762896640.0,
      "grad_norm": 1.9136401866541985,
      "language_loss": 0.75716317,
      "learning_rate": 2.0942206824874845e-06,
      "loss": 0.77915311,
      "num_input_tokens_seen": 89735885,
      "step": 4162,
      "time_per_iteration": 2.757193088531494
    },
    {
      "auxiliary_loss_clip": 0.01189715,
      "auxiliary_loss_mlp": 0.01027495,
      "balance_loss_clip": 1.05571496,
      "balance_loss_mlp": 1.01838136,
      "epoch": 0.5005711537305357,
      "flos": 14976186796800.0,
      "grad_norm": 2.448390615101562,
      "language_loss": 0.7895788,
      "learning_rate": 2.093442568612105e-06,
      "loss": 0.81175083,
      "num_input_tokens_seen": 89753690,
      "step": 4163,
      "time_per_iteration": 2.592289924621582
    },
    {
      "auxiliary_loss_clip": 0.01165069,
      "auxiliary_loss_mlp": 0.01035229,
      "balance_loss_clip": 1.05376482,
      "balance_loss_mlp": 1.02616882,
      "epoch": 0.5006913966211748,
      "flos": 26503259978880.0,
      "grad_norm": 1.518251581199556,
      "language_loss": 0.85124481,
      "learning_rate": 2.0926644405615613e-06,
      "loss": 0.87324774,
      "num_input_tokens_seen": 89774590,
      "step": 4164,
      "time_per_iteration": 2.816166400909424
    },
    {
      "auxiliary_loss_clip": 0.01188637,
      "auxiliary_loss_mlp": 0.01032563,
      "balance_loss_clip": 1.0560441,
      "balance_loss_mlp": 1.02382529,
      "epoch": 0.5008116395118138,
      "flos": 20449295971200.0,
      "grad_norm": 2.12053230057217,
      "language_loss": 0.81559873,
      "learning_rate": 2.091886298453897e-06,
      "loss": 0.83781075,
      "num_input_tokens_seen": 89792775,
      "step": 4165,
      "time_per_iteration": 2.6170711517333984
    },
    {
      "auxiliary_loss_clip": 0.011802,
      "auxiliary_loss_mlp": 0.01029393,
      "balance_loss_clip": 1.05653358,
      "balance_loss_mlp": 1.0205183,
      "epoch": 0.500931882402453,
      "flos": 21579871524480.0,
      "grad_norm": 3.0196556078606935,
      "language_loss": 0.73271,
      "learning_rate": 2.091108142407153e-06,
      "loss": 0.75480592,
      "num_input_tokens_seen": 89811515,
      "step": 4166,
      "time_per_iteration": 2.7297942638397217
    },
    {
      "auxiliary_loss_clip": 0.01069019,
      "auxiliary_loss_mlp": 0.01007508,
      "balance_loss_clip": 1.01331103,
      "balance_loss_mlp": 1.00626266,
      "epoch": 0.5010521252930921,
      "flos": 57785011925760.0,
      "grad_norm": 0.8437140243739624,
      "language_loss": 0.62323463,
      "learning_rate": 2.090329972539377e-06,
      "loss": 0.64399993,
      "num_input_tokens_seen": 89870080,
      "step": 4167,
      "time_per_iteration": 3.262648105621338
    },
    {
      "auxiliary_loss_clip": 0.01173493,
      "auxiliary_loss_mlp": 0.01032561,
      "balance_loss_clip": 1.05547106,
      "balance_loss_mlp": 1.02379346,
      "epoch": 0.5011723681837311,
      "flos": 18625500864000.0,
      "grad_norm": 1.7323544955569063,
      "language_loss": 0.68584192,
      "learning_rate": 2.089551788968616e-06,
      "loss": 0.70790243,
      "num_input_tokens_seen": 89888045,
      "step": 4168,
      "time_per_iteration": 2.6090943813323975
    },
    {
      "auxiliary_loss_clip": 0.0107703,
      "auxiliary_loss_mlp": 0.01005261,
      "balance_loss_clip": 1.01289463,
      "balance_loss_mlp": 1.00416422,
      "epoch": 0.5012926110743702,
      "flos": 55883146608000.0,
      "grad_norm": 0.8605758427954106,
      "language_loss": 0.60647583,
      "learning_rate": 2.08877359181292e-06,
      "loss": 0.62729871,
      "num_input_tokens_seen": 89944610,
      "step": 4169,
      "time_per_iteration": 3.203305244445801
    },
    {
      "auxiliary_loss_clip": 0.01180535,
      "auxiliary_loss_mlp": 0.01027938,
      "balance_loss_clip": 1.05593753,
      "balance_loss_mlp": 1.01829398,
      "epoch": 0.5014128539650093,
      "flos": 24238266117120.0,
      "grad_norm": 3.0192916721305556,
      "language_loss": 0.85909909,
      "learning_rate": 2.0879953811903396e-06,
      "loss": 0.88118386,
      "num_input_tokens_seen": 89959495,
      "step": 4170,
      "time_per_iteration": 2.6907780170440674
    },
    {
      "auxiliary_loss_clip": 0.01193522,
      "auxiliary_loss_mlp": 0.01033313,
      "balance_loss_clip": 1.05902147,
      "balance_loss_mlp": 1.024593,
      "epoch": 0.5015330968556484,
      "flos": 27527468382720.0,
      "grad_norm": 1.9829530876052455,
      "language_loss": 0.79140604,
      "learning_rate": 2.08721715721893e-06,
      "loss": 0.81367433,
      "num_input_tokens_seen": 89978820,
      "step": 4171,
      "time_per_iteration": 3.667344570159912
    },
    {
      "auxiliary_loss_clip": 0.01173484,
      "auxiliary_loss_mlp": 0.01030924,
      "balance_loss_clip": 1.05514073,
      "balance_loss_mlp": 1.02188766,
      "epoch": 0.5016533397462875,
      "flos": 23800802376960.0,
      "grad_norm": 2.249668634024799,
      "language_loss": 0.77357352,
      "learning_rate": 2.0864389200167477e-06,
      "loss": 0.79561758,
      "num_input_tokens_seen": 89997075,
      "step": 4172,
      "time_per_iteration": 2.715132474899292
    },
    {
      "auxiliary_loss_clip": 0.01176091,
      "auxiliary_loss_mlp": 0.01032119,
      "balance_loss_clip": 1.05958176,
      "balance_loss_mlp": 1.0226841,
      "epoch": 0.5017735826369266,
      "flos": 25295009264640.0,
      "grad_norm": 2.2587318654136177,
      "language_loss": 0.79005229,
      "learning_rate": 2.0856606697018504e-06,
      "loss": 0.81213444,
      "num_input_tokens_seen": 90015085,
      "step": 4173,
      "time_per_iteration": 4.617856025695801
    },
    {
      "auxiliary_loss_clip": 0.01162699,
      "auxiliary_loss_mlp": 0.0103297,
      "balance_loss_clip": 1.05452847,
      "balance_loss_mlp": 1.02337956,
      "epoch": 0.5018938255275657,
      "flos": 16873203778560.0,
      "grad_norm": 2.002831701215863,
      "language_loss": 0.73198873,
      "learning_rate": 2.084882406392297e-06,
      "loss": 0.75394541,
      "num_input_tokens_seen": 90033045,
      "step": 4174,
      "time_per_iteration": 2.67281436920166
    },
    {
      "auxiliary_loss_clip": 0.01179435,
      "auxiliary_loss_mlp": 0.01031212,
      "balance_loss_clip": 1.0570116,
      "balance_loss_mlp": 1.02280164,
      "epoch": 0.5020140684182047,
      "flos": 25515429073920.0,
      "grad_norm": 2.801128507485884,
      "language_loss": 0.71115923,
      "learning_rate": 2.0841041302061496e-06,
      "loss": 0.73326564,
      "num_input_tokens_seen": 90052505,
      "step": 4175,
      "time_per_iteration": 2.8152542114257812
    },
    {
      "auxiliary_loss_clip": 0.01154347,
      "auxiliary_loss_mlp": 0.0103789,
      "balance_loss_clip": 1.05409765,
      "balance_loss_mlp": 1.02884197,
      "epoch": 0.5021343113088439,
      "flos": 23659278791040.0,
      "grad_norm": 19.698542542808653,
      "language_loss": 0.75937986,
      "learning_rate": 2.083325841261473e-06,
      "loss": 0.78130221,
      "num_input_tokens_seen": 90071565,
      "step": 4176,
      "time_per_iteration": 2.732106924057007
    },
    {
      "auxiliary_loss_clip": 0.01175554,
      "auxiliary_loss_mlp": 0.01024953,
      "balance_loss_clip": 1.0560168,
      "balance_loss_mlp": 1.01619148,
      "epoch": 0.502254554199483,
      "flos": 24534673148160.0,
      "grad_norm": 2.625650068429101,
      "language_loss": 0.6670543,
      "learning_rate": 2.0825475396763322e-06,
      "loss": 0.68905938,
      "num_input_tokens_seen": 90092215,
      "step": 4177,
      "time_per_iteration": 2.7200145721435547
    },
    {
      "auxiliary_loss_clip": 0.01192794,
      "auxiliary_loss_mlp": 0.01035609,
      "balance_loss_clip": 1.05971682,
      "balance_loss_mlp": 1.02706194,
      "epoch": 0.502374797090122,
      "flos": 34240285607040.0,
      "grad_norm": 1.5886428333481464,
      "language_loss": 0.65761942,
      "learning_rate": 2.081769225568796e-06,
      "loss": 0.67990345,
      "num_input_tokens_seen": 90114665,
      "step": 4178,
      "time_per_iteration": 2.742154359817505
    },
    {
      "auxiliary_loss_clip": 0.01180151,
      "auxiliary_loss_mlp": 0.01034068,
      "balance_loss_clip": 1.05588722,
      "balance_loss_mlp": 1.02522266,
      "epoch": 0.5024950399807612,
      "flos": 26031106679040.0,
      "grad_norm": 1.9253571260516884,
      "language_loss": 0.76295543,
      "learning_rate": 2.0809908990569327e-06,
      "loss": 0.7850976,
      "num_input_tokens_seen": 90136445,
      "step": 4179,
      "time_per_iteration": 2.7795915603637695
    },
    {
      "auxiliary_loss_clip": 0.01192584,
      "auxiliary_loss_mlp": 0.00763545,
      "balance_loss_clip": 1.05908871,
      "balance_loss_mlp": 1.00115013,
      "epoch": 0.5026152828714002,
      "flos": 21252438120960.0,
      "grad_norm": 1.836832121855767,
      "language_loss": 0.7928201,
      "learning_rate": 2.0802125602588146e-06,
      "loss": 0.81238133,
      "num_input_tokens_seen": 90155710,
      "step": 4180,
      "time_per_iteration": 3.6344120502471924
    },
    {
      "auxiliary_loss_clip": 0.01180243,
      "auxiliary_loss_mlp": 0.01032393,
      "balance_loss_clip": 1.05584478,
      "balance_loss_mlp": 1.02297544,
      "epoch": 0.5027355257620393,
      "flos": 30956111245440.0,
      "grad_norm": 3.5167497697468457,
      "language_loss": 0.66726875,
      "learning_rate": 2.0794342092925146e-06,
      "loss": 0.68939519,
      "num_input_tokens_seen": 90176845,
      "step": 4181,
      "time_per_iteration": 2.746587038040161
    },
    {
      "auxiliary_loss_clip": 0.0116953,
      "auxiliary_loss_mlp": 0.01032934,
      "balance_loss_clip": 1.05662203,
      "balance_loss_mlp": 1.02398181,
      "epoch": 0.5028557686526784,
      "flos": 24791147233920.0,
      "grad_norm": 2.1454018541461863,
      "language_loss": 0.68024004,
      "learning_rate": 2.078655846276108e-06,
      "loss": 0.70226473,
      "num_input_tokens_seen": 90197175,
      "step": 4182,
      "time_per_iteration": 2.7469465732574463
    },
    {
      "auxiliary_loss_clip": 0.01155066,
      "auxiliary_loss_mlp": 0.01030615,
      "balance_loss_clip": 1.05056977,
      "balance_loss_mlp": 1.02222264,
      "epoch": 0.5029760115433175,
      "flos": 22966992990720.0,
      "grad_norm": 4.017587000424839,
      "language_loss": 0.69237626,
      "learning_rate": 2.0778774713276727e-06,
      "loss": 0.7142331,
      "num_input_tokens_seen": 90216650,
      "step": 4183,
      "time_per_iteration": 2.7006218433380127
    },
    {
      "auxiliary_loss_clip": 0.01178076,
      "auxiliary_loss_mlp": 0.01032834,
      "balance_loss_clip": 1.05458379,
      "balance_loss_mlp": 1.02301097,
      "epoch": 0.5030962544339566,
      "flos": 15305164485120.0,
      "grad_norm": 2.684940486478004,
      "language_loss": 0.68013066,
      "learning_rate": 2.077099084565287e-06,
      "loss": 0.70223981,
      "num_input_tokens_seen": 90234055,
      "step": 4184,
      "time_per_iteration": 2.678521156311035
    },
    {
      "auxiliary_loss_clip": 0.0117277,
      "auxiliary_loss_mlp": 0.01029232,
      "balance_loss_clip": 1.05330336,
      "balance_loss_mlp": 1.0199399,
      "epoch": 0.5032164973245957,
      "flos": 24494847943680.0,
      "grad_norm": 2.3222395933617554,
      "language_loss": 0.65858281,
      "learning_rate": 2.0763206861070313e-06,
      "loss": 0.68060285,
      "num_input_tokens_seen": 90253115,
      "step": 4185,
      "time_per_iteration": 2.6790173053741455
    },
    {
      "auxiliary_loss_clip": 0.01182376,
      "auxiliary_loss_mlp": 0.01040304,
      "balance_loss_clip": 1.05794048,
      "balance_loss_mlp": 1.03122592,
      "epoch": 0.5033367402152348,
      "flos": 16213452721920.0,
      "grad_norm": 1.9773850052947044,
      "language_loss": 0.75280887,
      "learning_rate": 2.0755422760709876e-06,
      "loss": 0.77503568,
      "num_input_tokens_seen": 90270515,
      "step": 4186,
      "time_per_iteration": 2.6910674571990967
    },
    {
      "auxiliary_loss_clip": 0.01192241,
      "auxiliary_loss_mlp": 0.0103162,
      "balance_loss_clip": 1.05844223,
      "balance_loss_mlp": 1.02262568,
      "epoch": 0.5034569831058738,
      "flos": 21391375927680.0,
      "grad_norm": 2.145072912467405,
      "language_loss": 0.7691102,
      "learning_rate": 2.0747638545752417e-06,
      "loss": 0.79134881,
      "num_input_tokens_seen": 90289075,
      "step": 4187,
      "time_per_iteration": 2.6384615898132324
    },
    {
      "auxiliary_loss_clip": 0.01151446,
      "auxiliary_loss_mlp": 0.01037099,
      "balance_loss_clip": 1.04942298,
      "balance_loss_mlp": 1.02830112,
      "epoch": 0.503577225996513,
      "flos": 20558751690240.0,
      "grad_norm": 2.698316101540089,
      "language_loss": 0.83585447,
      "learning_rate": 2.073985421737878e-06,
      "loss": 0.85773993,
      "num_input_tokens_seen": 90306385,
      "step": 4188,
      "time_per_iteration": 2.709934949874878
    },
    {
      "auxiliary_loss_clip": 0.01194792,
      "auxiliary_loss_mlp": 0.01027892,
      "balance_loss_clip": 1.059219,
      "balance_loss_mlp": 1.01913047,
      "epoch": 0.5036974688871521,
      "flos": 27229157930880.0,
      "grad_norm": 2.916599056211085,
      "language_loss": 0.74128062,
      "learning_rate": 2.0732069776769844e-06,
      "loss": 0.76350749,
      "num_input_tokens_seen": 90323795,
      "step": 4189,
      "time_per_iteration": 2.6287145614624023
    },
    {
      "auxiliary_loss_clip": 0.01193943,
      "auxiliary_loss_mlp": 0.01031657,
      "balance_loss_clip": 1.05951333,
      "balance_loss_mlp": 1.02243662,
      "epoch": 0.5038177117777911,
      "flos": 20412164286720.0,
      "grad_norm": 8.789171358795306,
      "language_loss": 0.73377472,
      "learning_rate": 2.072428522510651e-06,
      "loss": 0.7560308,
      "num_input_tokens_seen": 90340360,
      "step": 4190,
      "time_per_iteration": 2.695374011993408
    },
    {
      "auxiliary_loss_clip": 0.01158378,
      "auxiliary_loss_mlp": 0.01032117,
      "balance_loss_clip": 1.05373192,
      "balance_loss_mlp": 1.0237546,
      "epoch": 0.5039379546684303,
      "flos": 21907987286400.0,
      "grad_norm": 2.793435140665997,
      "language_loss": 0.76144278,
      "learning_rate": 2.071650056356968e-06,
      "loss": 0.78334779,
      "num_input_tokens_seen": 90357900,
      "step": 4191,
      "time_per_iteration": 2.6788179874420166
    },
    {
      "auxiliary_loss_clip": 0.01191163,
      "auxiliary_loss_mlp": 0.01030395,
      "balance_loss_clip": 1.05712199,
      "balance_loss_mlp": 1.02160978,
      "epoch": 0.5040581975590693,
      "flos": 20010718909440.0,
      "grad_norm": 12.584066191393518,
      "language_loss": 0.79929149,
      "learning_rate": 2.070871579334028e-06,
      "loss": 0.82150704,
      "num_input_tokens_seen": 90377010,
      "step": 4192,
      "time_per_iteration": 2.7714319229125977
    },
    {
      "auxiliary_loss_clip": 0.0119044,
      "auxiliary_loss_mlp": 0.01033237,
      "balance_loss_clip": 1.05631125,
      "balance_loss_mlp": 1.02423096,
      "epoch": 0.5041784404497084,
      "flos": 20959837931520.0,
      "grad_norm": 2.0411981080997763,
      "language_loss": 0.71773142,
      "learning_rate": 2.0700930915599264e-06,
      "loss": 0.73996818,
      "num_input_tokens_seen": 90396740,
      "step": 4193,
      "time_per_iteration": 2.6735219955444336
    },
    {
      "auxiliary_loss_clip": 0.01163972,
      "auxiliary_loss_mlp": 0.01026306,
      "balance_loss_clip": 1.05512702,
      "balance_loss_mlp": 1.01837826,
      "epoch": 0.5042986833403476,
      "flos": 12495082757760.0,
      "grad_norm": 1.9649344202931003,
      "language_loss": 0.78459907,
      "learning_rate": 2.0693145931527583e-06,
      "loss": 0.80650187,
      "num_input_tokens_seen": 90413220,
      "step": 4194,
      "time_per_iteration": 2.705106258392334
    },
    {
      "auxiliary_loss_clip": 0.01165832,
      "auxiliary_loss_mlp": 0.01027374,
      "balance_loss_clip": 1.05302465,
      "balance_loss_mlp": 1.01867151,
      "epoch": 0.5044189262309866,
      "flos": 29202305788800.0,
      "grad_norm": 1.719465053033623,
      "language_loss": 0.78256118,
      "learning_rate": 2.068536084230622e-06,
      "loss": 0.80449319,
      "num_input_tokens_seen": 90435085,
      "step": 4195,
      "time_per_iteration": 2.8046512603759766
    },
    {
      "auxiliary_loss_clip": 0.01160938,
      "auxiliary_loss_mlp": 0.01035532,
      "balance_loss_clip": 1.05288339,
      "balance_loss_mlp": 1.02570319,
      "epoch": 0.5045391691216257,
      "flos": 23873198238720.0,
      "grad_norm": 2.5498823505420685,
      "language_loss": 0.89220381,
      "learning_rate": 2.067757564911616e-06,
      "loss": 0.91416854,
      "num_input_tokens_seen": 90453660,
      "step": 4196,
      "time_per_iteration": 2.7548305988311768
    },
    {
      "auxiliary_loss_clip": 0.01176463,
      "auxiliary_loss_mlp": 0.01028774,
      "balance_loss_clip": 1.0572753,
      "balance_loss_mlp": 1.0197556,
      "epoch": 0.5046594120122648,
      "flos": 24644990793600.0,
      "grad_norm": 3.4990523223378367,
      "language_loss": 0.92714542,
      "learning_rate": 2.0669790353138407e-06,
      "loss": 0.94919777,
      "num_input_tokens_seen": 90472625,
      "step": 4197,
      "time_per_iteration": 3.541628837585449
    },
    {
      "auxiliary_loss_clip": 0.01194372,
      "auxiliary_loss_mlp": 0.00763115,
      "balance_loss_clip": 1.0587306,
      "balance_loss_mlp": 1.00125527,
      "epoch": 0.5047796549029039,
      "flos": 23362835846400.0,
      "grad_norm": 2.2393320477038094,
      "language_loss": 0.73679078,
      "learning_rate": 2.0662004955553995e-06,
      "loss": 0.75636566,
      "num_input_tokens_seen": 90492325,
      "step": 4198,
      "time_per_iteration": 2.634474754333496
    },
    {
      "auxiliary_loss_clip": 0.01186797,
      "auxiliary_loss_mlp": 0.01032748,
      "balance_loss_clip": 1.05583525,
      "balance_loss_mlp": 1.02362871,
      "epoch": 0.5048998977935429,
      "flos": 17304095329920.0,
      "grad_norm": 4.728315680528848,
      "language_loss": 0.76887524,
      "learning_rate": 2.065421945754395e-06,
      "loss": 0.7910707,
      "num_input_tokens_seen": 90510055,
      "step": 4199,
      "time_per_iteration": 4.482813358306885
    },
    {
      "auxiliary_loss_clip": 0.01191912,
      "auxiliary_loss_mlp": 0.01029171,
      "balance_loss_clip": 1.05936897,
      "balance_loss_mlp": 1.01990879,
      "epoch": 0.505020140684182,
      "flos": 34856979235200.0,
      "grad_norm": 1.6735108968917296,
      "language_loss": 0.78105831,
      "learning_rate": 2.0646433860289344e-06,
      "loss": 0.80326915,
      "num_input_tokens_seen": 90528980,
      "step": 4200,
      "time_per_iteration": 2.7609121799468994
    },
    {
      "auxiliary_loss_clip": 0.01176568,
      "auxiliary_loss_mlp": 0.01033863,
      "balance_loss_clip": 1.05489254,
      "balance_loss_mlp": 1.0242008,
      "epoch": 0.5051403835748212,
      "flos": 24863974058880.0,
      "grad_norm": 2.1759238420150404,
      "language_loss": 0.82741904,
      "learning_rate": 2.0638648164971233e-06,
      "loss": 0.84952337,
      "num_input_tokens_seen": 90547445,
      "step": 4201,
      "time_per_iteration": 2.6392595767974854
    },
    {
      "auxiliary_loss_clip": 0.01187903,
      "auxiliary_loss_mlp": 0.01031878,
      "balance_loss_clip": 1.05593657,
      "balance_loss_mlp": 1.02336609,
      "epoch": 0.5052606264654602,
      "flos": 20959694277120.0,
      "grad_norm": 2.0080660930930385,
      "language_loss": 0.89019614,
      "learning_rate": 2.06308623727707e-06,
      "loss": 0.91239399,
      "num_input_tokens_seen": 90567545,
      "step": 4202,
      "time_per_iteration": 2.657195568084717
    },
    {
      "auxiliary_loss_clip": 0.01178139,
      "auxiliary_loss_mlp": 0.01031253,
      "balance_loss_clip": 1.05718911,
      "balance_loss_mlp": 1.02209806,
      "epoch": 0.5053808693560993,
      "flos": 19642382893440.0,
      "grad_norm": 4.377143116805959,
      "language_loss": 0.76857698,
      "learning_rate": 2.0623076484868846e-06,
      "loss": 0.79067087,
      "num_input_tokens_seen": 90585000,
      "step": 4203,
      "time_per_iteration": 2.637214422225952
    },
    {
      "auxiliary_loss_clip": 0.01062244,
      "auxiliary_loss_mlp": 0.01000178,
      "balance_loss_clip": 1.0112133,
      "balance_loss_mlp": 0.99909908,
      "epoch": 0.5055011122467384,
      "flos": 67504915019520.0,
      "grad_norm": 0.8325025286479011,
      "language_loss": 0.60632461,
      "learning_rate": 2.061529050244679e-06,
      "loss": 0.62694883,
      "num_input_tokens_seen": 90644745,
      "step": 4204,
      "time_per_iteration": 3.200782299041748
    },
    {
      "auxiliary_loss_clip": 0.01193143,
      "auxiliary_loss_mlp": 0.01034489,
      "balance_loss_clip": 1.0589112,
      "balance_loss_mlp": 1.02592945,
      "epoch": 0.5056213551373775,
      "flos": 16872952383360.0,
      "grad_norm": 2.987664487454686,
      "language_loss": 0.74367881,
      "learning_rate": 2.060750442668565e-06,
      "loss": 0.76595515,
      "num_input_tokens_seen": 90662500,
      "step": 4205,
      "time_per_iteration": 2.6281721591949463
    },
    {
      "auxiliary_loss_clip": 0.0117655,
      "auxiliary_loss_mlp": 0.01028877,
      "balance_loss_clip": 1.05921054,
      "balance_loss_mlp": 1.02030027,
      "epoch": 0.5057415980280165,
      "flos": 15334179696000.0,
      "grad_norm": 2.2140188610862186,
      "language_loss": 0.64115918,
      "learning_rate": 2.059971825876657e-06,
      "loss": 0.66321349,
      "num_input_tokens_seen": 90677010,
      "step": 4206,
      "time_per_iteration": 3.579533815383911
    },
    {
      "auxiliary_loss_clip": 0.01189192,
      "auxiliary_loss_mlp": 0.01030503,
      "balance_loss_clip": 1.05627203,
      "balance_loss_mlp": 1.02165771,
      "epoch": 0.5058618409186557,
      "flos": 19025976574080.0,
      "grad_norm": 1.9176040410896698,
      "language_loss": 0.76949108,
      "learning_rate": 2.0591931999870713e-06,
      "loss": 0.79168802,
      "num_input_tokens_seen": 90695935,
      "step": 4207,
      "time_per_iteration": 2.712742805480957
    },
    {
      "auxiliary_loss_clip": 0.0107496,
      "auxiliary_loss_mlp": 0.01002994,
      "balance_loss_clip": 1.01084781,
      "balance_loss_mlp": 1.00196922,
      "epoch": 0.5059820838092948,
      "flos": 63453114080640.0,
      "grad_norm": 0.8423071216981858,
      "language_loss": 0.5754503,
      "learning_rate": 2.0584145651179234e-06,
      "loss": 0.59622979,
      "num_input_tokens_seen": 90751645,
      "step": 4208,
      "time_per_iteration": 3.2069966793060303
    },
    {
      "auxiliary_loss_clip": 0.01177212,
      "auxiliary_loss_mlp": 0.01033315,
      "balance_loss_clip": 1.05746686,
      "balance_loss_mlp": 1.02367663,
      "epoch": 0.5061023266999338,
      "flos": 15441803821440.0,
      "grad_norm": 3.998577141408444,
      "language_loss": 0.79774296,
      "learning_rate": 2.0576359213873327e-06,
      "loss": 0.8198483,
      "num_input_tokens_seen": 90766795,
      "step": 4209,
      "time_per_iteration": 2.6175994873046875
    },
    {
      "auxiliary_loss_clip": 0.0118415,
      "auxiliary_loss_mlp": 0.0103012,
      "balance_loss_clip": 1.05508947,
      "balance_loss_mlp": 1.02072072,
      "epoch": 0.506222569590573,
      "flos": 22451063990400.0,
      "grad_norm": 2.709874186403115,
      "language_loss": 0.70660937,
      "learning_rate": 2.056857268913419e-06,
      "loss": 0.72875202,
      "num_input_tokens_seen": 90786845,
      "step": 4210,
      "time_per_iteration": 2.704380750656128
    },
    {
      "auxiliary_loss_clip": 0.01192936,
      "auxiliary_loss_mlp": 0.0103352,
      "balance_loss_clip": 1.0582155,
      "balance_loss_mlp": 1.02464449,
      "epoch": 0.506342812481212,
      "flos": 17558665994880.0,
      "grad_norm": 15.866484273682294,
      "language_loss": 0.83995926,
      "learning_rate": 2.056078607814303e-06,
      "loss": 0.86222386,
      "num_input_tokens_seen": 90802630,
      "step": 4211,
      "time_per_iteration": 2.570909261703491
    },
    {
      "auxiliary_loss_clip": 0.01180004,
      "auxiliary_loss_mlp": 0.0103229,
      "balance_loss_clip": 1.05549085,
      "balance_loss_mlp": 1.02250326,
      "epoch": 0.5064630553718511,
      "flos": 23402050519680.0,
      "grad_norm": 2.363625736755643,
      "language_loss": 0.78522265,
      "learning_rate": 2.055299938208106e-06,
      "loss": 0.80734563,
      "num_input_tokens_seen": 90823620,
      "step": 4212,
      "time_per_iteration": 2.7083740234375
    },
    {
      "auxiliary_loss_clip": 0.01196817,
      "auxiliary_loss_mlp": 0.0076326,
      "balance_loss_clip": 1.060009,
      "balance_loss_mlp": 1.00127053,
      "epoch": 0.5065832982624903,
      "flos": 23987035416960.0,
      "grad_norm": 2.061488610563865,
      "language_loss": 0.8636654,
      "learning_rate": 2.0545212602129526e-06,
      "loss": 0.88326621,
      "num_input_tokens_seen": 90843475,
      "step": 4213,
      "time_per_iteration": 2.5893867015838623
    },
    {
      "auxiliary_loss_clip": 0.01176575,
      "auxiliary_loss_mlp": 0.01034747,
      "balance_loss_clip": 1.05797374,
      "balance_loss_mlp": 1.02544284,
      "epoch": 0.5067035411531293,
      "flos": 21503058289920.0,
      "grad_norm": 7.19032286822976,
      "language_loss": 0.66634512,
      "learning_rate": 2.0537425739469673e-06,
      "loss": 0.68845832,
      "num_input_tokens_seen": 90862410,
      "step": 4214,
      "time_per_iteration": 2.694554567337036
    },
    {
      "auxiliary_loss_clip": 0.01048289,
      "auxiliary_loss_mlp": 0.01003026,
      "balance_loss_clip": 1.01196539,
      "balance_loss_mlp": 1.00199509,
      "epoch": 0.5068237840437684,
      "flos": 65934397687680.0,
      "grad_norm": 0.8406527394036508,
      "language_loss": 0.59404635,
      "learning_rate": 2.052963879528276e-06,
      "loss": 0.61455953,
      "num_input_tokens_seen": 90922280,
      "step": 4215,
      "time_per_iteration": 3.224320650100708
    },
    {
      "auxiliary_loss_clip": 0.01160496,
      "auxiliary_loss_mlp": 0.01031263,
      "balance_loss_clip": 1.05729985,
      "balance_loss_mlp": 1.02290678,
      "epoch": 0.5069440269344075,
      "flos": 27264206626560.0,
      "grad_norm": 3.1777665067023104,
      "language_loss": 0.76886469,
      "learning_rate": 2.052185177075007e-06,
      "loss": 0.79078233,
      "num_input_tokens_seen": 90941850,
      "step": 4216,
      "time_per_iteration": 2.734719753265381
    },
    {
      "auxiliary_loss_clip": 0.01173014,
      "auxiliary_loss_mlp": 0.0103625,
      "balance_loss_clip": 1.05501115,
      "balance_loss_mlp": 1.02662945,
      "epoch": 0.5070642698250466,
      "flos": 23366319465600.0,
      "grad_norm": 1.780381806187783,
      "language_loss": 0.83147109,
      "learning_rate": 2.051406466705288e-06,
      "loss": 0.85356373,
      "num_input_tokens_seen": 90961390,
      "step": 4217,
      "time_per_iteration": 2.645664930343628
    },
    {
      "auxiliary_loss_clip": 0.01175873,
      "auxiliary_loss_mlp": 0.01023758,
      "balance_loss_clip": 1.05269575,
      "balance_loss_mlp": 1.01493669,
      "epoch": 0.5071845127156857,
      "flos": 20340127560960.0,
      "grad_norm": 2.121966402304375,
      "language_loss": 0.81397307,
      "learning_rate": 2.0506277485372486e-06,
      "loss": 0.83596945,
      "num_input_tokens_seen": 90980215,
      "step": 4218,
      "time_per_iteration": 2.6543381214141846
    },
    {
      "auxiliary_loss_clip": 0.01165306,
      "auxiliary_loss_mlp": 0.0102883,
      "balance_loss_clip": 1.05759549,
      "balance_loss_mlp": 1.01971602,
      "epoch": 0.5073047556063248,
      "flos": 12092955022080.0,
      "grad_norm": 3.8295824090440864,
      "language_loss": 0.66739631,
      "learning_rate": 2.04984902268902e-06,
      "loss": 0.68933761,
      "num_input_tokens_seen": 90997415,
      "step": 4219,
      "time_per_iteration": 2.612363815307617
    },
    {
      "auxiliary_loss_clip": 0.01195764,
      "auxiliary_loss_mlp": 0.01028362,
      "balance_loss_clip": 1.05721784,
      "balance_loss_mlp": 1.01909924,
      "epoch": 0.5074249984969639,
      "flos": 19682854542720.0,
      "grad_norm": 2.8862976017922453,
      "language_loss": 0.75653648,
      "learning_rate": 2.0490702892787345e-06,
      "loss": 0.77877778,
      "num_input_tokens_seen": 91016475,
      "step": 4220,
      "time_per_iteration": 2.6624960899353027
    },
    {
      "auxiliary_loss_clip": 0.01171214,
      "auxiliary_loss_mlp": 0.01027147,
      "balance_loss_clip": 1.05383325,
      "balance_loss_mlp": 1.01897478,
      "epoch": 0.5075452413876029,
      "flos": 28765703975040.0,
      "grad_norm": 3.234016875557291,
      "language_loss": 0.62736309,
      "learning_rate": 2.0482915484245246e-06,
      "loss": 0.64934671,
      "num_input_tokens_seen": 91038095,
      "step": 4221,
      "time_per_iteration": 2.7397823333740234
    },
    {
      "auxiliary_loss_clip": 0.01169673,
      "auxiliary_loss_mlp": 0.01032907,
      "balance_loss_clip": 1.05841088,
      "balance_loss_mlp": 1.02391815,
      "epoch": 0.5076654842782421,
      "flos": 20339445202560.0,
      "grad_norm": 3.2590453355420093,
      "language_loss": 0.84317267,
      "learning_rate": 2.047512800244526e-06,
      "loss": 0.86519849,
      "num_input_tokens_seen": 91053360,
      "step": 4222,
      "time_per_iteration": 2.6575117111206055
    },
    {
      "auxiliary_loss_clip": 0.01195092,
      "auxiliary_loss_mlp": 0.01035884,
      "balance_loss_clip": 1.05958247,
      "balance_loss_mlp": 1.02654958,
      "epoch": 0.5077857271688812,
      "flos": 26359653404160.0,
      "grad_norm": 2.033476802276585,
      "language_loss": 0.79034007,
      "learning_rate": 2.046734044856873e-06,
      "loss": 0.81264979,
      "num_input_tokens_seen": 91072770,
      "step": 4223,
      "time_per_iteration": 3.5927839279174805
    },
    {
      "auxiliary_loss_clip": 0.01174971,
      "auxiliary_loss_mlp": 0.01031916,
      "balance_loss_clip": 1.0551089,
      "balance_loss_mlp": 1.02331507,
      "epoch": 0.5079059700595202,
      "flos": 21798962530560.0,
      "grad_norm": 1.991362642616682,
      "language_loss": 0.8135432,
      "learning_rate": 2.045955282379702e-06,
      "loss": 0.83561206,
      "num_input_tokens_seen": 91091430,
      "step": 4224,
      "time_per_iteration": 2.6835100650787354
    },
    {
      "auxiliary_loss_clip": 0.01195724,
      "auxiliary_loss_mlp": 0.01033054,
      "balance_loss_clip": 1.05879319,
      "balance_loss_mlp": 1.02343988,
      "epoch": 0.5080262129501594,
      "flos": 13187943175680.0,
      "grad_norm": 3.46660838557409,
      "language_loss": 0.76125383,
      "learning_rate": 2.045176512931152e-06,
      "loss": 0.78354156,
      "num_input_tokens_seen": 91106060,
      "step": 4225,
      "time_per_iteration": 4.492693901062012
    },
    {
      "auxiliary_loss_clip": 0.01188886,
      "auxiliary_loss_mlp": 0.01035823,
      "balance_loss_clip": 1.05700541,
      "balance_loss_mlp": 1.02705479,
      "epoch": 0.5081464558407984,
      "flos": 25301473712640.0,
      "grad_norm": 1.8661741382146182,
      "language_loss": 0.76650494,
      "learning_rate": 2.0443977366293604e-06,
      "loss": 0.78875202,
      "num_input_tokens_seen": 91124100,
      "step": 4226,
      "time_per_iteration": 2.716216564178467
    },
    {
      "auxiliary_loss_clip": 0.01182015,
      "auxiliary_loss_mlp": 0.01038487,
      "balance_loss_clip": 1.05573511,
      "balance_loss_mlp": 1.02882528,
      "epoch": 0.5082666987314375,
      "flos": 30951226995840.0,
      "grad_norm": 1.5998489364071706,
      "language_loss": 0.77119571,
      "learning_rate": 2.043618953592468e-06,
      "loss": 0.7934007,
      "num_input_tokens_seen": 91146555,
      "step": 4227,
      "time_per_iteration": 2.6752727031707764
    },
    {
      "auxiliary_loss_clip": 0.01179972,
      "auxiliary_loss_mlp": 0.01036446,
      "balance_loss_clip": 1.05712104,
      "balance_loss_mlp": 1.02688539,
      "epoch": 0.5083869416220766,
      "flos": 19682495406720.0,
      "grad_norm": 1.8221367696791895,
      "language_loss": 0.81388223,
      "learning_rate": 2.0428401639386144e-06,
      "loss": 0.83604646,
      "num_input_tokens_seen": 91167120,
      "step": 4228,
      "time_per_iteration": 2.7230608463287354
    },
    {
      "auxiliary_loss_clip": 0.0107641,
      "auxiliary_loss_mlp": 0.01000316,
      "balance_loss_clip": 1.01209199,
      "balance_loss_mlp": 0.99930251,
      "epoch": 0.5085071845127157,
      "flos": 71817535589760.0,
      "grad_norm": 0.8182235356847336,
      "language_loss": 0.58015418,
      "learning_rate": 2.042061367785943e-06,
      "loss": 0.60092145,
      "num_input_tokens_seen": 91220260,
      "step": 4229,
      "time_per_iteration": 3.1439719200134277
    },
    {
      "auxiliary_loss_clip": 0.01191538,
      "auxiliary_loss_mlp": 0.01027314,
      "balance_loss_clip": 1.05634952,
      "balance_loss_mlp": 1.0180639,
      "epoch": 0.5086274274033548,
      "flos": 35951608252800.0,
      "grad_norm": 2.4066169953058267,
      "language_loss": 0.74994135,
      "learning_rate": 2.041282565252594e-06,
      "loss": 0.77212977,
      "num_input_tokens_seen": 91240425,
      "step": 4230,
      "time_per_iteration": 2.7044143676757812
    },
    {
      "auxiliary_loss_clip": 0.01193577,
      "auxiliary_loss_mlp": 0.01025136,
      "balance_loss_clip": 1.0586257,
      "balance_loss_mlp": 1.01649952,
      "epoch": 0.5087476702939938,
      "flos": 23513732881920.0,
      "grad_norm": 1.868765376688365,
      "language_loss": 0.77341139,
      "learning_rate": 2.040503756456714e-06,
      "loss": 0.79559857,
      "num_input_tokens_seen": 91259635,
      "step": 4231,
      "time_per_iteration": 2.675261974334717
    },
    {
      "auxiliary_loss_clip": 0.01162078,
      "auxiliary_loss_mlp": 0.01036565,
      "balance_loss_clip": 1.04988468,
      "balance_loss_mlp": 1.02754664,
      "epoch": 0.508867913184633,
      "flos": 15122091841920.0,
      "grad_norm": 2.4575229100911833,
      "language_loss": 0.7875843,
      "learning_rate": 2.0397249415164456e-06,
      "loss": 0.80957073,
      "num_input_tokens_seen": 91276990,
      "step": 4232,
      "time_per_iteration": 3.5251805782318115
    },
    {
      "auxiliary_loss_clip": 0.01179708,
      "auxiliary_loss_mlp": 0.01033033,
      "balance_loss_clip": 1.05708838,
      "balance_loss_mlp": 1.02344227,
      "epoch": 0.508988156075272,
      "flos": 25885309374720.0,
      "grad_norm": 1.694649681710504,
      "language_loss": 0.80149758,
      "learning_rate": 2.0389461205499354e-06,
      "loss": 0.82362497,
      "num_input_tokens_seen": 91296125,
      "step": 4233,
      "time_per_iteration": 2.7115559577941895
    },
    {
      "auxiliary_loss_clip": 0.01178106,
      "auxiliary_loss_mlp": 0.01027735,
      "balance_loss_clip": 1.05497587,
      "balance_loss_mlp": 1.01918185,
      "epoch": 0.5091083989659111,
      "flos": 13844857057920.0,
      "grad_norm": 2.001443473126748,
      "language_loss": 0.73534477,
      "learning_rate": 2.03816729367533e-06,
      "loss": 0.75740319,
      "num_input_tokens_seen": 91314280,
      "step": 4234,
      "time_per_iteration": 2.599410057067871
    },
    {
      "auxiliary_loss_clip": 0.01179382,
      "auxiliary_loss_mlp": 0.01034573,
      "balance_loss_clip": 1.05776572,
      "balance_loss_mlp": 1.02513194,
      "epoch": 0.5092286418565503,
      "flos": 21104881050240.0,
      "grad_norm": 2.5109913399891246,
      "language_loss": 0.71759021,
      "learning_rate": 2.0373884610107765e-06,
      "loss": 0.73972976,
      "num_input_tokens_seen": 91334595,
      "step": 4235,
      "time_per_iteration": 2.656693458557129
    },
    {
      "auxiliary_loss_clip": 0.01166834,
      "auxiliary_loss_mlp": 0.0102934,
      "balance_loss_clip": 1.05086255,
      "balance_loss_mlp": 1.02020907,
      "epoch": 0.5093488847471893,
      "flos": 18621298972800.0,
      "grad_norm": 2.408409864395338,
      "language_loss": 0.69544333,
      "learning_rate": 2.0366096226744225e-06,
      "loss": 0.71740508,
      "num_input_tokens_seen": 91349790,
      "step": 4236,
      "time_per_iteration": 2.6050143241882324
    },
    {
      "auxiliary_loss_clip": 0.01169865,
      "auxiliary_loss_mlp": 0.01028041,
      "balance_loss_clip": 1.05338621,
      "balance_loss_mlp": 1.01865315,
      "epoch": 0.5094691276378284,
      "flos": 23803783205760.0,
      "grad_norm": 2.0848962350533125,
      "language_loss": 0.76956058,
      "learning_rate": 2.035830778784418e-06,
      "loss": 0.79153967,
      "num_input_tokens_seen": 91370465,
      "step": 4237,
      "time_per_iteration": 2.7291183471679688
    },
    {
      "auxiliary_loss_clip": 0.0119236,
      "auxiliary_loss_mlp": 0.01037456,
      "balance_loss_clip": 1.05887842,
      "balance_loss_mlp": 1.02790141,
      "epoch": 0.5095893705284675,
      "flos": 17420410546560.0,
      "grad_norm": 2.3032049986417666,
      "language_loss": 0.80309296,
      "learning_rate": 2.0350519294589134e-06,
      "loss": 0.82539117,
      "num_input_tokens_seen": 91388505,
      "step": 4238,
      "time_per_iteration": 2.516361713409424
    },
    {
      "auxiliary_loss_clip": 0.01169212,
      "auxiliary_loss_mlp": 0.0103204,
      "balance_loss_clip": 1.05183399,
      "balance_loss_mlp": 1.0219605,
      "epoch": 0.5097096134191066,
      "flos": 25849362839040.0,
      "grad_norm": 1.888807909362668,
      "language_loss": 0.82740271,
      "learning_rate": 2.0342730748160588e-06,
      "loss": 0.84941524,
      "num_input_tokens_seen": 91408970,
      "step": 4239,
      "time_per_iteration": 2.670586109161377
    },
    {
      "auxiliary_loss_clip": 0.01183265,
      "auxiliary_loss_mlp": 0.01031416,
      "balance_loss_clip": 1.05538297,
      "balance_loss_mlp": 1.02168822,
      "epoch": 0.5098298563097456,
      "flos": 27745122844800.0,
      "grad_norm": 2.274705890217665,
      "language_loss": 0.70721012,
      "learning_rate": 2.033494214974006e-06,
      "loss": 0.729357,
      "num_input_tokens_seen": 91430115,
      "step": 4240,
      "time_per_iteration": 2.6585822105407715
    },
    {
      "auxiliary_loss_clip": 0.01176721,
      "auxiliary_loss_mlp": 0.01033144,
      "balance_loss_clip": 1.05416071,
      "balance_loss_mlp": 1.02499616,
      "epoch": 0.5099500992003848,
      "flos": 21358913011200.0,
      "grad_norm": 3.3138065924834277,
      "language_loss": 0.83814871,
      "learning_rate": 2.0327153500509067e-06,
      "loss": 0.86024737,
      "num_input_tokens_seen": 91449140,
      "step": 4241,
      "time_per_iteration": 2.651729106903076
    },
    {
      "auxiliary_loss_clip": 0.01194634,
      "auxiliary_loss_mlp": 0.01031129,
      "balance_loss_clip": 1.05853009,
      "balance_loss_mlp": 1.02126408,
      "epoch": 0.5100703420910239,
      "flos": 19865999013120.0,
      "grad_norm": 3.116458180299132,
      "language_loss": 0.84963316,
      "learning_rate": 2.031936480164916e-06,
      "loss": 0.87189078,
      "num_input_tokens_seen": 91466880,
      "step": 4242,
      "time_per_iteration": 2.7247812747955322
    },
    {
      "auxiliary_loss_clip": 0.01193715,
      "auxiliary_loss_mlp": 0.01028356,
      "balance_loss_clip": 1.05951262,
      "balance_loss_mlp": 1.02021384,
      "epoch": 0.5101905849816629,
      "flos": 24648797635200.0,
      "grad_norm": 4.9509001989320955,
      "language_loss": 0.80552322,
      "learning_rate": 2.0311576054341857e-06,
      "loss": 0.82774389,
      "num_input_tokens_seen": 91487495,
      "step": 4243,
      "time_per_iteration": 2.6737825870513916
    },
    {
      "auxiliary_loss_clip": 0.01195423,
      "auxiliary_loss_mlp": 0.01025272,
      "balance_loss_clip": 1.06078029,
      "balance_loss_mlp": 1.01600957,
      "epoch": 0.5103108278723021,
      "flos": 22930076787840.0,
      "grad_norm": 2.378952625019656,
      "language_loss": 0.6306746,
      "learning_rate": 2.0303787259768715e-06,
      "loss": 0.65288162,
      "num_input_tokens_seen": 91508395,
      "step": 4244,
      "time_per_iteration": 2.6265668869018555
    },
    {
      "auxiliary_loss_clip": 0.01175836,
      "auxiliary_loss_mlp": 0.0076374,
      "balance_loss_clip": 1.05836892,
      "balance_loss_mlp": 1.00117445,
      "epoch": 0.5104310707629411,
      "flos": 21506613736320.0,
      "grad_norm": 3.2789933322866798,
      "language_loss": 0.69130993,
      "learning_rate": 2.0295998419111294e-06,
      "loss": 0.71070564,
      "num_input_tokens_seen": 91525685,
      "step": 4245,
      "time_per_iteration": 2.6165459156036377
    },
    {
      "auxiliary_loss_clip": 0.0117894,
      "auxiliary_loss_mlp": 0.01029889,
      "balance_loss_clip": 1.05705333,
      "balance_loss_mlp": 1.02090013,
      "epoch": 0.5105513136535802,
      "flos": 14903180403840.0,
      "grad_norm": 2.3518147295986864,
      "language_loss": 0.73144388,
      "learning_rate": 2.028820953355115e-06,
      "loss": 0.75353211,
      "num_input_tokens_seen": 91543785,
      "step": 4246,
      "time_per_iteration": 2.6407887935638428
    },
    {
      "auxiliary_loss_clip": 0.01194695,
      "auxiliary_loss_mlp": 0.01033411,
      "balance_loss_clip": 1.05755842,
      "balance_loss_mlp": 1.02376056,
      "epoch": 0.5106715565442194,
      "flos": 22602212421120.0,
      "grad_norm": 1.8871658402418767,
      "language_loss": 0.7857765,
      "learning_rate": 2.0280420604269834e-06,
      "loss": 0.80805755,
      "num_input_tokens_seen": 91563325,
      "step": 4247,
      "time_per_iteration": 2.5787830352783203
    },
    {
      "auxiliary_loss_clip": 0.01077003,
      "auxiliary_loss_mlp": 0.01003551,
      "balance_loss_clip": 1.01287532,
      "balance_loss_mlp": 1.0024904,
      "epoch": 0.5107917994348584,
      "flos": 71027645558400.0,
      "grad_norm": 0.7082305346861494,
      "language_loss": 0.58867157,
      "learning_rate": 2.027263163244895e-06,
      "loss": 0.6094771,
      "num_input_tokens_seen": 91632450,
      "step": 4248,
      "time_per_iteration": 3.404087543487549
    },
    {
      "auxiliary_loss_clip": 0.01177252,
      "auxiliary_loss_mlp": 0.01033504,
      "balance_loss_clip": 1.05808604,
      "balance_loss_mlp": 1.02474856,
      "epoch": 0.5109120423254975,
      "flos": 24827416992000.0,
      "grad_norm": 2.102264808615688,
      "language_loss": 0.74501193,
      "learning_rate": 2.026484261927005e-06,
      "loss": 0.76711953,
      "num_input_tokens_seen": 91651945,
      "step": 4249,
      "time_per_iteration": 3.621330499649048
    },
    {
      "auxiliary_loss_clip": 0.01186072,
      "auxiliary_loss_mlp": 0.01034595,
      "balance_loss_clip": 1.05868399,
      "balance_loss_mlp": 1.02498102,
      "epoch": 0.5110322852161366,
      "flos": 21247661612160.0,
      "grad_norm": 2.690994487780425,
      "language_loss": 0.74258089,
      "learning_rate": 2.025705356591475e-06,
      "loss": 0.76478767,
      "num_input_tokens_seen": 91669635,
      "step": 4250,
      "time_per_iteration": 2.720048189163208
    },
    {
      "auxiliary_loss_clip": 0.01077738,
      "auxiliary_loss_mlp": 0.01003413,
      "balance_loss_clip": 1.01357651,
      "balance_loss_mlp": 1.00240541,
      "epoch": 0.5111525281067757,
      "flos": 66457114358400.0,
      "grad_norm": 0.7642255156211482,
      "language_loss": 0.57941198,
      "learning_rate": 2.024926447356462e-06,
      "loss": 0.60022348,
      "num_input_tokens_seen": 91731920,
      "step": 4251,
      "time_per_iteration": 4.968192100524902
    },
    {
      "auxiliary_loss_clip": 0.01179074,
      "auxiliary_loss_mlp": 0.01035291,
      "balance_loss_clip": 1.05565703,
      "balance_loss_mlp": 1.0250448,
      "epoch": 0.5112727709974147,
      "flos": 14866731077760.0,
      "grad_norm": 2.191611331971667,
      "language_loss": 0.79503477,
      "learning_rate": 2.024147534340127e-06,
      "loss": 0.81717843,
      "num_input_tokens_seen": 91749780,
      "step": 4252,
      "time_per_iteration": 2.6646997928619385
    },
    {
      "auxiliary_loss_clip": 0.011773,
      "auxiliary_loss_mlp": 0.01030678,
      "balance_loss_clip": 1.05492449,
      "balance_loss_mlp": 1.02219045,
      "epoch": 0.5113930138880539,
      "flos": 21177600134400.0,
      "grad_norm": 1.658455195438162,
      "language_loss": 0.80030775,
      "learning_rate": 2.02336861766063e-06,
      "loss": 0.82238758,
      "num_input_tokens_seen": 91768840,
      "step": 4253,
      "time_per_iteration": 2.632490873336792
    },
    {
      "auxiliary_loss_clip": 0.0118798,
      "auxiliary_loss_mlp": 0.01039281,
      "balance_loss_clip": 1.0588268,
      "balance_loss_mlp": 1.02865994,
      "epoch": 0.511513256778693,
      "flos": 20409111630720.0,
      "grad_norm": 2.9311227760961116,
      "language_loss": 0.78812683,
      "learning_rate": 2.0225896974361327e-06,
      "loss": 0.81039941,
      "num_input_tokens_seen": 91788945,
      "step": 4254,
      "time_per_iteration": 2.6830906867980957
    },
    {
      "auxiliary_loss_clip": 0.01057056,
      "auxiliary_loss_mlp": 0.01007688,
      "balance_loss_clip": 1.02388537,
      "balance_loss_mlp": 1.00661492,
      "epoch": 0.511633499669332,
      "flos": 69879975131520.0,
      "grad_norm": 0.8556832273457184,
      "language_loss": 0.59919,
      "learning_rate": 2.0218107737847962e-06,
      "loss": 0.6198374,
      "num_input_tokens_seen": 91850990,
      "step": 4255,
      "time_per_iteration": 3.334580898284912
    },
    {
      "auxiliary_loss_clip": 0.01173639,
      "auxiliary_loss_mlp": 0.01033832,
      "balance_loss_clip": 1.05687451,
      "balance_loss_mlp": 1.02499843,
      "epoch": 0.5117537425599712,
      "flos": 24097855852800.0,
      "grad_norm": 3.0971941800365164,
      "language_loss": 0.75015533,
      "learning_rate": 2.0210318468247826e-06,
      "loss": 0.77223003,
      "num_input_tokens_seen": 91869960,
      "step": 4256,
      "time_per_iteration": 2.7097063064575195
    },
    {
      "auxiliary_loss_clip": 0.01186955,
      "auxiliary_loss_mlp": 0.01039231,
      "balance_loss_clip": 1.05412757,
      "balance_loss_mlp": 1.0296762,
      "epoch": 0.5118739854506102,
      "flos": 20959550622720.0,
      "grad_norm": 3.6025368815858667,
      "language_loss": 0.81925297,
      "learning_rate": 2.020252916674255e-06,
      "loss": 0.84151483,
      "num_input_tokens_seen": 91889075,
      "step": 4257,
      "time_per_iteration": 2.580331563949585
    },
    {
      "auxiliary_loss_clip": 0.01147944,
      "auxiliary_loss_mlp": 0.0103992,
      "balance_loss_clip": 1.05656052,
      "balance_loss_mlp": 1.0305562,
      "epoch": 0.5119942283412493,
      "flos": 17457326749440.0,
      "grad_norm": 5.718164477387293,
      "language_loss": 0.80950058,
      "learning_rate": 2.019473983451375e-06,
      "loss": 0.83137929,
      "num_input_tokens_seen": 91907495,
      "step": 4258,
      "time_per_iteration": 2.873483180999756
    },
    {
      "auxiliary_loss_clip": 0.01193723,
      "auxiliary_loss_mlp": 0.01036209,
      "balance_loss_clip": 1.0569706,
      "balance_loss_mlp": 1.02694631,
      "epoch": 0.5121144712318885,
      "flos": 21066743784960.0,
      "grad_norm": 1.8280184979026552,
      "language_loss": 0.71589184,
      "learning_rate": 2.0186950472743076e-06,
      "loss": 0.73819113,
      "num_input_tokens_seen": 91927400,
      "step": 4259,
      "time_per_iteration": 3.5687413215637207
    },
    {
      "auxiliary_loss_clip": 0.01173356,
      "auxiliary_loss_mlp": 0.01029276,
      "balance_loss_clip": 1.05421352,
      "balance_loss_mlp": 1.02019238,
      "epoch": 0.5122347141225275,
      "flos": 19860791541120.0,
      "grad_norm": 1.72058802952863,
      "language_loss": 0.74151886,
      "learning_rate": 2.0179161082612162e-06,
      "loss": 0.76354516,
      "num_input_tokens_seen": 91946790,
      "step": 4260,
      "time_per_iteration": 2.642179489135742
    },
    {
      "auxiliary_loss_clip": 0.01162536,
      "auxiliary_loss_mlp": 0.01037355,
      "balance_loss_clip": 1.05390453,
      "balance_loss_mlp": 1.02809799,
      "epoch": 0.5123549570131666,
      "flos": 22528487756160.0,
      "grad_norm": 2.494093771319493,
      "language_loss": 0.72951114,
      "learning_rate": 2.017137166530266e-06,
      "loss": 0.75151002,
      "num_input_tokens_seen": 91966325,
      "step": 4261,
      "time_per_iteration": 2.6731979846954346
    },
    {
      "auxiliary_loss_clip": 0.01177772,
      "auxiliary_loss_mlp": 0.01032548,
      "balance_loss_clip": 1.0571599,
      "balance_loss_mlp": 1.02355385,
      "epoch": 0.5124751999038056,
      "flos": 20333375804160.0,
      "grad_norm": 18.262582927248797,
      "language_loss": 0.80184853,
      "learning_rate": 2.0163582221996213e-06,
      "loss": 0.82395178,
      "num_input_tokens_seen": 91984700,
      "step": 4262,
      "time_per_iteration": 2.594475269317627
    },
    {
      "auxiliary_loss_clip": 0.01192145,
      "auxiliary_loss_mlp": 0.01034252,
      "balance_loss_clip": 1.05644071,
      "balance_loss_mlp": 1.02487636,
      "epoch": 0.5125954427944448,
      "flos": 39785970211200.0,
      "grad_norm": 9.493717200785792,
      "language_loss": 0.68256414,
      "learning_rate": 2.015579275387446e-06,
      "loss": 0.70482808,
      "num_input_tokens_seen": 92010020,
      "step": 4263,
      "time_per_iteration": 2.773897171020508
    },
    {
      "auxiliary_loss_clip": 0.01161141,
      "auxiliary_loss_mlp": 0.01032448,
      "balance_loss_clip": 1.05370021,
      "balance_loss_mlp": 1.02347755,
      "epoch": 0.5127156856850839,
      "flos": 29205394358400.0,
      "grad_norm": 2.3503149929678604,
      "language_loss": 0.69082999,
      "learning_rate": 2.0148003262119085e-06,
      "loss": 0.71276593,
      "num_input_tokens_seen": 92030990,
      "step": 4264,
      "time_per_iteration": 2.706711530685425
    },
    {
      "auxiliary_loss_clip": 0.01174051,
      "auxiliary_loss_mlp": 0.01045458,
      "balance_loss_clip": 1.05527925,
      "balance_loss_mlp": 1.03550458,
      "epoch": 0.5128359285757229,
      "flos": 13553693412480.0,
      "grad_norm": 2.5956631304616176,
      "language_loss": 0.76598656,
      "learning_rate": 2.0140213747911728e-06,
      "loss": 0.78818166,
      "num_input_tokens_seen": 92049525,
      "step": 4265,
      "time_per_iteration": 2.622758388519287
    },
    {
      "auxiliary_loss_clip": 0.01180498,
      "auxiliary_loss_mlp": 0.01032943,
      "balance_loss_clip": 1.05845666,
      "balance_loss_mlp": 1.02370977,
      "epoch": 0.5129561714663621,
      "flos": 25192089820800.0,
      "grad_norm": 2.2094753087256525,
      "language_loss": 0.81087726,
      "learning_rate": 2.013242421243406e-06,
      "loss": 0.83301169,
      "num_input_tokens_seen": 92068430,
      "step": 4266,
      "time_per_iteration": 2.7011232376098633
    },
    {
      "auxiliary_loss_clip": 0.01166323,
      "auxiliary_loss_mlp": 0.01033705,
      "balance_loss_clip": 1.0541687,
      "balance_loss_mlp": 1.02450824,
      "epoch": 0.5130764143570011,
      "flos": 18150223080960.0,
      "grad_norm": 1.578487162410511,
      "language_loss": 0.79439098,
      "learning_rate": 2.012463465686774e-06,
      "loss": 0.81639123,
      "num_input_tokens_seen": 92088180,
      "step": 4267,
      "time_per_iteration": 2.698204755783081
    },
    {
      "auxiliary_loss_clip": 0.01077099,
      "auxiliary_loss_mlp": 0.01009998,
      "balance_loss_clip": 1.01294184,
      "balance_loss_mlp": 1.0089426,
      "epoch": 0.5131966572476402,
      "flos": 59794896418560.0,
      "grad_norm": 0.8237550307397675,
      "language_loss": 0.54674143,
      "learning_rate": 2.0116845082394446e-06,
      "loss": 0.56761235,
      "num_input_tokens_seen": 92153015,
      "step": 4268,
      "time_per_iteration": 3.176621913909912
    },
    {
      "auxiliary_loss_clip": 0.01179292,
      "auxiliary_loss_mlp": 0.01031954,
      "balance_loss_clip": 1.0576098,
      "balance_loss_mlp": 1.02313256,
      "epoch": 0.5133169001382794,
      "flos": 18515219132160.0,
      "grad_norm": 2.825757108119924,
      "language_loss": 0.78819966,
      "learning_rate": 2.0109055490195836e-06,
      "loss": 0.81031203,
      "num_input_tokens_seen": 92171470,
      "step": 4269,
      "time_per_iteration": 2.649559736251831
    },
    {
      "auxiliary_loss_clip": 0.01192577,
      "auxiliary_loss_mlp": 0.01031986,
      "balance_loss_clip": 1.05705118,
      "balance_loss_mlp": 1.02258646,
      "epoch": 0.5134371430289184,
      "flos": 15523537219200.0,
      "grad_norm": 2.715394455320574,
      "language_loss": 0.64675862,
      "learning_rate": 2.0101265881453605e-06,
      "loss": 0.66900426,
      "num_input_tokens_seen": 92189945,
      "step": 4270,
      "time_per_iteration": 2.62930965423584
    },
    {
      "auxiliary_loss_clip": 0.01191585,
      "auxiliary_loss_mlp": 0.01034231,
      "balance_loss_clip": 1.06022453,
      "balance_loss_mlp": 1.02606511,
      "epoch": 0.5135573859195575,
      "flos": 21433786911360.0,
      "grad_norm": 2.4563375046896785,
      "language_loss": 0.78483737,
      "learning_rate": 2.009347625734941e-06,
      "loss": 0.80709553,
      "num_input_tokens_seen": 92209855,
      "step": 4271,
      "time_per_iteration": 2.6135172843933105
    },
    {
      "auxiliary_loss_clip": 0.01182779,
      "auxiliary_loss_mlp": 0.01034847,
      "balance_loss_clip": 1.05699396,
      "balance_loss_mlp": 1.02582872,
      "epoch": 0.5136776288101966,
      "flos": 17712651600000.0,
      "grad_norm": 2.47627555355915,
      "language_loss": 0.75928009,
      "learning_rate": 2.0085686619064954e-06,
      "loss": 0.78145647,
      "num_input_tokens_seen": 92226295,
      "step": 4272,
      "time_per_iteration": 2.679234027862549
    },
    {
      "auxiliary_loss_clip": 0.01183285,
      "auxiliary_loss_mlp": 0.01033547,
      "balance_loss_clip": 1.05936742,
      "balance_loss_mlp": 1.02373576,
      "epoch": 0.5137978717008357,
      "flos": 16581680997120.0,
      "grad_norm": 2.1242644860047006,
      "language_loss": 0.82681715,
      "learning_rate": 2.00778969677819e-06,
      "loss": 0.84898555,
      "num_input_tokens_seen": 92243330,
      "step": 4273,
      "time_per_iteration": 2.6053435802459717
    },
    {
      "auxiliary_loss_clip": 0.01190739,
      "auxiliary_loss_mlp": 0.00763178,
      "balance_loss_clip": 1.05683899,
      "balance_loss_mlp": 1.00132012,
      "epoch": 0.5139181145914747,
      "flos": 20668243322880.0,
      "grad_norm": 2.6529474057619007,
      "language_loss": 0.64427233,
      "learning_rate": 2.0070107304681934e-06,
      "loss": 0.6638115,
      "num_input_tokens_seen": 92262285,
      "step": 4274,
      "time_per_iteration": 2.8543496131896973
    },
    {
      "auxiliary_loss_clip": 0.01165739,
      "auxiliary_loss_mlp": 0.01027704,
      "balance_loss_clip": 1.05229354,
      "balance_loss_mlp": 1.01929379,
      "epoch": 0.5140383574821139,
      "flos": 32926996546560.0,
      "grad_norm": 1.8146939179243033,
      "language_loss": 0.78315568,
      "learning_rate": 2.006231763094675e-06,
      "loss": 0.80509013,
      "num_input_tokens_seen": 92283305,
      "step": 4275,
      "time_per_iteration": 3.6561853885650635
    },
    {
      "auxiliary_loss_clip": 0.01179876,
      "auxiliary_loss_mlp": 0.01031433,
      "balance_loss_clip": 1.05701172,
      "balance_loss_mlp": 1.02276051,
      "epoch": 0.514158600372753,
      "flos": 19537093152000.0,
      "grad_norm": 2.3502661608320627,
      "language_loss": 0.87977207,
      "learning_rate": 2.0054527947758027e-06,
      "loss": 0.90188509,
      "num_input_tokens_seen": 92302105,
      "step": 4276,
      "time_per_iteration": 2.6031062602996826
    },
    {
      "auxiliary_loss_clip": 0.01074906,
      "auxiliary_loss_mlp": 0.01002923,
      "balance_loss_clip": 1.01118541,
      "balance_loss_mlp": 1.00190949,
      "epoch": 0.514278843263392,
      "flos": 62523855279360.0,
      "grad_norm": 0.7230548799371945,
      "language_loss": 0.55848229,
      "learning_rate": 2.004673825629746e-06,
      "loss": 0.57926059,
      "num_input_tokens_seen": 92362885,
      "step": 4277,
      "time_per_iteration": 5.135640621185303
    },
    {
      "auxiliary_loss_clip": 0.01160364,
      "auxiliary_loss_mlp": 0.01030081,
      "balance_loss_clip": 1.05519438,
      "balance_loss_mlp": 1.02114642,
      "epoch": 0.5143990861540312,
      "flos": 25882328545920.0,
      "grad_norm": 2.095992137878352,
      "language_loss": 0.72281063,
      "learning_rate": 2.0038948557746744e-06,
      "loss": 0.74471509,
      "num_input_tokens_seen": 92384740,
      "step": 4278,
      "time_per_iteration": 2.773432493209839
    },
    {
      "auxiliary_loss_clip": 0.01191216,
      "auxiliary_loss_mlp": 0.01031953,
      "balance_loss_clip": 1.05653906,
      "balance_loss_mlp": 1.02288771,
      "epoch": 0.5145193290446702,
      "flos": 23330660238720.0,
      "grad_norm": 2.780419342497299,
      "language_loss": 0.75094956,
      "learning_rate": 2.0031158853287558e-06,
      "loss": 0.77318126,
      "num_input_tokens_seen": 92405175,
      "step": 4279,
      "time_per_iteration": 2.6206226348876953
    },
    {
      "auxiliary_loss_clip": 0.01192734,
      "auxiliary_loss_mlp": 0.01036572,
      "balance_loss_clip": 1.05817604,
      "balance_loss_mlp": 1.02807212,
      "epoch": 0.5146395719353093,
      "flos": 22856603518080.0,
      "grad_norm": 3.0446521431029825,
      "language_loss": 0.70885003,
      "learning_rate": 2.0023369144101593e-06,
      "loss": 0.73114306,
      "num_input_tokens_seen": 92423345,
      "step": 4280,
      "time_per_iteration": 2.7161104679107666
    },
    {
      "auxiliary_loss_clip": 0.01152685,
      "auxiliary_loss_mlp": 0.01027535,
      "balance_loss_clip": 1.05053222,
      "balance_loss_mlp": 1.0187428,
      "epoch": 0.5147598148259485,
      "flos": 26391577616640.0,
      "grad_norm": 1.7639016737597544,
      "language_loss": 0.76771706,
      "learning_rate": 2.0015579431370555e-06,
      "loss": 0.78951925,
      "num_input_tokens_seen": 92445025,
      "step": 4281,
      "time_per_iteration": 2.8288540840148926
    },
    {
      "auxiliary_loss_clip": 0.01178991,
      "auxiliary_loss_mlp": 0.01032041,
      "balance_loss_clip": 1.05821514,
      "balance_loss_mlp": 1.02364326,
      "epoch": 0.5148800577165875,
      "flos": 29965694561280.0,
      "grad_norm": 2.4707942410695485,
      "language_loss": 0.70421779,
      "learning_rate": 2.000778971627612e-06,
      "loss": 0.72632802,
      "num_input_tokens_seen": 92464490,
      "step": 4282,
      "time_per_iteration": 2.7610068321228027
    },
    {
      "auxiliary_loss_clip": 0.01177736,
      "auxiliary_loss_mlp": 0.01036758,
      "balance_loss_clip": 1.05498815,
      "balance_loss_mlp": 1.0279665,
      "epoch": 0.5150003006072266,
      "flos": 17931383470080.0,
      "grad_norm": 8.223009785235746,
      "language_loss": 0.90381038,
      "learning_rate": 2e-06,
      "loss": 0.9259553,
      "num_input_tokens_seen": 92482085,
      "step": 4283,
      "time_per_iteration": 2.667969226837158
    },
    {
      "auxiliary_loss_clip": 0.01189152,
      "auxiliary_loss_mlp": 0.01030061,
      "balance_loss_clip": 1.05711699,
      "balance_loss_mlp": 1.02115619,
      "epoch": 0.5151205434978657,
      "flos": 18478733892480.0,
      "grad_norm": 1.881023104164794,
      "language_loss": 0.85886359,
      "learning_rate": 1.9992210283723878e-06,
      "loss": 0.88105577,
      "num_input_tokens_seen": 92499325,
      "step": 4284,
      "time_per_iteration": 2.6265690326690674
    },
    {
      "auxiliary_loss_clip": 0.01187711,
      "auxiliary_loss_mlp": 0.01026258,
      "balance_loss_clip": 1.05679846,
      "balance_loss_mlp": 1.01785636,
      "epoch": 0.5152407863885048,
      "flos": 25341263003520.0,
      "grad_norm": 1.6501787506245678,
      "language_loss": 0.7940048,
      "learning_rate": 1.9984420568629448e-06,
      "loss": 0.81614447,
      "num_input_tokens_seen": 92522090,
      "step": 4285,
      "time_per_iteration": 3.629162549972534
    },
    {
      "auxiliary_loss_clip": 0.01164391,
      "auxiliary_loss_mlp": 0.01031012,
      "balance_loss_clip": 1.05405164,
      "balance_loss_mlp": 1.02160025,
      "epoch": 0.5153610292791438,
      "flos": 18329740277760.0,
      "grad_norm": 2.0632942559214187,
      "language_loss": 0.79100269,
      "learning_rate": 1.9976630855898405e-06,
      "loss": 0.81295675,
      "num_input_tokens_seen": 92539845,
      "step": 4286,
      "time_per_iteration": 2.6640944480895996
    },
    {
      "auxiliary_loss_clip": 0.01140924,
      "auxiliary_loss_mlp": 0.01033623,
      "balance_loss_clip": 1.04782844,
      "balance_loss_mlp": 1.02557623,
      "epoch": 0.515481272169783,
      "flos": 30409945971840.0,
      "grad_norm": 8.555335679755945,
      "language_loss": 0.74991798,
      "learning_rate": 1.9968841146712445e-06,
      "loss": 0.77166343,
      "num_input_tokens_seen": 92559460,
      "step": 4287,
      "time_per_iteration": 2.8705546855926514
    },
    {
      "auxiliary_loss_clip": 0.01164185,
      "auxiliary_loss_mlp": 0.01030103,
      "balance_loss_clip": 1.05396855,
      "balance_loss_mlp": 1.02129376,
      "epoch": 0.5156015150604221,
      "flos": 23037305863680.0,
      "grad_norm": 2.464647826228789,
      "language_loss": 0.71506369,
      "learning_rate": 1.996105144225326e-06,
      "loss": 0.73700655,
      "num_input_tokens_seen": 92579695,
      "step": 4288,
      "time_per_iteration": 2.6803433895111084
    },
    {
      "auxiliary_loss_clip": 0.01190596,
      "auxiliary_loss_mlp": 0.01031593,
      "balance_loss_clip": 1.05724835,
      "balance_loss_mlp": 1.02320671,
      "epoch": 0.5157217579510611,
      "flos": 17858556645120.0,
      "grad_norm": 1.7947054939015212,
      "language_loss": 0.7909531,
      "learning_rate": 1.995326174370254e-06,
      "loss": 0.81317496,
      "num_input_tokens_seen": 92598795,
      "step": 4289,
      "time_per_iteration": 2.6462008953094482
    },
    {
      "auxiliary_loss_clip": 0.01168572,
      "auxiliary_loss_mlp": 0.01031687,
      "balance_loss_clip": 1.05363131,
      "balance_loss_mlp": 1.02403414,
      "epoch": 0.5158420008417003,
      "flos": 19171486569600.0,
      "grad_norm": 1.5999731701319315,
      "language_loss": 0.73334479,
      "learning_rate": 1.994547205224197e-06,
      "loss": 0.75534737,
      "num_input_tokens_seen": 92617700,
      "step": 4290,
      "time_per_iteration": 2.5959115028381348
    },
    {
      "auxiliary_loss_clip": 0.01175957,
      "auxiliary_loss_mlp": 0.01029645,
      "balance_loss_clip": 1.05350876,
      "balance_loss_mlp": 1.02143168,
      "epoch": 0.5159622437323393,
      "flos": 22419534827520.0,
      "grad_norm": 2.85290998516251,
      "language_loss": 0.67737389,
      "learning_rate": 1.993768236905325e-06,
      "loss": 0.69942993,
      "num_input_tokens_seen": 92638370,
      "step": 4291,
      "time_per_iteration": 2.775796890258789
    },
    {
      "auxiliary_loss_clip": 0.01174132,
      "auxiliary_loss_mlp": 0.01029864,
      "balance_loss_clip": 1.05170774,
      "balance_loss_mlp": 1.02083957,
      "epoch": 0.5160824866229784,
      "flos": 24603010773120.0,
      "grad_norm": 6.348022522110698,
      "language_loss": 0.66233987,
      "learning_rate": 1.992989269531807e-06,
      "loss": 0.68437982,
      "num_input_tokens_seen": 92657180,
      "step": 4292,
      "time_per_iteration": 2.6906142234802246
    },
    {
      "auxiliary_loss_clip": 0.01177431,
      "auxiliary_loss_mlp": 0.01031373,
      "balance_loss_clip": 1.05658865,
      "balance_loss_mlp": 1.02308202,
      "epoch": 0.5162027295136175,
      "flos": 18002737837440.0,
      "grad_norm": 2.605655593689613,
      "language_loss": 0.68961966,
      "learning_rate": 1.99221030322181e-06,
      "loss": 0.71170771,
      "num_input_tokens_seen": 92673985,
      "step": 4293,
      "time_per_iteration": 2.6250665187835693
    },
    {
      "auxiliary_loss_clip": 0.0112929,
      "auxiliary_loss_mlp": 0.01041777,
      "balance_loss_clip": 1.04831672,
      "balance_loss_mlp": 1.03311062,
      "epoch": 0.5163229724042566,
      "flos": 27344611221120.0,
      "grad_norm": 1.6099140417403721,
      "language_loss": 0.80860525,
      "learning_rate": 1.991431338093505e-06,
      "loss": 0.83031589,
      "num_input_tokens_seen": 92696340,
      "step": 4294,
      "time_per_iteration": 2.810410261154175
    },
    {
      "auxiliary_loss_clip": 0.01191404,
      "auxiliary_loss_mlp": 0.01028671,
      "balance_loss_clip": 1.05990624,
      "balance_loss_mlp": 1.019629,
      "epoch": 0.5164432152948957,
      "flos": 21762764599680.0,
      "grad_norm": 1.8999104367185933,
      "language_loss": 0.795443,
      "learning_rate": 1.9906523742650587e-06,
      "loss": 0.81764376,
      "num_input_tokens_seen": 92715200,
      "step": 4295,
      "time_per_iteration": 2.6065433025360107
    },
    {
      "auxiliary_loss_clip": 0.01172707,
      "auxiliary_loss_mlp": 0.01033136,
      "balance_loss_clip": 1.05148876,
      "balance_loss_mlp": 1.02409446,
      "epoch": 0.5165634581855347,
      "flos": 25550334115200.0,
      "grad_norm": 2.5325678757839882,
      "language_loss": 0.77549982,
      "learning_rate": 1.9898734118546397e-06,
      "loss": 0.79755819,
      "num_input_tokens_seen": 92735150,
      "step": 4296,
      "time_per_iteration": 2.7346441745758057
    },
    {
      "auxiliary_loss_clip": 0.0119222,
      "auxiliary_loss_mlp": 0.01041099,
      "balance_loss_clip": 1.05870557,
      "balance_loss_mlp": 1.03146696,
      "epoch": 0.5166837010761739,
      "flos": 19901191363200.0,
      "grad_norm": 1.6276016115885443,
      "language_loss": 0.80358803,
      "learning_rate": 1.989094450980416e-06,
      "loss": 0.82592124,
      "num_input_tokens_seen": 92755250,
      "step": 4297,
      "time_per_iteration": 2.606447696685791
    },
    {
      "auxiliary_loss_clip": 0.01188313,
      "auxiliary_loss_mlp": 0.01031011,
      "balance_loss_clip": 1.05578232,
      "balance_loss_mlp": 1.02283359,
      "epoch": 0.516803943966813,
      "flos": 26646076454400.0,
      "grad_norm": 2.847294453697517,
      "language_loss": 0.77140796,
      "learning_rate": 1.9883154917605556e-06,
      "loss": 0.79360121,
      "num_input_tokens_seen": 92774460,
      "step": 4298,
      "time_per_iteration": 2.696582794189453
    },
    {
      "auxiliary_loss_clip": 0.0116459,
      "auxiliary_loss_mlp": 0.01028821,
      "balance_loss_clip": 1.05219281,
      "balance_loss_mlp": 1.02028525,
      "epoch": 0.516924186857452,
      "flos": 19682854542720.0,
      "grad_norm": 1.9350781544157587,
      "language_loss": 0.83277529,
      "learning_rate": 1.9875365343132262e-06,
      "loss": 0.85470933,
      "num_input_tokens_seen": 92791580,
      "step": 4299,
      "time_per_iteration": 2.624462604522705
    },
    {
      "auxiliary_loss_clip": 0.01194115,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 1.05955946,
      "balance_loss_mlp": 1.01925635,
      "epoch": 0.5170444297480912,
      "flos": 15956583586560.0,
      "grad_norm": 2.63340277555296,
      "language_loss": 0.84822667,
      "learning_rate": 1.9867575787565946e-06,
      "loss": 0.87044734,
      "num_input_tokens_seen": 92806240,
      "step": 4300,
      "time_per_iteration": 2.635941982269287
    },
    {
      "auxiliary_loss_clip": 0.01174614,
      "auxiliary_loss_mlp": 0.01027435,
      "balance_loss_clip": 1.05573416,
      "balance_loss_mlp": 1.0185833,
      "epoch": 0.5171646726387302,
      "flos": 14174157968640.0,
      "grad_norm": 2.0750558606232783,
      "language_loss": 0.86521971,
      "learning_rate": 1.9859786252088275e-06,
      "loss": 0.88724029,
      "num_input_tokens_seen": 92823420,
      "step": 4301,
      "time_per_iteration": 3.3824641704559326
    },
    {
      "auxiliary_loss_clip": 0.01194974,
      "auxiliary_loss_mlp": 0.01032314,
      "balance_loss_clip": 1.05770397,
      "balance_loss_mlp": 1.02314711,
      "epoch": 0.5172849155293693,
      "flos": 23578550974080.0,
      "grad_norm": 3.16994046929751,
      "language_loss": 0.66505086,
      "learning_rate": 1.9851996737880914e-06,
      "loss": 0.68732369,
      "num_input_tokens_seen": 92838605,
      "step": 4302,
      "time_per_iteration": 2.6750190258026123
    },
    {
      "auxiliary_loss_clip": 0.01178339,
      "auxiliary_loss_mlp": 0.01034907,
      "balance_loss_clip": 1.05682385,
      "balance_loss_mlp": 1.02603757,
      "epoch": 0.5174051584200084,
      "flos": 14283541860480.0,
      "grad_norm": 3.8496693562872193,
      "language_loss": 0.74997211,
      "learning_rate": 1.9844207246125537e-06,
      "loss": 0.77210462,
      "num_input_tokens_seen": 92855185,
      "step": 4303,
      "time_per_iteration": 2.6715097427368164
    },
    {
      "auxiliary_loss_clip": 0.01176396,
      "auxiliary_loss_mlp": 0.0103403,
      "balance_loss_clip": 1.05818796,
      "balance_loss_mlp": 1.02564907,
      "epoch": 0.5175254013106475,
      "flos": 37889384192640.0,
      "grad_norm": 2.161040913174511,
      "language_loss": 0.6902771,
      "learning_rate": 1.983641777800379e-06,
      "loss": 0.7123813,
      "num_input_tokens_seen": 92877830,
      "step": 4304,
      "time_per_iteration": 3.74688982963562
    },
    {
      "auxiliary_loss_clip": 0.01073024,
      "auxiliary_loss_mlp": 0.01004121,
      "balance_loss_clip": 1.00955749,
      "balance_loss_mlp": 1.00311399,
      "epoch": 0.5176456442012866,
      "flos": 68549737829760.0,
      "grad_norm": 0.7404764622298394,
      "language_loss": 0.58707726,
      "learning_rate": 1.9828628334697343e-06,
      "loss": 0.6078487,
      "num_input_tokens_seen": 92945040,
      "step": 4305,
      "time_per_iteration": 3.340167999267578
    },
    {
      "auxiliary_loss_clip": 0.01073414,
      "auxiliary_loss_mlp": 0.0100398,
      "balance_loss_clip": 1.00984991,
      "balance_loss_mlp": 1.0029366,
      "epoch": 0.5177658870919257,
      "flos": 64084137235200.0,
      "grad_norm": 0.7605371972888926,
      "language_loss": 0.54635417,
      "learning_rate": 1.982083891738784e-06,
      "loss": 0.56712806,
      "num_input_tokens_seen": 93005910,
      "step": 4306,
      "time_per_iteration": 3.243309736251831
    },
    {
      "auxiliary_loss_clip": 0.01170068,
      "auxiliary_loss_mlp": 0.01028069,
      "balance_loss_clip": 1.05557156,
      "balance_loss_mlp": 1.01993251,
      "epoch": 0.5178861299825648,
      "flos": 26651248012800.0,
      "grad_norm": 1.4867864001987832,
      "language_loss": 0.82852018,
      "learning_rate": 1.9813049527256923e-06,
      "loss": 0.85050154,
      "num_input_tokens_seen": 93026305,
      "step": 4307,
      "time_per_iteration": 2.6819841861724854
    },
    {
      "auxiliary_loss_clip": 0.0119206,
      "auxiliary_loss_mlp": 0.01039184,
      "balance_loss_clip": 1.05810177,
      "balance_loss_mlp": 1.03062773,
      "epoch": 0.5180063728732038,
      "flos": 17931886260480.0,
      "grad_norm": 2.313894467639496,
      "language_loss": 0.82141626,
      "learning_rate": 1.9805260165486252e-06,
      "loss": 0.84372866,
      "num_input_tokens_seen": 93045675,
      "step": 4308,
      "time_per_iteration": 2.7480597496032715
    },
    {
      "auxiliary_loss_clip": 0.01158877,
      "auxiliary_loss_mlp": 0.01031008,
      "balance_loss_clip": 1.05739999,
      "balance_loss_mlp": 1.02243114,
      "epoch": 0.518126615763843,
      "flos": 19500895221120.0,
      "grad_norm": 3.740645675531796,
      "language_loss": 0.86628234,
      "learning_rate": 1.9797470833257457e-06,
      "loss": 0.88818121,
      "num_input_tokens_seen": 93065375,
      "step": 4309,
      "time_per_iteration": 2.687279462814331
    },
    {
      "auxiliary_loss_clip": 0.01184789,
      "auxiliary_loss_mlp": 0.01031815,
      "balance_loss_clip": 1.05860806,
      "balance_loss_mlp": 1.02298117,
      "epoch": 0.5182468586544821,
      "flos": 20704082117760.0,
      "grad_norm": 3.69417283218253,
      "language_loss": 0.7791906,
      "learning_rate": 1.9789681531752177e-06,
      "loss": 0.80135667,
      "num_input_tokens_seen": 93085595,
      "step": 4310,
      "time_per_iteration": 2.6688969135284424
    },
    {
      "auxiliary_loss_clip": 0.01174917,
      "auxiliary_loss_mlp": 0.01035449,
      "balance_loss_clip": 1.0558939,
      "balance_loss_mlp": 1.02714562,
      "epoch": 0.5183671015451211,
      "flos": 23112107936640.0,
      "grad_norm": 2.026736944524809,
      "language_loss": 0.72698665,
      "learning_rate": 1.978189226215204e-06,
      "loss": 0.74909031,
      "num_input_tokens_seen": 93106140,
      "step": 4311,
      "time_per_iteration": 2.6698195934295654
    },
    {
      "auxiliary_loss_clip": 0.0119271,
      "auxiliary_loss_mlp": 0.01029352,
      "balance_loss_clip": 1.05866957,
      "balance_loss_mlp": 1.02047062,
      "epoch": 0.5184873444357603,
      "flos": 17597090568960.0,
      "grad_norm": 2.1140514751520154,
      "language_loss": 0.77169168,
      "learning_rate": 1.9774103025638675e-06,
      "loss": 0.79391229,
      "num_input_tokens_seen": 93124265,
      "step": 4312,
      "time_per_iteration": 3.581346273422241
    },
    {
      "auxiliary_loss_clip": 0.01179348,
      "auxiliary_loss_mlp": 0.01036381,
      "balance_loss_clip": 1.06092978,
      "balance_loss_mlp": 1.02684402,
      "epoch": 0.5186075873263993,
      "flos": 24936800883840.0,
      "grad_norm": 2.596005181785897,
      "language_loss": 0.7656014,
      "learning_rate": 1.9766313823393696e-06,
      "loss": 0.78775865,
      "num_input_tokens_seen": 93145130,
      "step": 4313,
      "time_per_iteration": 2.927459716796875
    },
    {
      "auxiliary_loss_clip": 0.01165843,
      "auxiliary_loss_mlp": 0.01036389,
      "balance_loss_clip": 1.05318427,
      "balance_loss_mlp": 1.02752566,
      "epoch": 0.5187278302170384,
      "flos": 15190106244480.0,
      "grad_norm": 2.67498282743556,
      "language_loss": 0.69279748,
      "learning_rate": 1.975852465659873e-06,
      "loss": 0.71481979,
      "num_input_tokens_seen": 93161110,
      "step": 4314,
      "time_per_iteration": 2.637559652328491
    },
    {
      "auxiliary_loss_clip": 0.01174212,
      "auxiliary_loss_mlp": 0.01035268,
      "balance_loss_clip": 1.0567596,
      "balance_loss_mlp": 1.02573681,
      "epoch": 0.5188480731076776,
      "flos": 25009412227200.0,
      "grad_norm": 4.475145777866261,
      "language_loss": 0.70398784,
      "learning_rate": 1.9750735526435377e-06,
      "loss": 0.72608262,
      "num_input_tokens_seen": 93178055,
      "step": 4315,
      "time_per_iteration": 2.6998109817504883
    },
    {
      "auxiliary_loss_clip": 0.01190453,
      "auxiliary_loss_mlp": 0.01035627,
      "balance_loss_clip": 1.05592418,
      "balance_loss_mlp": 1.02666247,
      "epoch": 0.5189683159983166,
      "flos": 24790141653120.0,
      "grad_norm": 4.925954020178933,
      "language_loss": 0.79068792,
      "learning_rate": 1.974294643408525e-06,
      "loss": 0.8129487,
      "num_input_tokens_seen": 93195850,
      "step": 4316,
      "time_per_iteration": 2.742828130722046
    },
    {
      "auxiliary_loss_clip": 0.01193805,
      "auxiliary_loss_mlp": 0.01033359,
      "balance_loss_clip": 1.05858791,
      "balance_loss_mlp": 1.02370286,
      "epoch": 0.5190885588889557,
      "flos": 24754266944640.0,
      "grad_norm": 4.370694025425269,
      "language_loss": 0.67412621,
      "learning_rate": 1.9735157380729947e-06,
      "loss": 0.69639784,
      "num_input_tokens_seen": 93216260,
      "step": 4317,
      "time_per_iteration": 2.682921886444092
    },
    {
      "auxiliary_loss_clip": 0.01155837,
      "auxiliary_loss_mlp": 0.01032099,
      "balance_loss_clip": 1.05190325,
      "balance_loss_mlp": 1.02321172,
      "epoch": 0.5192088017795948,
      "flos": 24712646060160.0,
      "grad_norm": 1.9104101529431081,
      "language_loss": 0.84398448,
      "learning_rate": 1.9727368367551053e-06,
      "loss": 0.86586386,
      "num_input_tokens_seen": 93234810,
      "step": 4318,
      "time_per_iteration": 2.672076940536499
    },
    {
      "auxiliary_loss_clip": 0.01175639,
      "auxiliary_loss_mlp": 0.01026342,
      "balance_loss_clip": 1.05571699,
      "balance_loss_mlp": 1.01755631,
      "epoch": 0.5193290446702339,
      "flos": 27229588894080.0,
      "grad_norm": 2.641456317078131,
      "language_loss": 0.68575025,
      "learning_rate": 1.9719579395730164e-06,
      "loss": 0.70777005,
      "num_input_tokens_seen": 93254185,
      "step": 4319,
      "time_per_iteration": 2.8213086128234863
    },
    {
      "auxiliary_loss_clip": 0.0119462,
      "auxiliary_loss_mlp": 0.0103899,
      "balance_loss_clip": 1.06176186,
      "balance_loss_mlp": 1.02989388,
      "epoch": 0.5194492875608729,
      "flos": 11473352392320.0,
      "grad_norm": 2.6537485500356746,
      "language_loss": 0.93721503,
      "learning_rate": 1.9711790466448854e-06,
      "loss": 0.95955122,
      "num_input_tokens_seen": 93268205,
      "step": 4320,
      "time_per_iteration": 2.5817127227783203
    },
    {
      "auxiliary_loss_clip": 0.01181746,
      "auxiliary_loss_mlp": 0.01032894,
      "balance_loss_clip": 1.05968916,
      "balance_loss_mlp": 1.02352428,
      "epoch": 0.5195695304515121,
      "flos": 20338906498560.0,
      "grad_norm": 4.688758104381833,
      "language_loss": 0.71273196,
      "learning_rate": 1.9704001580888704e-06,
      "loss": 0.73487836,
      "num_input_tokens_seen": 93286945,
      "step": 4321,
      "time_per_iteration": 2.703122138977051
    },
    {
      "auxiliary_loss_clip": 0.01171229,
      "auxiliary_loss_mlp": 0.01036436,
      "balance_loss_clip": 1.05343151,
      "balance_loss_mlp": 1.02801442,
      "epoch": 0.5196897733421512,
      "flos": 20048317470720.0,
      "grad_norm": 1.8010210327246834,
      "language_loss": 0.86877489,
      "learning_rate": 1.9696212740231283e-06,
      "loss": 0.89085156,
      "num_input_tokens_seen": 93305595,
      "step": 4322,
      "time_per_iteration": 2.648063898086548
    },
    {
      "auxiliary_loss_clip": 0.01177153,
      "auxiliary_loss_mlp": 0.01036472,
      "balance_loss_clip": 1.05477285,
      "balance_loss_mlp": 1.02707267,
      "epoch": 0.5198100162327902,
      "flos": 23805507058560.0,
      "grad_norm": 2.8648732781266304,
      "language_loss": 0.82415581,
      "learning_rate": 1.9688423945658146e-06,
      "loss": 0.84629202,
      "num_input_tokens_seen": 93326460,
      "step": 4323,
      "time_per_iteration": 2.6945557594299316
    },
    {
      "auxiliary_loss_clip": 0.01165711,
      "auxiliary_loss_mlp": 0.01036085,
      "balance_loss_clip": 1.05299509,
      "balance_loss_mlp": 1.02716804,
      "epoch": 0.5199302591234293,
      "flos": 24023951619840.0,
      "grad_norm": 2.077056178642022,
      "language_loss": 0.72396302,
      "learning_rate": 1.9680635198350845e-06,
      "loss": 0.74598098,
      "num_input_tokens_seen": 93346170,
      "step": 4324,
      "time_per_iteration": 2.715053081512451
    },
    {
      "auxiliary_loss_clip": 0.01196123,
      "auxiliary_loss_mlp": 0.01035725,
      "balance_loss_clip": 1.05875754,
      "balance_loss_mlp": 1.02657008,
      "epoch": 0.5200505020140684,
      "flos": 26359366095360.0,
      "grad_norm": 2.757644140825921,
      "language_loss": 0.7284658,
      "learning_rate": 1.967284649949093e-06,
      "loss": 0.75078428,
      "num_input_tokens_seen": 93365380,
      "step": 4325,
      "time_per_iteration": 2.7255945205688477
    },
    {
      "auxiliary_loss_clip": 0.01187002,
      "auxiliary_loss_mlp": 0.01032013,
      "balance_loss_clip": 1.05469251,
      "balance_loss_mlp": 1.0229826,
      "epoch": 0.5201707449047075,
      "flos": 39604262284800.0,
      "grad_norm": 6.476767856353214,
      "language_loss": 0.72907305,
      "learning_rate": 1.966505785025994e-06,
      "loss": 0.7512632,
      "num_input_tokens_seen": 93387285,
      "step": 4326,
      "time_per_iteration": 2.814375162124634
    },
    {
      "auxiliary_loss_clip": 0.01157395,
      "auxiliary_loss_mlp": 0.01037666,
      "balance_loss_clip": 1.05274808,
      "balance_loss_mlp": 1.02814162,
      "epoch": 0.5202909877953465,
      "flos": 53682788292480.0,
      "grad_norm": 1.965957004491182,
      "language_loss": 0.76097506,
      "learning_rate": 1.965726925183941e-06,
      "loss": 0.78292572,
      "num_input_tokens_seen": 93410390,
      "step": 4327,
      "time_per_iteration": 3.8708627223968506
    },
    {
      "auxiliary_loss_clip": 0.01192387,
      "auxiliary_loss_mlp": 0.01036097,
      "balance_loss_clip": 1.05880952,
      "balance_loss_mlp": 1.02645922,
      "epoch": 0.5204112306859857,
      "flos": 19537021324800.0,
      "grad_norm": 2.426893040534437,
      "language_loss": 0.84646463,
      "learning_rate": 1.964948070541087e-06,
      "loss": 0.8687495,
      "num_input_tokens_seen": 93429050,
      "step": 4328,
      "time_per_iteration": 2.6557960510253906
    },
    {
      "auxiliary_loss_clip": 0.01167983,
      "auxiliary_loss_mlp": 0.01033218,
      "balance_loss_clip": 1.05339026,
      "balance_loss_mlp": 1.0239141,
      "epoch": 0.5205314735766248,
      "flos": 15304697608320.0,
      "grad_norm": 2.212055342449076,
      "language_loss": 0.69615924,
      "learning_rate": 1.9641692212155816e-06,
      "loss": 0.71817124,
      "num_input_tokens_seen": 93446815,
      "step": 4329,
      "time_per_iteration": 2.655977964401245
    },
    {
      "auxiliary_loss_clip": 0.01190096,
      "auxiliary_loss_mlp": 0.01028435,
      "balance_loss_clip": 1.0572437,
      "balance_loss_mlp": 1.0198822,
      "epoch": 0.5206517164672638,
      "flos": 59263701160320.0,
      "grad_norm": 1.8765004319345389,
      "language_loss": 0.72574109,
      "learning_rate": 1.9633903773255777e-06,
      "loss": 0.74792635,
      "num_input_tokens_seen": 93469130,
      "step": 4330,
      "time_per_iteration": 3.8266119956970215
    },
    {
      "auxiliary_loss_clip": 0.01188812,
      "auxiliary_loss_mlp": 0.01036065,
      "balance_loss_clip": 1.05620527,
      "balance_loss_mlp": 1.02672482,
      "epoch": 0.520771959357903,
      "flos": 26871129118080.0,
      "grad_norm": 1.9249238422218906,
      "language_loss": 0.75300813,
      "learning_rate": 1.9626115389892237e-06,
      "loss": 0.77525699,
      "num_input_tokens_seen": 93489920,
      "step": 4331,
      "time_per_iteration": 2.695477247238159
    },
    {
      "auxiliary_loss_clip": 0.01175924,
      "auxiliary_loss_mlp": 0.01033818,
      "balance_loss_clip": 1.0581522,
      "balance_loss_mlp": 1.02471662,
      "epoch": 0.520892202248542,
      "flos": 26907075653760.0,
      "grad_norm": 1.8944427957730714,
      "language_loss": 0.85392827,
      "learning_rate": 1.96183270632467e-06,
      "loss": 0.87602568,
      "num_input_tokens_seen": 93509770,
      "step": 4332,
      "time_per_iteration": 2.713088035583496
    },
    {
      "auxiliary_loss_clip": 0.01181728,
      "auxiliary_loss_mlp": 0.01040424,
      "balance_loss_clip": 1.0565784,
      "balance_loss_mlp": 1.0309056,
      "epoch": 0.5210124451391811,
      "flos": 25849434666240.0,
      "grad_norm": 1.7446596579552915,
      "language_loss": 0.79465652,
      "learning_rate": 1.9610538794500644e-06,
      "loss": 0.81687808,
      "num_input_tokens_seen": 93529320,
      "step": 4333,
      "time_per_iteration": 2.7764453887939453
    },
    {
      "auxiliary_loss_clip": 0.01072668,
      "auxiliary_loss_mlp": 0.01006351,
      "balance_loss_clip": 1.00985479,
      "balance_loss_mlp": 1.00531352,
      "epoch": 0.5211326880298203,
      "flos": 70553804319360.0,
      "grad_norm": 0.7729605798428805,
      "language_loss": 0.59378827,
      "learning_rate": 1.9602750584835542e-06,
      "loss": 0.61457849,
      "num_input_tokens_seen": 93595255,
      "step": 4334,
      "time_per_iteration": 3.3215696811676025
    },
    {
      "auxiliary_loss_clip": 0.01156648,
      "auxiliary_loss_mlp": 0.0102472,
      "balance_loss_clip": 1.05403984,
      "balance_loss_mlp": 1.01635134,
      "epoch": 0.5212529309204593,
      "flos": 15628898787840.0,
      "grad_norm": 2.400946279179996,
      "language_loss": 0.82827312,
      "learning_rate": 1.959496243543286e-06,
      "loss": 0.85008681,
      "num_input_tokens_seen": 93613135,
      "step": 4335,
      "time_per_iteration": 2.837514638900757
    },
    {
      "auxiliary_loss_clip": 0.01193683,
      "auxiliary_loss_mlp": 0.01034736,
      "balance_loss_clip": 1.05972862,
      "balance_loss_mlp": 1.02522886,
      "epoch": 0.5213731738110984,
      "flos": 26242655829120.0,
      "grad_norm": 2.5103066786133623,
      "language_loss": 0.79770827,
      "learning_rate": 1.9587174347474057e-06,
      "loss": 0.81999254,
      "num_input_tokens_seen": 93629645,
      "step": 4336,
      "time_per_iteration": 2.6893045902252197
    },
    {
      "auxiliary_loss_clip": 0.01158293,
      "auxiliary_loss_mlp": 0.01033634,
      "balance_loss_clip": 1.05322003,
      "balance_loss_mlp": 1.02398968,
      "epoch": 0.5214934167017375,
      "flos": 19418407637760.0,
      "grad_norm": 3.2295479644295306,
      "language_loss": 0.81938422,
      "learning_rate": 1.9579386322140574e-06,
      "loss": 0.84130347,
      "num_input_tokens_seen": 93645325,
      "step": 4337,
      "time_per_iteration": 2.6706151962280273
    },
    {
      "auxiliary_loss_clip": 0.01182018,
      "auxiliary_loss_mlp": 0.01030676,
      "balance_loss_clip": 1.05539191,
      "balance_loss_mlp": 1.02168787,
      "epoch": 0.5216136595923766,
      "flos": 30955788023040.0,
      "grad_norm": 1.7988563044578365,
      "language_loss": 0.80967307,
      "learning_rate": 1.9571598360613854e-06,
      "loss": 0.83179998,
      "num_input_tokens_seen": 93668200,
      "step": 4338,
      "time_per_iteration": 3.675325393676758
    },
    {
      "auxiliary_loss_clip": 0.01187732,
      "auxiliary_loss_mlp": 0.01030774,
      "balance_loss_clip": 1.05530524,
      "balance_loss_mlp": 1.02185154,
      "epoch": 0.5217339024830157,
      "flos": 21945047143680.0,
      "grad_norm": 2.766121001081325,
      "language_loss": 0.69949704,
      "learning_rate": 1.956381046407532e-06,
      "loss": 0.72168207,
      "num_input_tokens_seen": 93688495,
      "step": 4339,
      "time_per_iteration": 2.611656904220581
    },
    {
      "auxiliary_loss_clip": 0.01189328,
      "auxiliary_loss_mlp": 0.01036056,
      "balance_loss_clip": 1.05842829,
      "balance_loss_mlp": 1.02698994,
      "epoch": 0.5218541453736548,
      "flos": 20923209037440.0,
      "grad_norm": 1.8552775915477397,
      "language_loss": 0.86749589,
      "learning_rate": 1.9556022633706394e-06,
      "loss": 0.88974977,
      "num_input_tokens_seen": 93707285,
      "step": 4340,
      "time_per_iteration": 2.708678960800171
    },
    {
      "auxiliary_loss_clip": 0.01164947,
      "auxiliary_loss_mlp": 0.01031428,
      "balance_loss_clip": 1.05415905,
      "balance_loss_mlp": 1.02264237,
      "epoch": 0.5219743882642939,
      "flos": 23951663498880.0,
      "grad_norm": 1.7045296434393662,
      "language_loss": 0.8004837,
      "learning_rate": 1.954823487068848e-06,
      "loss": 0.82244748,
      "num_input_tokens_seen": 93727495,
      "step": 4341,
      "time_per_iteration": 2.771872043609619
    },
    {
      "auxiliary_loss_clip": 0.01188539,
      "auxiliary_loss_mlp": 0.01032498,
      "balance_loss_clip": 1.05755043,
      "balance_loss_mlp": 1.02381337,
      "epoch": 0.5220946311549329,
      "flos": 28799280213120.0,
      "grad_norm": 1.751292642189208,
      "language_loss": 0.81148517,
      "learning_rate": 1.9540447176202976e-06,
      "loss": 0.83369559,
      "num_input_tokens_seen": 93748740,
      "step": 4342,
      "time_per_iteration": 2.7099835872650146
    },
    {
      "auxiliary_loss_clip": 0.01071173,
      "auxiliary_loss_mlp": 0.01012668,
      "balance_loss_clip": 1.02500844,
      "balance_loss_mlp": 1.01172066,
      "epoch": 0.5222148740455721,
      "flos": 67189369017600.0,
      "grad_norm": 0.8769681221149522,
      "language_loss": 0.60647488,
      "learning_rate": 1.9532659551431272e-06,
      "loss": 0.62731332,
      "num_input_tokens_seen": 93815770,
      "step": 4343,
      "time_per_iteration": 3.4167416095733643
    },
    {
      "auxiliary_loss_clip": 0.0117342,
      "auxiliary_loss_mlp": 0.01031908,
      "balance_loss_clip": 1.05456924,
      "balance_loss_mlp": 1.02353942,
      "epoch": 0.5223351169362112,
      "flos": 61856164339200.0,
      "grad_norm": 1.5666606939222467,
      "language_loss": 0.67765141,
      "learning_rate": 1.9524871997554744e-06,
      "loss": 0.69970471,
      "num_input_tokens_seen": 93843530,
      "step": 4344,
      "time_per_iteration": 3.044875144958496
    },
    {
      "auxiliary_loss_clip": 0.01180868,
      "auxiliary_loss_mlp": 0.01028755,
      "balance_loss_clip": 1.05482185,
      "balance_loss_mlp": 1.01915324,
      "epoch": 0.5224553598268502,
      "flos": 14647388676480.0,
      "grad_norm": 3.6464356209413427,
      "language_loss": 0.80694455,
      "learning_rate": 1.951708451575475e-06,
      "loss": 0.82904083,
      "num_input_tokens_seen": 93860595,
      "step": 4345,
      "time_per_iteration": 2.662555456161499
    },
    {
      "auxiliary_loss_clip": 0.01174856,
      "auxiliary_loss_mlp": 0.01029837,
      "balance_loss_clip": 1.05710185,
      "balance_loss_mlp": 1.02015078,
      "epoch": 0.5225756027174894,
      "flos": 14826043946880.0,
      "grad_norm": 2.1941928855706583,
      "language_loss": 0.81739277,
      "learning_rate": 1.9509297107212657e-06,
      "loss": 0.83943969,
      "num_input_tokens_seen": 93877365,
      "step": 4346,
      "time_per_iteration": 2.619264602661133
    },
    {
      "auxiliary_loss_clip": 0.01189133,
      "auxiliary_loss_mlp": 0.00762946,
      "balance_loss_clip": 1.05656147,
      "balance_loss_mlp": 1.00138783,
      "epoch": 0.5226958456081284,
      "flos": 23512009029120.0,
      "grad_norm": 1.7970875443355754,
      "language_loss": 0.79267281,
      "learning_rate": 1.95015097731098e-06,
      "loss": 0.81219363,
      "num_input_tokens_seen": 93896855,
      "step": 4347,
      "time_per_iteration": 2.695678472518921
    },
    {
      "auxiliary_loss_clip": 0.01179552,
      "auxiliary_loss_mlp": 0.01030616,
      "balance_loss_clip": 1.05595493,
      "balance_loss_mlp": 1.02081728,
      "epoch": 0.5228160884987675,
      "flos": 19062928690560.0,
      "grad_norm": 3.4572805222369904,
      "language_loss": 0.8233676,
      "learning_rate": 1.949372251462751e-06,
      "loss": 0.84546924,
      "num_input_tokens_seen": 93914270,
      "step": 4348,
      "time_per_iteration": 2.667877197265625
    },
    {
      "auxiliary_loss_clip": 0.0117174,
      "auxiliary_loss_mlp": 0.01028389,
      "balance_loss_clip": 1.05544758,
      "balance_loss_mlp": 1.0202713,
      "epoch": 0.5229363313894067,
      "flos": 21063224252160.0,
      "grad_norm": 2.074671339022501,
      "language_loss": 0.83054143,
      "learning_rate": 1.9485935332947124e-06,
      "loss": 0.85254276,
      "num_input_tokens_seen": 93932180,
      "step": 4349,
      "time_per_iteration": 2.6752254962921143
    },
    {
      "auxiliary_loss_clip": 0.01171355,
      "auxiliary_loss_mlp": 0.01029231,
      "balance_loss_clip": 1.05378425,
      "balance_loss_mlp": 1.02141428,
      "epoch": 0.5230565742800457,
      "flos": 14830389492480.0,
      "grad_norm": 2.3995448721518873,
      "language_loss": 0.84090871,
      "learning_rate": 1.947814822924993e-06,
      "loss": 0.86291456,
      "num_input_tokens_seen": 93949690,
      "step": 4350,
      "time_per_iteration": 2.6737751960754395
    },
    {
      "auxiliary_loss_clip": 0.01190878,
      "auxiliary_loss_mlp": 0.01038889,
      "balance_loss_clip": 1.05817175,
      "balance_loss_mlp": 1.03024006,
      "epoch": 0.5231768171706848,
      "flos": 25813021253760.0,
      "grad_norm": 2.2812467044781046,
      "language_loss": 0.83081341,
      "learning_rate": 1.9470361204717236e-06,
      "loss": 0.85311109,
      "num_input_tokens_seen": 93968830,
      "step": 4351,
      "time_per_iteration": 2.638747215270996
    },
    {
      "auxiliary_loss_clip": 0.01174274,
      "auxiliary_loss_mlp": 0.0076323,
      "balance_loss_clip": 1.05630803,
      "balance_loss_mlp": 1.00136483,
      "epoch": 0.5232970600613239,
      "flos": 22743807834240.0,
      "grad_norm": 2.744058378382618,
      "language_loss": 0.81051129,
      "learning_rate": 1.9462574260530326e-06,
      "loss": 0.82988632,
      "num_input_tokens_seen": 93989110,
      "step": 4352,
      "time_per_iteration": 2.721604347229004
    },
    {
      "auxiliary_loss_clip": 0.01188831,
      "auxiliary_loss_mlp": 0.01027345,
      "balance_loss_clip": 1.05675054,
      "balance_loss_mlp": 1.01866651,
      "epoch": 0.523417302951963,
      "flos": 17310703432320.0,
      "grad_norm": 2.0829910212503497,
      "language_loss": 0.81186604,
      "learning_rate": 1.9454787397870472e-06,
      "loss": 0.83402777,
      "num_input_tokens_seen": 94006430,
      "step": 4353,
      "time_per_iteration": 3.579228162765503
    },
    {
      "auxiliary_loss_clip": 0.01173233,
      "auxiliary_loss_mlp": 0.01032894,
      "balance_loss_clip": 1.05536413,
      "balance_loss_mlp": 1.02391768,
      "epoch": 0.523537545842602,
      "flos": 18551740285440.0,
      "grad_norm": 2.0203174845991425,
      "language_loss": 0.7201643,
      "learning_rate": 1.944700061791894e-06,
      "loss": 0.74222553,
      "num_input_tokens_seen": 94024825,
      "step": 4354,
      "time_per_iteration": 2.67828631401062
    },
    {
      "auxiliary_loss_clip": 0.0118969,
      "auxiliary_loss_mlp": 0.01029867,
      "balance_loss_clip": 1.05588841,
      "balance_loss_mlp": 1.02103412,
      "epoch": 0.5236577887332411,
      "flos": 19719267955200.0,
      "grad_norm": 2.592060164326165,
      "language_loss": 0.65433216,
      "learning_rate": 1.943921392185698e-06,
      "loss": 0.67652774,
      "num_input_tokens_seen": 94043450,
      "step": 4355,
      "time_per_iteration": 3.538728713989258
    },
    {
      "auxiliary_loss_clip": 0.01165406,
      "auxiliary_loss_mlp": 0.01027155,
      "balance_loss_clip": 1.05601025,
      "balance_loss_mlp": 1.01912642,
      "epoch": 0.5237780316238803,
      "flos": 23550218121600.0,
      "grad_norm": 1.9937818458701984,
      "language_loss": 0.77353495,
      "learning_rate": 1.9431427310865814e-06,
      "loss": 0.79546058,
      "num_input_tokens_seen": 94063055,
      "step": 4356,
      "time_per_iteration": 3.6481380462646484
    },
    {
      "auxiliary_loss_clip": 0.011716,
      "auxiliary_loss_mlp": 0.01033799,
      "balance_loss_clip": 1.05245864,
      "balance_loss_mlp": 1.02555585,
      "epoch": 0.5238982745145193,
      "flos": 22491894775680.0,
      "grad_norm": 1.9195622939646972,
      "language_loss": 0.78612489,
      "learning_rate": 1.942364078612667e-06,
      "loss": 0.8081789,
      "num_input_tokens_seen": 94081785,
      "step": 4357,
      "time_per_iteration": 2.685590982437134
    },
    {
      "auxiliary_loss_clip": 0.01143685,
      "auxiliary_loss_mlp": 0.01027967,
      "balance_loss_clip": 1.05004144,
      "balance_loss_mlp": 1.01920485,
      "epoch": 0.5240185174051584,
      "flos": 27088927234560.0,
      "grad_norm": 2.0898563613645473,
      "language_loss": 0.75797391,
      "learning_rate": 1.9415854348820765e-06,
      "loss": 0.77969038,
      "num_input_tokens_seen": 94101635,
      "step": 4358,
      "time_per_iteration": 2.782945394515991
    },
    {
      "auxiliary_loss_clip": 0.01164672,
      "auxiliary_loss_mlp": 0.01028425,
      "balance_loss_clip": 1.05403018,
      "balance_loss_mlp": 1.02014017,
      "epoch": 0.5241387602957975,
      "flos": 22674680110080.0,
      "grad_norm": 4.680090214814519,
      "language_loss": 0.69163918,
      "learning_rate": 1.940806800012929e-06,
      "loss": 0.71357012,
      "num_input_tokens_seen": 94121705,
      "step": 4359,
      "time_per_iteration": 2.7635223865509033
    },
    {
      "auxiliary_loss_clip": 0.0118411,
      "auxiliary_loss_mlp": 0.01027407,
      "balance_loss_clip": 1.05837727,
      "balance_loss_mlp": 1.01853824,
      "epoch": 0.5242590031864366,
      "flos": 40553453134080.0,
      "grad_norm": 2.3655392475365002,
      "language_loss": 0.63978142,
      "learning_rate": 1.9400281741233432e-06,
      "loss": 0.66189659,
      "num_input_tokens_seen": 94146595,
      "step": 4360,
      "time_per_iteration": 2.858527660369873
    },
    {
      "auxiliary_loss_clip": 0.01061622,
      "auxiliary_loss_mlp": 0.0101296,
      "balance_loss_clip": 1.00991738,
      "balance_loss_mlp": 1.01194692,
      "epoch": 0.5243792460770756,
      "flos": 66676313105280.0,
      "grad_norm": 0.6854396585378845,
      "language_loss": 0.5243271,
      "learning_rate": 1.939249557331435e-06,
      "loss": 0.54507291,
      "num_input_tokens_seen": 94212410,
      "step": 4361,
      "time_per_iteration": 3.262873649597168
    },
    {
      "auxiliary_loss_clip": 0.01189573,
      "auxiliary_loss_mlp": 0.01026412,
      "balance_loss_clip": 1.05752325,
      "balance_loss_mlp": 1.01847255,
      "epoch": 0.5244994889677148,
      "flos": 28183663992960.0,
      "grad_norm": 2.1760068449108587,
      "language_loss": 0.73414242,
      "learning_rate": 1.938470949755321e-06,
      "loss": 0.75630224,
      "num_input_tokens_seen": 94232290,
      "step": 4362,
      "time_per_iteration": 2.730630874633789
    },
    {
      "auxiliary_loss_clip": 0.01072932,
      "auxiliary_loss_mlp": 0.01006101,
      "balance_loss_clip": 1.00985372,
      "balance_loss_mlp": 1.00505149,
      "epoch": 0.5246197318583539,
      "flos": 65950379239680.0,
      "grad_norm": 0.8045375163985079,
      "language_loss": 0.55567265,
      "learning_rate": 1.937692351513115e-06,
      "loss": 0.57646292,
      "num_input_tokens_seen": 94291285,
      "step": 4363,
      "time_per_iteration": 3.1414361000061035
    },
    {
      "auxiliary_loss_clip": 0.01191872,
      "auxiliary_loss_mlp": 0.01028886,
      "balance_loss_clip": 1.05727589,
      "balance_loss_mlp": 1.02023125,
      "epoch": 0.5247399747489929,
      "flos": 21033490769280.0,
      "grad_norm": 1.8433738191636857,
      "language_loss": 0.80689484,
      "learning_rate": 1.9369137627229297e-06,
      "loss": 0.8291024,
      "num_input_tokens_seen": 94309685,
      "step": 4364,
      "time_per_iteration": 3.6039392948150635
    },
    {
      "auxiliary_loss_clip": 0.01174084,
      "auxiliary_loss_mlp": 0.01032012,
      "balance_loss_clip": 1.05560982,
      "balance_loss_mlp": 1.02341652,
      "epoch": 0.5248602176396321,
      "flos": 19025940660480.0,
      "grad_norm": 2.257437054194611,
      "language_loss": 0.88927084,
      "learning_rate": 1.936135183502877e-06,
      "loss": 0.91133177,
      "num_input_tokens_seen": 94326985,
      "step": 4365,
      "time_per_iteration": 2.6932754516601562
    },
    {
      "auxiliary_loss_clip": 0.01167639,
      "auxiliary_loss_mlp": 0.01037354,
      "balance_loss_clip": 1.05356574,
      "balance_loss_mlp": 1.02864563,
      "epoch": 0.5249804605302711,
      "flos": 22200084685440.0,
      "grad_norm": 3.0324300445817167,
      "language_loss": 0.80403292,
      "learning_rate": 1.935356613971066e-06,
      "loss": 0.82608283,
      "num_input_tokens_seen": 94347645,
      "step": 4366,
      "time_per_iteration": 2.6659066677093506
    },
    {
      "auxiliary_loss_clip": 0.01187459,
      "auxiliary_loss_mlp": 0.01032174,
      "balance_loss_clip": 1.05626702,
      "balance_loss_mlp": 1.02402627,
      "epoch": 0.5251007034209102,
      "flos": 23805686626560.0,
      "grad_norm": 1.796012643541197,
      "language_loss": 0.76674581,
      "learning_rate": 1.9345780542456047e-06,
      "loss": 0.7889421,
      "num_input_tokens_seen": 94367020,
      "step": 4367,
      "time_per_iteration": 2.6828770637512207
    },
    {
      "auxiliary_loss_clip": 0.01185694,
      "auxiliary_loss_mlp": 0.01026979,
      "balance_loss_clip": 1.05639088,
      "balance_loss_mlp": 1.01850343,
      "epoch": 0.5252209463115494,
      "flos": 23294605962240.0,
      "grad_norm": 2.4082303487006103,
      "language_loss": 0.71334541,
      "learning_rate": 1.9337995044446007e-06,
      "loss": 0.73547214,
      "num_input_tokens_seen": 94385860,
      "step": 4368,
      "time_per_iteration": 2.665112018585205
    },
    {
      "auxiliary_loss_clip": 0.0118852,
      "auxiliary_loss_mlp": 0.01034292,
      "balance_loss_clip": 1.05624819,
      "balance_loss_mlp": 1.02637029,
      "epoch": 0.5253411892021884,
      "flos": 19828687760640.0,
      "grad_norm": 2.3480945203456502,
      "language_loss": 0.80408156,
      "learning_rate": 1.9330209646861596e-06,
      "loss": 0.82630968,
      "num_input_tokens_seen": 94405010,
      "step": 4369,
      "time_per_iteration": 2.675260305404663
    },
    {
      "auxiliary_loss_clip": 0.01172349,
      "auxiliary_loss_mlp": 0.01028235,
      "balance_loss_clip": 1.05217648,
      "balance_loss_mlp": 1.01966953,
      "epoch": 0.5254614320928275,
      "flos": 24133730561280.0,
      "grad_norm": 1.998000505934656,
      "language_loss": 0.78101438,
      "learning_rate": 1.9322424350883843e-06,
      "loss": 0.80302024,
      "num_input_tokens_seen": 94426845,
      "step": 4370,
      "time_per_iteration": 2.649041175842285
    },
    {
      "auxiliary_loss_clip": 0.01188751,
      "auxiliary_loss_mlp": 0.01031542,
      "balance_loss_clip": 1.05786991,
      "balance_loss_mlp": 1.02325702,
      "epoch": 0.5255816749834666,
      "flos": 24644954880000.0,
      "grad_norm": 2.2411314186838505,
      "language_loss": 0.78814173,
      "learning_rate": 1.931463915769379e-06,
      "loss": 0.8103447,
      "num_input_tokens_seen": 94446960,
      "step": 4371,
      "time_per_iteration": 2.6920459270477295
    },
    {
      "auxiliary_loss_clip": 0.01171334,
      "auxiliary_loss_mlp": 0.01029431,
      "balance_loss_clip": 1.05725145,
      "balance_loss_mlp": 1.02076435,
      "epoch": 0.5257019178741057,
      "flos": 14136595320960.0,
      "grad_norm": 2.3790250454997723,
      "language_loss": 0.7384516,
      "learning_rate": 1.930685406847242e-06,
      "loss": 0.76045918,
      "num_input_tokens_seen": 94461535,
      "step": 4372,
      "time_per_iteration": 2.700415849685669
    },
    {
      "auxiliary_loss_clip": 0.01174576,
      "auxiliary_loss_mlp": 0.01028832,
      "balance_loss_clip": 1.05313849,
      "balance_loss_mlp": 1.02068365,
      "epoch": 0.5258221607647448,
      "flos": 23548961145600.0,
      "grad_norm": 1.638170763010689,
      "language_loss": 0.81970406,
      "learning_rate": 1.9299069084400734e-06,
      "loss": 0.8417381,
      "num_input_tokens_seen": 94482395,
      "step": 4373,
      "time_per_iteration": 2.7137160301208496
    },
    {
      "auxiliary_loss_clip": 0.01168619,
      "auxiliary_loss_mlp": 0.01028796,
      "balance_loss_clip": 1.05458462,
      "balance_loss_mlp": 1.01930714,
      "epoch": 0.5259424036553839,
      "flos": 24966103403520.0,
      "grad_norm": 2.5859540230460714,
      "language_loss": 0.69651258,
      "learning_rate": 1.9291284206659717e-06,
      "loss": 0.71848679,
      "num_input_tokens_seen": 94500580,
      "step": 4374,
      "time_per_iteration": 2.6463027000427246
    },
    {
      "auxiliary_loss_clip": 0.01189784,
      "auxiliary_loss_mlp": 0.01035561,
      "balance_loss_clip": 1.05705214,
      "balance_loss_mlp": 1.0267036,
      "epoch": 0.526062646546023,
      "flos": 28763908295040.0,
      "grad_norm": 2.032759992170467,
      "language_loss": 0.71677345,
      "learning_rate": 1.928349943643032e-06,
      "loss": 0.7390269,
      "num_input_tokens_seen": 94519680,
      "step": 4375,
      "time_per_iteration": 2.699676990509033
    },
    {
      "auxiliary_loss_clip": 0.01192238,
      "auxiliary_loss_mlp": 0.0103387,
      "balance_loss_clip": 1.06083703,
      "balance_loss_mlp": 1.02529931,
      "epoch": 0.526182889436662,
      "flos": 22821375254400.0,
      "grad_norm": 1.8280236919189676,
      "language_loss": 0.8186062,
      "learning_rate": 1.9275714774893493e-06,
      "loss": 0.84086722,
      "num_input_tokens_seen": 94539135,
      "step": 4376,
      "time_per_iteration": 2.6285812854766846
    },
    {
      "auxiliary_loss_clip": 0.01190026,
      "auxiliary_loss_mlp": 0.00762789,
      "balance_loss_clip": 1.05671942,
      "balance_loss_mlp": 1.00131679,
      "epoch": 0.5263031323273012,
      "flos": 22929466256640.0,
      "grad_norm": 3.204366442366217,
      "language_loss": 0.73287034,
      "learning_rate": 1.9267930223230154e-06,
      "loss": 0.75239849,
      "num_input_tokens_seen": 94557610,
      "step": 4377,
      "time_per_iteration": 2.6911983489990234
    },
    {
      "auxiliary_loss_clip": 0.01164798,
      "auxiliary_loss_mlp": 0.01035193,
      "balance_loss_clip": 1.05195951,
      "balance_loss_mlp": 1.02717662,
      "epoch": 0.5264233752179402,
      "flos": 17748634049280.0,
      "grad_norm": 2.6940380759258806,
      "language_loss": 0.78578091,
      "learning_rate": 1.9260145782621224e-06,
      "loss": 0.80778074,
      "num_input_tokens_seen": 94575390,
      "step": 4378,
      "time_per_iteration": 2.7485413551330566
    },
    {
      "auxiliary_loss_clip": 0.01190626,
      "auxiliary_loss_mlp": 0.01031861,
      "balance_loss_clip": 1.0575937,
      "balance_loss_mlp": 1.02370751,
      "epoch": 0.5265436181085793,
      "flos": 24421626069120.0,
      "grad_norm": 1.8577756968150798,
      "language_loss": 0.88204801,
      "learning_rate": 1.925236145424758e-06,
      "loss": 0.90427291,
      "num_input_tokens_seen": 94594210,
      "step": 4379,
      "time_per_iteration": 2.665095090866089
    },
    {
      "auxiliary_loss_clip": 0.0107297,
      "auxiliary_loss_mlp": 0.01007182,
      "balance_loss_clip": 1.0102272,
      "balance_loss_mlp": 1.00608492,
      "epoch": 0.5266638609992185,
      "flos": 69207298156800.0,
      "grad_norm": 0.6977480407517135,
      "language_loss": 0.57533371,
      "learning_rate": 1.924457723929012e-06,
      "loss": 0.59613526,
      "num_input_tokens_seen": 94665020,
      "step": 4380,
      "time_per_iteration": 4.290636777877808
    },
    {
      "auxiliary_loss_clip": 0.0118913,
      "auxiliary_loss_mlp": 0.01027201,
      "balance_loss_clip": 1.05770266,
      "balance_loss_mlp": 1.0188508,
      "epoch": 0.5267841038898575,
      "flos": 20738699850240.0,
      "grad_norm": 1.6406031509528427,
      "language_loss": 0.82889819,
      "learning_rate": 1.9236793138929685e-06,
      "loss": 0.85106146,
      "num_input_tokens_seen": 94684290,
      "step": 4381,
      "time_per_iteration": 3.6188971996307373
    },
    {
      "auxiliary_loss_clip": 0.01190348,
      "auxiliary_loss_mlp": 0.01023652,
      "balance_loss_clip": 1.0572089,
      "balance_loss_mlp": 1.0151521,
      "epoch": 0.5269043467804966,
      "flos": 17234392988160.0,
      "grad_norm": 2.2530785500037545,
      "language_loss": 0.81564152,
      "learning_rate": 1.9229009154347133e-06,
      "loss": 0.83778155,
      "num_input_tokens_seen": 94701880,
      "step": 4382,
      "time_per_iteration": 3.529649496078491
    },
    {
      "auxiliary_loss_clip": 0.01189564,
      "auxiliary_loss_mlp": 0.01028466,
      "balance_loss_clip": 1.05826986,
      "balance_loss_mlp": 1.02026391,
      "epoch": 0.5270245896711357,
      "flos": 18223157646720.0,
      "grad_norm": 2.6649917556695195,
      "language_loss": 0.80787241,
      "learning_rate": 1.922122528672327e-06,
      "loss": 0.83005273,
      "num_input_tokens_seen": 94720545,
      "step": 4383,
      "time_per_iteration": 2.6781983375549316
    },
    {
      "auxiliary_loss_clip": 0.01186756,
      "auxiliary_loss_mlp": 0.01030635,
      "balance_loss_clip": 1.05672348,
      "balance_loss_mlp": 1.02256465,
      "epoch": 0.5271448325617748,
      "flos": 21287558643840.0,
      "grad_norm": 6.204785363276911,
      "language_loss": 0.78270793,
      "learning_rate": 1.9213441537238914e-06,
      "loss": 0.80488181,
      "num_input_tokens_seen": 94737420,
      "step": 4384,
      "time_per_iteration": 2.600571870803833
    },
    {
      "auxiliary_loss_clip": 0.01062061,
      "auxiliary_loss_mlp": 0.01001848,
      "balance_loss_clip": 1.01153064,
      "balance_loss_mlp": 1.00072765,
      "epoch": 0.5272650754524139,
      "flos": 65495497403520.0,
      "grad_norm": 0.9547327124591716,
      "language_loss": 0.5730778,
      "learning_rate": 1.920565790707485e-06,
      "loss": 0.59371686,
      "num_input_tokens_seen": 94802810,
      "step": 4385,
      "time_per_iteration": 3.359647512435913
    },
    {
      "auxiliary_loss_clip": 0.01170944,
      "auxiliary_loss_mlp": 0.01033171,
      "balance_loss_clip": 1.05481458,
      "balance_loss_mlp": 1.02433157,
      "epoch": 0.527385318343053,
      "flos": 19676426008320.0,
      "grad_norm": 2.226162949493698,
      "language_loss": 0.65930223,
      "learning_rate": 1.9197874397411853e-06,
      "loss": 0.68134338,
      "num_input_tokens_seen": 94819440,
      "step": 4386,
      "time_per_iteration": 2.6460371017456055
    },
    {
      "auxiliary_loss_clip": 0.01174949,
      "auxiliary_loss_mlp": 0.00763098,
      "balance_loss_clip": 1.05713701,
      "balance_loss_mlp": 1.0012778,
      "epoch": 0.5275055612336921,
      "flos": 12712018947840.0,
      "grad_norm": 4.5383122398785005,
      "language_loss": 0.66530985,
      "learning_rate": 1.919009100943067e-06,
      "loss": 0.68469036,
      "num_input_tokens_seen": 94835130,
      "step": 4387,
      "time_per_iteration": 2.720395088195801
    },
    {
      "auxiliary_loss_clip": 0.01177357,
      "auxiliary_loss_mlp": 0.01035625,
      "balance_loss_clip": 1.05702245,
      "balance_loss_mlp": 1.02650547,
      "epoch": 0.5276258041243311,
      "flos": 17749029098880.0,
      "grad_norm": 4.771691805261118,
      "language_loss": 0.65919745,
      "learning_rate": 1.9182307744312043e-06,
      "loss": 0.68132728,
      "num_input_tokens_seen": 94852235,
      "step": 4388,
      "time_per_iteration": 2.736330986022949
    },
    {
      "auxiliary_loss_clip": 0.01173194,
      "auxiliary_loss_mlp": 0.01029928,
      "balance_loss_clip": 1.05618715,
      "balance_loss_mlp": 1.02098703,
      "epoch": 0.5277460470149702,
      "flos": 22710447077760.0,
      "grad_norm": 1.8043200624916438,
      "language_loss": 0.76764059,
      "learning_rate": 1.9174524603236676e-06,
      "loss": 0.78967184,
      "num_input_tokens_seen": 94871185,
      "step": 4389,
      "time_per_iteration": 2.6409518718719482
    },
    {
      "auxiliary_loss_clip": 0.01162604,
      "auxiliary_loss_mlp": 0.01029025,
      "balance_loss_clip": 1.05146551,
      "balance_loss_mlp": 1.02012062,
      "epoch": 0.5278662899056094,
      "flos": 19902699734400.0,
      "grad_norm": 5.486486303464693,
      "language_loss": 0.76856726,
      "learning_rate": 1.916674158738527e-06,
      "loss": 0.79048353,
      "num_input_tokens_seen": 94890090,
      "step": 4390,
      "time_per_iteration": 3.757918119430542
    },
    {
      "auxiliary_loss_clip": 0.01161577,
      "auxiliary_loss_mlp": 0.01034644,
      "balance_loss_clip": 1.05070341,
      "balance_loss_mlp": 1.02538729,
      "epoch": 0.5279865327962484,
      "flos": 18005215875840.0,
      "grad_norm": 2.1578103131161863,
      "language_loss": 0.60411453,
      "learning_rate": 1.9158958697938506e-06,
      "loss": 0.6260767,
      "num_input_tokens_seen": 94908470,
      "step": 4391,
      "time_per_iteration": 2.7172348499298096
    },
    {
      "auxiliary_loss_clip": 0.01189939,
      "auxiliary_loss_mlp": 0.01029704,
      "balance_loss_clip": 1.05839396,
      "balance_loss_mlp": 1.02102554,
      "epoch": 0.5281067756868875,
      "flos": 15924443892480.0,
      "grad_norm": 2.945799710229932,
      "language_loss": 0.863702,
      "learning_rate": 1.9151175936077032e-06,
      "loss": 0.88589841,
      "num_input_tokens_seen": 94923440,
      "step": 4392,
      "time_per_iteration": 2.599797248840332
    },
    {
      "auxiliary_loss_clip": 0.01186417,
      "auxiliary_loss_mlp": 0.01027043,
      "balance_loss_clip": 1.05541646,
      "balance_loss_mlp": 1.01905632,
      "epoch": 0.5282270185775266,
      "flos": 19426488197760.0,
      "grad_norm": 2.133786785991138,
      "language_loss": 0.79346234,
      "learning_rate": 1.9143393302981507e-06,
      "loss": 0.81559694,
      "num_input_tokens_seen": 94941125,
      "step": 4393,
      "time_per_iteration": 2.7584965229034424
    },
    {
      "auxiliary_loss_clip": 0.01190496,
      "auxiliary_loss_mlp": 0.01033841,
      "balance_loss_clip": 1.05824864,
      "balance_loss_mlp": 1.02573133,
      "epoch": 0.5283472614681657,
      "flos": 16399613934720.0,
      "grad_norm": 2.003265417944864,
      "language_loss": 0.83425188,
      "learning_rate": 1.913561079983252e-06,
      "loss": 0.85649526,
      "num_input_tokens_seen": 94959950,
      "step": 4394,
      "time_per_iteration": 2.6012637615203857
    },
    {
      "auxiliary_loss_clip": 0.01176159,
      "auxiliary_loss_mlp": 0.01037564,
      "balance_loss_clip": 1.05541694,
      "balance_loss_mlp": 1.02795589,
      "epoch": 0.5284675043588047,
      "flos": 26760524163840.0,
      "grad_norm": 2.400358846518709,
      "language_loss": 0.75063407,
      "learning_rate": 1.9127828427810693e-06,
      "loss": 0.77277136,
      "num_input_tokens_seen": 94980515,
      "step": 4395,
      "time_per_iteration": 2.736443519592285
    },
    {
      "auxiliary_loss_clip": 0.0118019,
      "auxiliary_loss_mlp": 0.01028645,
      "balance_loss_clip": 1.05812931,
      "balance_loss_mlp": 1.0197103,
      "epoch": 0.5285877472494439,
      "flos": 19899898473600.0,
      "grad_norm": 2.0664975342747693,
      "language_loss": 0.80722952,
      "learning_rate": 1.9120046188096607e-06,
      "loss": 0.82931787,
      "num_input_tokens_seen": 94998560,
      "step": 4396,
      "time_per_iteration": 2.626624822616577
    },
    {
      "auxiliary_loss_clip": 0.01173444,
      "auxiliary_loss_mlp": 0.01029455,
      "balance_loss_clip": 1.0529933,
      "balance_loss_mlp": 1.02128291,
      "epoch": 0.528707990140083,
      "flos": 20011257613440.0,
      "grad_norm": 2.780516117646687,
      "language_loss": 0.73940223,
      "learning_rate": 1.9112264081870804e-06,
      "loss": 0.76143122,
      "num_input_tokens_seen": 95016950,
      "step": 4397,
      "time_per_iteration": 2.736741065979004
    },
    {
      "auxiliary_loss_clip": 0.01191536,
      "auxiliary_loss_mlp": 0.01028666,
      "balance_loss_clip": 1.05504489,
      "balance_loss_mlp": 1.01944542,
      "epoch": 0.528828233030722,
      "flos": 20667956014080.0,
      "grad_norm": 2.5822381304382556,
      "language_loss": 0.75788283,
      "learning_rate": 1.9104482110313843e-06,
      "loss": 0.78008485,
      "num_input_tokens_seen": 95036540,
      "step": 4398,
      "time_per_iteration": 2.637944459915161
    },
    {
      "auxiliary_loss_clip": 0.0115333,
      "auxiliary_loss_mlp": 0.0103007,
      "balance_loss_clip": 1.0512569,
      "balance_loss_mlp": 1.0215224,
      "epoch": 0.5289484759213612,
      "flos": 25192448956800.0,
      "grad_norm": 2.020806265193084,
      "language_loss": 0.74510205,
      "learning_rate": 1.909670027460623e-06,
      "loss": 0.76693606,
      "num_input_tokens_seen": 95053840,
      "step": 4399,
      "time_per_iteration": 2.74651837348938
    },
    {
      "auxiliary_loss_clip": 0.01193092,
      "auxiliary_loss_mlp": 0.010331,
      "balance_loss_clip": 1.06000555,
      "balance_loss_mlp": 1.0242722,
      "epoch": 0.5290687188120002,
      "flos": 31139255715840.0,
      "grad_norm": 4.0592731974333205,
      "language_loss": 0.72081959,
      "learning_rate": 1.908891857592847e-06,
      "loss": 0.74308151,
      "num_input_tokens_seen": 95074910,
      "step": 4400,
      "time_per_iteration": 2.699597120285034
    },
    {
      "auxiliary_loss_clip": 0.01188572,
      "auxiliary_loss_mlp": 0.01036383,
      "balance_loss_clip": 1.055475,
      "balance_loss_mlp": 1.02749014,
      "epoch": 0.5291889617026393,
      "flos": 20119851406080.0,
      "grad_norm": 2.3285925084578607,
      "language_loss": 0.898996,
      "learning_rate": 1.9081137015461034e-06,
      "loss": 0.92124546,
      "num_input_tokens_seen": 95090985,
      "step": 4401,
      "time_per_iteration": 2.6972250938415527
    },
    {
      "auxiliary_loss_clip": 0.0115623,
      "auxiliary_loss_mlp": 0.01035952,
      "balance_loss_clip": 1.05531573,
      "balance_loss_mlp": 1.02705276,
      "epoch": 0.5293092045932785,
      "flos": 19643747610240.0,
      "grad_norm": 1.8415847103311886,
      "language_loss": 0.90744621,
      "learning_rate": 1.9073355594384383e-06,
      "loss": 0.92936802,
      "num_input_tokens_seen": 95109225,
      "step": 4402,
      "time_per_iteration": 2.707036018371582
    },
    {
      "auxiliary_loss_clip": 0.01172986,
      "auxiliary_loss_mlp": 0.01032252,
      "balance_loss_clip": 1.05701196,
      "balance_loss_mlp": 1.0231626,
      "epoch": 0.5294294474839175,
      "flos": 24317736958080.0,
      "grad_norm": 2.8012755067834627,
      "language_loss": 0.80580944,
      "learning_rate": 1.906557431387895e-06,
      "loss": 0.82786185,
      "num_input_tokens_seen": 95128215,
      "step": 4403,
      "time_per_iteration": 2.717489242553711
    },
    {
      "auxiliary_loss_clip": 0.01187978,
      "auxiliary_loss_mlp": 0.00763428,
      "balance_loss_clip": 1.05684638,
      "balance_loss_mlp": 1.00129044,
      "epoch": 0.5295496903745566,
      "flos": 18875941464960.0,
      "grad_norm": 2.1849076883274026,
      "language_loss": 0.7891292,
      "learning_rate": 1.905779317512516e-06,
      "loss": 0.80864322,
      "num_input_tokens_seen": 95145760,
      "step": 4404,
      "time_per_iteration": 2.6395788192749023
    },
    {
      "auxiliary_loss_clip": 0.011887,
      "auxiliary_loss_mlp": 0.0102562,
      "balance_loss_clip": 1.05728316,
      "balance_loss_mlp": 1.01781178,
      "epoch": 0.5296699332651957,
      "flos": 20923101296640.0,
      "grad_norm": 2.135712215286102,
      "language_loss": 0.80564451,
      "learning_rate": 1.9050012179303385e-06,
      "loss": 0.82778776,
      "num_input_tokens_seen": 95164270,
      "step": 4405,
      "time_per_iteration": 2.6291327476501465
    },
    {
      "auxiliary_loss_clip": 0.01190195,
      "auxiliary_loss_mlp": 0.01034115,
      "balance_loss_clip": 1.05575025,
      "balance_loss_mlp": 1.02478051,
      "epoch": 0.5297901761558348,
      "flos": 22046745525120.0,
      "grad_norm": 2.8108983695289322,
      "language_loss": 0.68888992,
      "learning_rate": 1.904223132759401e-06,
      "loss": 0.711133,
      "num_input_tokens_seen": 95182870,
      "step": 4406,
      "time_per_iteration": 3.6628096103668213
    },
    {
      "auxiliary_loss_clip": 0.01147315,
      "auxiliary_loss_mlp": 0.01029576,
      "balance_loss_clip": 1.05192876,
      "balance_loss_mlp": 1.0212853,
      "epoch": 0.5299104190464738,
      "flos": 21798495653760.0,
      "grad_norm": 3.6301945265145537,
      "language_loss": 0.69333577,
      "learning_rate": 1.9034450621177383e-06,
      "loss": 0.7151047,
      "num_input_tokens_seen": 95201190,
      "step": 4407,
      "time_per_iteration": 3.777672529220581
    },
    {
      "auxiliary_loss_clip": 0.01178513,
      "auxiliary_loss_mlp": 0.01035708,
      "balance_loss_clip": 1.05880046,
      "balance_loss_mlp": 1.02640986,
      "epoch": 0.530030661937113,
      "flos": 14720790119040.0,
      "grad_norm": 4.413744143784475,
      "language_loss": 0.70567852,
      "learning_rate": 1.9026670061233824e-06,
      "loss": 0.72782075,
      "num_input_tokens_seen": 95218625,
      "step": 4408,
      "time_per_iteration": 3.573090076446533
    },
    {
      "auxiliary_loss_clip": 0.01172415,
      "auxiliary_loss_mlp": 0.01034592,
      "balance_loss_clip": 1.0569042,
      "balance_loss_mlp": 1.02624786,
      "epoch": 0.5301509048277521,
      "flos": 21251504367360.0,
      "grad_norm": 1.6488436732479055,
      "language_loss": 0.81111681,
      "learning_rate": 1.901888964894365e-06,
      "loss": 0.83318686,
      "num_input_tokens_seen": 95237665,
      "step": 4409,
      "time_per_iteration": 2.6680033206939697
    },
    {
      "auxiliary_loss_clip": 0.0119132,
      "auxiliary_loss_mlp": 0.01031173,
      "balance_loss_clip": 1.05727232,
      "balance_loss_mlp": 1.02173746,
      "epoch": 0.5302711477183911,
      "flos": 25957058791680.0,
      "grad_norm": 1.900352930233973,
      "language_loss": 0.68083721,
      "learning_rate": 1.9011109385487134e-06,
      "loss": 0.70306218,
      "num_input_tokens_seen": 95258915,
      "step": 4410,
      "time_per_iteration": 2.6103851795196533
    },
    {
      "auxiliary_loss_clip": 0.01164216,
      "auxiliary_loss_mlp": 0.01034582,
      "balance_loss_clip": 1.0541147,
      "balance_loss_mlp": 1.02589726,
      "epoch": 0.5303913906090303,
      "flos": 22273126992000.0,
      "grad_norm": 2.353706853947387,
      "language_loss": 0.65818501,
      "learning_rate": 1.900332927204454e-06,
      "loss": 0.68017292,
      "num_input_tokens_seen": 95277365,
      "step": 4411,
      "time_per_iteration": 2.7192468643188477
    },
    {
      "auxiliary_loss_clip": 0.01173608,
      "auxiliary_loss_mlp": 0.01033587,
      "balance_loss_clip": 1.05589676,
      "balance_loss_mlp": 1.02411556,
      "epoch": 0.5305116334996693,
      "flos": 24936010784640.0,
      "grad_norm": 1.9789027195401674,
      "language_loss": 0.76820862,
      "learning_rate": 1.8995549309796097e-06,
      "loss": 0.79028058,
      "num_input_tokens_seen": 95296670,
      "step": 4412,
      "time_per_iteration": 2.6834781169891357
    },
    {
      "auxiliary_loss_clip": 0.01190688,
      "auxiliary_loss_mlp": 0.0103518,
      "balance_loss_clip": 1.05896866,
      "balance_loss_mlp": 1.02611971,
      "epoch": 0.5306318763903084,
      "flos": 20189338266240.0,
      "grad_norm": 1.7415095629541455,
      "language_loss": 0.77199113,
      "learning_rate": 1.8987769499922028e-06,
      "loss": 0.79424989,
      "num_input_tokens_seen": 95315640,
      "step": 4413,
      "time_per_iteration": 2.7290735244750977
    },
    {
      "auxiliary_loss_clip": 0.01177222,
      "auxiliary_loss_mlp": 0.0103545,
      "balance_loss_clip": 1.05642295,
      "balance_loss_mlp": 1.02627122,
      "epoch": 0.5307521192809476,
      "flos": 20266366982400.0,
      "grad_norm": 2.695185824211639,
      "language_loss": 0.70747745,
      "learning_rate": 1.897998984360252e-06,
      "loss": 0.72960418,
      "num_input_tokens_seen": 95334610,
      "step": 4414,
      "time_per_iteration": 2.6839828491210938
    },
    {
      "auxiliary_loss_clip": 0.01170149,
      "auxiliary_loss_mlp": 0.0102907,
      "balance_loss_clip": 1.05660939,
      "balance_loss_mlp": 1.02086794,
      "epoch": 0.5308723621715866,
      "flos": 28844276976000.0,
      "grad_norm": 4.276033204368402,
      "language_loss": 0.78397864,
      "learning_rate": 1.897221034201775e-06,
      "loss": 0.80597079,
      "num_input_tokens_seen": 95358350,
      "step": 4415,
      "time_per_iteration": 2.7702221870422363
    },
    {
      "auxiliary_loss_clip": 0.01171643,
      "auxiliary_loss_mlp": 0.01032962,
      "balance_loss_clip": 1.05221486,
      "balance_loss_mlp": 1.02488518,
      "epoch": 0.5309926050622257,
      "flos": 27457766040960.0,
      "grad_norm": 1.7259178141089175,
      "language_loss": 0.66794574,
      "learning_rate": 1.8964430996347842e-06,
      "loss": 0.68999177,
      "num_input_tokens_seen": 95379900,
      "step": 4416,
      "time_per_iteration": 2.7351491451263428
    },
    {
      "auxiliary_loss_clip": 0.01182549,
      "auxiliary_loss_mlp": 0.01026778,
      "balance_loss_clip": 1.05713964,
      "balance_loss_mlp": 1.01783752,
      "epoch": 0.5311128479528648,
      "flos": 20514545026560.0,
      "grad_norm": 3.406050283470623,
      "language_loss": 0.82406265,
      "learning_rate": 1.8956651807772931e-06,
      "loss": 0.84615594,
      "num_input_tokens_seen": 95397935,
      "step": 4417,
      "time_per_iteration": 3.5887420177459717
    },
    {
      "auxiliary_loss_clip": 0.01187006,
      "auxiliary_loss_mlp": 0.01027902,
      "balance_loss_clip": 1.05697083,
      "balance_loss_mlp": 1.01953399,
      "epoch": 0.5312330908435039,
      "flos": 21397660807680.0,
      "grad_norm": 2.175987676887724,
      "language_loss": 0.83826554,
      "learning_rate": 1.8948872777473115e-06,
      "loss": 0.86041462,
      "num_input_tokens_seen": 95415890,
      "step": 4418,
      "time_per_iteration": 2.6799333095550537
    },
    {
      "auxiliary_loss_clip": 0.01188347,
      "auxiliary_loss_mlp": 0.01032804,
      "balance_loss_clip": 1.05726671,
      "balance_loss_mlp": 1.02404833,
      "epoch": 0.531353333734143,
      "flos": 24717350741760.0,
      "grad_norm": 2.957475749601298,
      "language_loss": 0.63812971,
      "learning_rate": 1.8941093906628458e-06,
      "loss": 0.66034126,
      "num_input_tokens_seen": 95433675,
      "step": 4419,
      "time_per_iteration": 2.6426656246185303
    },
    {
      "auxiliary_loss_clip": 0.01189192,
      "auxiliary_loss_mlp": 0.01033361,
      "balance_loss_clip": 1.05801582,
      "balance_loss_mlp": 1.02484965,
      "epoch": 0.531473576624782,
      "flos": 30480689808000.0,
      "grad_norm": 2.5050989574310414,
      "language_loss": 0.71226978,
      "learning_rate": 1.893331519641902e-06,
      "loss": 0.73449534,
      "num_input_tokens_seen": 95455820,
      "step": 4420,
      "time_per_iteration": 2.7109005451202393
    },
    {
      "auxiliary_loss_clip": 0.01190305,
      "auxiliary_loss_mlp": 0.01025966,
      "balance_loss_clip": 1.05803752,
      "balance_loss_mlp": 1.01717401,
      "epoch": 0.5315938195154212,
      "flos": 23002975440000.0,
      "grad_norm": 3.109120225410408,
      "language_loss": 0.73823559,
      "learning_rate": 1.8925536648024815e-06,
      "loss": 0.76039827,
      "num_input_tokens_seen": 95473240,
      "step": 4421,
      "time_per_iteration": 2.6771132946014404
    },
    {
      "auxiliary_loss_clip": 0.01176461,
      "auxiliary_loss_mlp": 0.01033981,
      "balance_loss_clip": 1.05398941,
      "balance_loss_mlp": 1.02558255,
      "epoch": 0.5317140624060602,
      "flos": 22748584343040.0,
      "grad_norm": 1.9845884540326957,
      "language_loss": 0.75465071,
      "learning_rate": 1.8917758262625849e-06,
      "loss": 0.77675509,
      "num_input_tokens_seen": 95493480,
      "step": 4422,
      "time_per_iteration": 2.6981565952301025
    },
    {
      "auxiliary_loss_clip": 0.01184678,
      "auxiliary_loss_mlp": 0.01031803,
      "balance_loss_clip": 1.05559683,
      "balance_loss_mlp": 1.02335048,
      "epoch": 0.5318343052966993,
      "flos": 22821087945600.0,
      "grad_norm": 1.6612604386247423,
      "language_loss": 0.81034005,
      "learning_rate": 1.8909980041402089e-06,
      "loss": 0.83250487,
      "num_input_tokens_seen": 95512075,
      "step": 4423,
      "time_per_iteration": 2.7202301025390625
    },
    {
      "auxiliary_loss_clip": 0.01156582,
      "auxiliary_loss_mlp": 0.01028014,
      "balance_loss_clip": 1.05429971,
      "balance_loss_mlp": 1.01959777,
      "epoch": 0.5319545481873384,
      "flos": 13626089274240.0,
      "grad_norm": 3.511105857487514,
      "language_loss": 0.65823853,
      "learning_rate": 1.8902201985533494e-06,
      "loss": 0.68008447,
      "num_input_tokens_seen": 95529340,
      "step": 4424,
      "time_per_iteration": 2.742732286453247
    },
    {
      "auxiliary_loss_clip": 0.01160754,
      "auxiliary_loss_mlp": 0.0102539,
      "balance_loss_clip": 1.05385613,
      "balance_loss_mlp": 1.01696157,
      "epoch": 0.5320747910779775,
      "flos": 22162522037760.0,
      "grad_norm": 2.195098057504112,
      "language_loss": 0.74559975,
      "learning_rate": 1.8894424096199983e-06,
      "loss": 0.76746118,
      "num_input_tokens_seen": 95548545,
      "step": 4425,
      "time_per_iteration": 2.7228572368621826
    },
    {
      "auxiliary_loss_clip": 0.01158907,
      "auxiliary_loss_mlp": 0.0103243,
      "balance_loss_clip": 1.05308437,
      "balance_loss_mlp": 1.02380526,
      "epoch": 0.5321950339686166,
      "flos": 18588081870720.0,
      "grad_norm": 2.392481476938838,
      "language_loss": 0.86110479,
      "learning_rate": 1.8886646374581463e-06,
      "loss": 0.88301814,
      "num_input_tokens_seen": 95567770,
      "step": 4426,
      "time_per_iteration": 2.6819939613342285
    },
    {
      "auxiliary_loss_clip": 0.01143501,
      "auxiliary_loss_mlp": 0.01032354,
      "balance_loss_clip": 1.04891145,
      "balance_loss_mlp": 1.02247739,
      "epoch": 0.5323152768592557,
      "flos": 22856818999680.0,
      "grad_norm": 1.767392169966738,
      "language_loss": 0.71535617,
      "learning_rate": 1.8878868821857795e-06,
      "loss": 0.73711479,
      "num_input_tokens_seen": 95587420,
      "step": 4427,
      "time_per_iteration": 2.792189598083496
    },
    {
      "auxiliary_loss_clip": 0.0118886,
      "auxiliary_loss_mlp": 0.01030926,
      "balance_loss_clip": 1.05571222,
      "balance_loss_mlp": 1.02237833,
      "epoch": 0.5324355197498948,
      "flos": 33948690998400.0,
      "grad_norm": 4.384484557508495,
      "language_loss": 0.75129217,
      "learning_rate": 1.8871091439208838e-06,
      "loss": 0.77349001,
      "num_input_tokens_seen": 95609030,
      "step": 4428,
      "time_per_iteration": 2.678220510482788
    },
    {
      "auxiliary_loss_clip": 0.01152014,
      "auxiliary_loss_mlp": 0.01032252,
      "balance_loss_clip": 1.05192041,
      "balance_loss_mlp": 1.02313852,
      "epoch": 0.5325557626405338,
      "flos": 23256720092160.0,
      "grad_norm": 12.1004355929958,
      "language_loss": 0.77100033,
      "learning_rate": 1.8863314227814414e-06,
      "loss": 0.79284298,
      "num_input_tokens_seen": 95627340,
      "step": 4429,
      "time_per_iteration": 2.783684015274048
    },
    {
      "auxiliary_loss_clip": 0.01175683,
      "auxiliary_loss_mlp": 0.01030019,
      "balance_loss_clip": 1.05860448,
      "balance_loss_mlp": 1.02141809,
      "epoch": 0.532676005531173,
      "flos": 26718687797760.0,
      "grad_norm": 2.7922242379410966,
      "language_loss": 0.49403113,
      "learning_rate": 1.8855537188854313e-06,
      "loss": 0.51608813,
      "num_input_tokens_seen": 95646315,
      "step": 4430,
      "time_per_iteration": 2.658017873764038
    },
    {
      "auxiliary_loss_clip": 0.01189411,
      "auxiliary_loss_mlp": 0.01040007,
      "balance_loss_clip": 1.05748773,
      "balance_loss_mlp": 1.03090525,
      "epoch": 0.5327962484218121,
      "flos": 17894610921600.0,
      "grad_norm": 2.1441853997593645,
      "language_loss": 0.78768551,
      "learning_rate": 1.8847760323508315e-06,
      "loss": 0.80997974,
      "num_input_tokens_seen": 95665220,
      "step": 4431,
      "time_per_iteration": 2.639779567718506
    },
    {
      "auxiliary_loss_clip": 0.01186057,
      "auxiliary_loss_mlp": 0.01032413,
      "balance_loss_clip": 1.05699515,
      "balance_loss_mlp": 1.02422929,
      "epoch": 0.5329164913124511,
      "flos": 17925385898880.0,
      "grad_norm": 1.95682929221893,
      "language_loss": 0.75772619,
      "learning_rate": 1.883998363295616e-06,
      "loss": 0.77991092,
      "num_input_tokens_seen": 95682700,
      "step": 4432,
      "time_per_iteration": 3.5459017753601074
    },
    {
      "auxiliary_loss_clip": 0.01075746,
      "auxiliary_loss_mlp": 0.01002269,
      "balance_loss_clip": 1.01311898,
      "balance_loss_mlp": 1.00126779,
      "epoch": 0.5330367342030903,
      "flos": 57254178781440.0,
      "grad_norm": 0.8749599793189338,
      "language_loss": 0.62615967,
      "learning_rate": 1.8832207118377565e-06,
      "loss": 0.64693987,
      "num_input_tokens_seen": 95738070,
      "step": 4433,
      "time_per_iteration": 4.051807641983032
    },
    {
      "auxiliary_loss_clip": 0.01158652,
      "auxiliary_loss_mlp": 0.01027165,
      "balance_loss_clip": 1.05324686,
      "balance_loss_mlp": 1.01902866,
      "epoch": 0.5331569770937293,
      "flos": 17420518287360.0,
      "grad_norm": 1.9078697574049692,
      "language_loss": 0.69212687,
      "learning_rate": 1.882443078095222e-06,
      "loss": 0.71398503,
      "num_input_tokens_seen": 95756950,
      "step": 4434,
      "time_per_iteration": 2.7198359966278076
    },
    {
      "auxiliary_loss_clip": 0.01056834,
      "auxiliary_loss_mlp": 0.01001054,
      "balance_loss_clip": 1.01250792,
      "balance_loss_mlp": 0.9999575,
      "epoch": 0.5332772199843684,
      "flos": 56750783627520.0,
      "grad_norm": 0.8540545620243167,
      "language_loss": 0.66766483,
      "learning_rate": 1.8816654621859794e-06,
      "loss": 0.68824369,
      "num_input_tokens_seen": 95816615,
      "step": 4435,
      "time_per_iteration": 4.170056581497192
    },
    {
      "auxiliary_loss_clip": 0.01187491,
      "auxiliary_loss_mlp": 0.01029541,
      "balance_loss_clip": 1.05636847,
      "balance_loss_mlp": 1.02120817,
      "epoch": 0.5333974628750076,
      "flos": 18697753071360.0,
      "grad_norm": 3.5789161202626874,
      "language_loss": 0.72272432,
      "learning_rate": 1.8808878642279915e-06,
      "loss": 0.74489462,
      "num_input_tokens_seen": 95832020,
      "step": 4436,
      "time_per_iteration": 2.5503733158111572
    },
    {
      "auxiliary_loss_clip": 0.01172427,
      "auxiliary_loss_mlp": 0.01031277,
      "balance_loss_clip": 1.05361891,
      "balance_loss_mlp": 1.02235401,
      "epoch": 0.5335177057656466,
      "flos": 23805507058560.0,
      "grad_norm": 2.3809758290440857,
      "language_loss": 0.64646268,
      "learning_rate": 1.8801102843392209e-06,
      "loss": 0.66849971,
      "num_input_tokens_seen": 95851425,
      "step": 4437,
      "time_per_iteration": 2.700587511062622
    },
    {
      "auxiliary_loss_clip": 0.0115029,
      "auxiliary_loss_mlp": 0.01029697,
      "balance_loss_clip": 1.0473032,
      "balance_loss_mlp": 1.02120876,
      "epoch": 0.5336379486562857,
      "flos": 25078683605760.0,
      "grad_norm": 3.0125685322231965,
      "language_loss": 0.85087955,
      "learning_rate": 1.8793327226376238e-06,
      "loss": 0.87267935,
      "num_input_tokens_seen": 95870745,
      "step": 4438,
      "time_per_iteration": 2.8326852321624756
    },
    {
      "auxiliary_loss_clip": 0.01187969,
      "auxiliary_loss_mlp": 0.01030671,
      "balance_loss_clip": 1.05787206,
      "balance_loss_mlp": 1.02207577,
      "epoch": 0.5337581915469248,
      "flos": 21396691140480.0,
      "grad_norm": 3.5861958853372453,
      "language_loss": 0.80469865,
      "learning_rate": 1.8785551792411569e-06,
      "loss": 0.82688504,
      "num_input_tokens_seen": 95889755,
      "step": 4439,
      "time_per_iteration": 2.6517982482910156
    },
    {
      "auxiliary_loss_clip": 0.01186984,
      "auxiliary_loss_mlp": 0.01027521,
      "balance_loss_clip": 1.05719352,
      "balance_loss_mlp": 1.01936734,
      "epoch": 0.5338784344375639,
      "flos": 14865905064960.0,
      "grad_norm": 2.7867220307057936,
      "language_loss": 0.82855022,
      "learning_rate": 1.8777776542677733e-06,
      "loss": 0.85069525,
      "num_input_tokens_seen": 95907805,
      "step": 4440,
      "time_per_iteration": 2.6270813941955566
    },
    {
      "auxiliary_loss_clip": 0.01187813,
      "auxiliary_loss_mlp": 0.01027592,
      "balance_loss_clip": 1.05597591,
      "balance_loss_mlp": 1.0193429,
      "epoch": 0.5339986773282029,
      "flos": 20813501923200.0,
      "grad_norm": 2.554322625848595,
      "language_loss": 0.73284185,
      "learning_rate": 1.8770001478354216e-06,
      "loss": 0.75499582,
      "num_input_tokens_seen": 95927480,
      "step": 4441,
      "time_per_iteration": 2.6386430263519287
    },
    {
      "auxiliary_loss_clip": 0.01191512,
      "auxiliary_loss_mlp": 0.01029873,
      "balance_loss_clip": 1.05779624,
      "balance_loss_mlp": 1.02102804,
      "epoch": 0.5341189202188421,
      "flos": 17969089772160.0,
      "grad_norm": 2.1518244722063016,
      "language_loss": 0.83975196,
      "learning_rate": 1.8762226600620504e-06,
      "loss": 0.86196578,
      "num_input_tokens_seen": 95946095,
      "step": 4442,
      "time_per_iteration": 2.6385293006896973
    },
    {
      "auxiliary_loss_clip": 0.01177246,
      "auxiliary_loss_mlp": 0.01028633,
      "balance_loss_clip": 1.05780053,
      "balance_loss_mlp": 1.01956725,
      "epoch": 0.5342391631094812,
      "flos": 11031866328960.0,
      "grad_norm": 3.0452684281535354,
      "language_loss": 0.59684151,
      "learning_rate": 1.8754451910656031e-06,
      "loss": 0.6189003,
      "num_input_tokens_seen": 95959995,
      "step": 4443,
      "time_per_iteration": 3.5382678508758545
    },
    {
      "auxiliary_loss_clip": 0.01157803,
      "auxiliary_loss_mlp": 0.01027747,
      "balance_loss_clip": 1.05680501,
      "balance_loss_mlp": 1.0193783,
      "epoch": 0.5343594060001202,
      "flos": 15339135772800.0,
      "grad_norm": 2.7347457097914294,
      "language_loss": 0.82885551,
      "learning_rate": 1.8746677409640212e-06,
      "loss": 0.85071099,
      "num_input_tokens_seen": 95977095,
      "step": 4444,
      "time_per_iteration": 2.7472691535949707
    },
    {
      "auxiliary_loss_clip": 0.01191551,
      "auxiliary_loss_mlp": 0.01039175,
      "balance_loss_clip": 1.06001973,
      "balance_loss_mlp": 1.03027022,
      "epoch": 0.5344796488907594,
      "flos": 26900898514560.0,
      "grad_norm": 2.2169027030139494,
      "language_loss": 0.84775436,
      "learning_rate": 1.8738903098752432e-06,
      "loss": 0.87006164,
      "num_input_tokens_seen": 95996225,
      "step": 4445,
      "time_per_iteration": 2.6404666900634766
    },
    {
      "auxiliary_loss_clip": 0.01173903,
      "auxiliary_loss_mlp": 0.01033627,
      "balance_loss_clip": 1.05701101,
      "balance_loss_mlp": 1.02543187,
      "epoch": 0.5345998917813984,
      "flos": 25411216740480.0,
      "grad_norm": 34.49214562437864,
      "language_loss": 0.73723304,
      "learning_rate": 1.8731128979172052e-06,
      "loss": 0.75930834,
      "num_input_tokens_seen": 96015425,
      "step": 4446,
      "time_per_iteration": 2.7754011154174805
    },
    {
      "auxiliary_loss_clip": 0.01167103,
      "auxiliary_loss_mlp": 0.01033291,
      "balance_loss_clip": 1.05074203,
      "balance_loss_mlp": 1.02564931,
      "epoch": 0.5347201346720375,
      "flos": 32853379622400.0,
      "grad_norm": 2.909214657545158,
      "language_loss": 0.67110705,
      "learning_rate": 1.8723355052078394e-06,
      "loss": 0.69311094,
      "num_input_tokens_seen": 96035460,
      "step": 4447,
      "time_per_iteration": 2.7319071292877197
    },
    {
      "auxiliary_loss_clip": 0.01184225,
      "auxiliary_loss_mlp": 0.01040694,
      "balance_loss_clip": 1.05905533,
      "balance_loss_mlp": 1.03099668,
      "epoch": 0.5348403775626767,
      "flos": 17967940536960.0,
      "grad_norm": 3.5344677161669686,
      "language_loss": 0.77540076,
      "learning_rate": 1.8715581318650765e-06,
      "loss": 0.79764998,
      "num_input_tokens_seen": 96054515,
      "step": 4448,
      "time_per_iteration": 2.747154474258423
    },
    {
      "auxiliary_loss_clip": 0.0119623,
      "auxiliary_loss_mlp": 0.01030034,
      "balance_loss_clip": 1.06192589,
      "balance_loss_mlp": 1.02034831,
      "epoch": 0.5349606204533157,
      "flos": 17603339535360.0,
      "grad_norm": 2.441154155739293,
      "language_loss": 0.81537646,
      "learning_rate": 1.8707807780068422e-06,
      "loss": 0.83763909,
      "num_input_tokens_seen": 96072330,
      "step": 4449,
      "time_per_iteration": 2.6674563884735107
    },
    {
      "auxiliary_loss_clip": 0.01176511,
      "auxiliary_loss_mlp": 0.01027692,
      "balance_loss_clip": 1.05420005,
      "balance_loss_mlp": 1.01937127,
      "epoch": 0.5350808633439548,
      "flos": 29167831710720.0,
      "grad_norm": 1.9695884798426726,
      "language_loss": 0.66435951,
      "learning_rate": 1.8700034437510611e-06,
      "loss": 0.68640155,
      "num_input_tokens_seen": 96092425,
      "step": 4450,
      "time_per_iteration": 2.746509552001953
    },
    {
      "auxiliary_loss_clip": 0.011739,
      "auxiliary_loss_mlp": 0.01041486,
      "balance_loss_clip": 1.05783904,
      "balance_loss_mlp": 1.032969,
      "epoch": 0.5352011062345938,
      "flos": 19499997381120.0,
      "grad_norm": 2.6074104573775627,
      "language_loss": 0.81514251,
      "learning_rate": 1.8692261292156549e-06,
      "loss": 0.83729637,
      "num_input_tokens_seen": 96111660,
      "step": 4451,
      "time_per_iteration": 2.7069151401519775
    },
    {
      "auxiliary_loss_clip": 0.01184962,
      "auxiliary_loss_mlp": 0.01034241,
      "balance_loss_clip": 1.05685377,
      "balance_loss_mlp": 1.02606344,
      "epoch": 0.535321349125233,
      "flos": 23477642691840.0,
      "grad_norm": 2.345672050805596,
      "language_loss": 0.81459075,
      "learning_rate": 1.8684488345185401e-06,
      "loss": 0.83678281,
      "num_input_tokens_seen": 96131835,
      "step": 4452,
      "time_per_iteration": 2.5459320545196533
    },
    {
      "auxiliary_loss_clip": 0.01176042,
      "auxiliary_loss_mlp": 0.01032658,
      "balance_loss_clip": 1.05998218,
      "balance_loss_mlp": 1.02403331,
      "epoch": 0.535441592015872,
      "flos": 20478059786880.0,
      "grad_norm": 2.071538957147415,
      "language_loss": 0.79138601,
      "learning_rate": 1.8676715597776332e-06,
      "loss": 0.81347311,
      "num_input_tokens_seen": 96150180,
      "step": 4453,
      "time_per_iteration": 2.682645559310913
    },
    {
      "auxiliary_loss_clip": 0.01172415,
      "auxiliary_loss_mlp": 0.01028083,
      "balance_loss_clip": 1.05651069,
      "balance_loss_mlp": 1.01984012,
      "epoch": 0.5355618349065111,
      "flos": 19573147428480.0,
      "grad_norm": 2.268142977224219,
      "language_loss": 0.76397073,
      "learning_rate": 1.8668943051108455e-06,
      "loss": 0.78597569,
      "num_input_tokens_seen": 96167485,
      "step": 4454,
      "time_per_iteration": 2.602931261062622
    },
    {
      "auxiliary_loss_clip": 0.01188429,
      "auxiliary_loss_mlp": 0.0102956,
      "balance_loss_clip": 1.05790305,
      "balance_loss_mlp": 1.02120388,
      "epoch": 0.5356820777971503,
      "flos": 24024633978240.0,
      "grad_norm": 1.8141896506291024,
      "language_loss": 0.76382136,
      "learning_rate": 1.8661170706360856e-06,
      "loss": 0.78600121,
      "num_input_tokens_seen": 96186650,
      "step": 4455,
      "time_per_iteration": 2.6930465698242188
    },
    {
      "auxiliary_loss_clip": 0.01169332,
      "auxiliary_loss_mlp": 0.01031329,
      "balance_loss_clip": 1.05617809,
      "balance_loss_mlp": 1.02342558,
      "epoch": 0.5358023206877893,
      "flos": 20884676722560.0,
      "grad_norm": 1.6057093206339557,
      "language_loss": 0.81688333,
      "learning_rate": 1.8653398564712594e-06,
      "loss": 0.83888996,
      "num_input_tokens_seen": 96205595,
      "step": 4456,
      "time_per_iteration": 2.631998062133789
    },
    {
      "auxiliary_loss_clip": 0.01173959,
      "auxiliary_loss_mlp": 0.01034373,
      "balance_loss_clip": 1.05680919,
      "balance_loss_mlp": 1.02623701,
      "epoch": 0.5359225635784284,
      "flos": 22418996123520.0,
      "grad_norm": 2.8886519077033794,
      "language_loss": 0.8227151,
      "learning_rate": 1.8645626627342704e-06,
      "loss": 0.84479845,
      "num_input_tokens_seen": 96226360,
      "step": 4457,
      "time_per_iteration": 2.7765698432922363
    },
    {
      "auxiliary_loss_clip": 0.01175243,
      "auxiliary_loss_mlp": 0.01027228,
      "balance_loss_clip": 1.05437613,
      "balance_loss_mlp": 1.01837671,
      "epoch": 0.5360428064690675,
      "flos": 24097784025600.0,
      "grad_norm": 2.45802692442425,
      "language_loss": 0.81195658,
      "learning_rate": 1.8637854895430172e-06,
      "loss": 0.83398128,
      "num_input_tokens_seen": 96245625,
      "step": 4458,
      "time_per_iteration": 3.6245670318603516
    },
    {
      "auxiliary_loss_clip": 0.01174055,
      "auxiliary_loss_mlp": 0.01034874,
      "balance_loss_clip": 1.05221891,
      "balance_loss_mlp": 1.02630234,
      "epoch": 0.5361630493597066,
      "flos": 21434505183360.0,
      "grad_norm": 3.4477414562555806,
      "language_loss": 0.69590521,
      "learning_rate": 1.8630083370153978e-06,
      "loss": 0.71799445,
      "num_input_tokens_seen": 96265265,
      "step": 4459,
      "time_per_iteration": 3.648620128631592
    },
    {
      "auxiliary_loss_clip": 0.01053345,
      "auxiliary_loss_mlp": 0.01000488,
      "balance_loss_clip": 1.01160026,
      "balance_loss_mlp": 0.99940294,
      "epoch": 0.5362832922503457,
      "flos": 68888696520960.0,
      "grad_norm": 0.7430068849638568,
      "language_loss": 0.55359042,
      "learning_rate": 1.8622312052693041e-06,
      "loss": 0.57412875,
      "num_input_tokens_seen": 96326445,
      "step": 4460,
      "time_per_iteration": 3.2941155433654785
    },
    {
      "auxiliary_loss_clip": 0.01154775,
      "auxiliary_loss_mlp": 0.01031122,
      "balance_loss_clip": 1.0510329,
      "balance_loss_mlp": 1.02277172,
      "epoch": 0.5364035351409848,
      "flos": 9793702563840.0,
      "grad_norm": 3.739759384600425,
      "language_loss": 0.73110378,
      "learning_rate": 1.8614540944226267e-06,
      "loss": 0.75296283,
      "num_input_tokens_seen": 96343115,
      "step": 4461,
      "time_per_iteration": 3.6963770389556885
    },
    {
      "auxiliary_loss_clip": 0.01186974,
      "auxiliary_loss_mlp": 0.01032214,
      "balance_loss_clip": 1.05870533,
      "balance_loss_mlp": 1.02463233,
      "epoch": 0.5365237780316239,
      "flos": 23290080848640.0,
      "grad_norm": 2.2822389509627783,
      "language_loss": 0.68454039,
      "learning_rate": 1.8606770045932537e-06,
      "loss": 0.70673227,
      "num_input_tokens_seen": 96362230,
      "step": 4462,
      "time_per_iteration": 2.61808705329895
    },
    {
      "auxiliary_loss_clip": 0.0117428,
      "auxiliary_loss_mlp": 0.010288,
      "balance_loss_clip": 1.05710077,
      "balance_loss_mlp": 1.01932859,
      "epoch": 0.5366440209222629,
      "flos": 26578133879040.0,
      "grad_norm": 3.669552083763479,
      "language_loss": 0.81533217,
      "learning_rate": 1.859899935899068e-06,
      "loss": 0.837363,
      "num_input_tokens_seen": 96382085,
      "step": 4463,
      "time_per_iteration": 2.737396001815796
    },
    {
      "auxiliary_loss_clip": 0.01187606,
      "auxiliary_loss_mlp": 0.01035696,
      "balance_loss_clip": 1.055444,
      "balance_loss_mlp": 1.02643967,
      "epoch": 0.5367642638129021,
      "flos": 19608052469760.0,
      "grad_norm": 1.5849630479787462,
      "language_loss": 0.79404986,
      "learning_rate": 1.8591228884579506e-06,
      "loss": 0.81628287,
      "num_input_tokens_seen": 96400580,
      "step": 4464,
      "time_per_iteration": 2.6402626037597656
    },
    {
      "auxiliary_loss_clip": 0.01184373,
      "auxiliary_loss_mlp": 0.01032696,
      "balance_loss_clip": 1.05439198,
      "balance_loss_mlp": 1.02447069,
      "epoch": 0.5368845067035412,
      "flos": 23915214172800.0,
      "grad_norm": 8.374246637865847,
      "language_loss": 0.81828249,
      "learning_rate": 1.8583458623877795e-06,
      "loss": 0.84045315,
      "num_input_tokens_seen": 96419680,
      "step": 4465,
      "time_per_iteration": 2.671107053756714
    },
    {
      "auxiliary_loss_clip": 0.01173284,
      "auxiliary_loss_mlp": 0.00762591,
      "balance_loss_clip": 1.05605519,
      "balance_loss_mlp": 1.00142956,
      "epoch": 0.5370047495941802,
      "flos": 16873131951360.0,
      "grad_norm": 1.7553464868513649,
      "language_loss": 0.74297076,
      "learning_rate": 1.8575688578064281e-06,
      "loss": 0.76232958,
      "num_input_tokens_seen": 96437805,
      "step": 4466,
      "time_per_iteration": 2.618847608566284
    },
    {
      "auxiliary_loss_clip": 0.01189752,
      "auxiliary_loss_mlp": 0.01031228,
      "balance_loss_clip": 1.05842888,
      "balance_loss_mlp": 1.02260911,
      "epoch": 0.5371249924848194,
      "flos": 20740926493440.0,
      "grad_norm": 1.8994069055201832,
      "language_loss": 0.77188885,
      "learning_rate": 1.8567918748317674e-06,
      "loss": 0.79409862,
      "num_input_tokens_seen": 96457155,
      "step": 4467,
      "time_per_iteration": 2.6633846759796143
    },
    {
      "auxiliary_loss_clip": 0.01147418,
      "auxiliary_loss_mlp": 0.01032372,
      "balance_loss_clip": 1.05109787,
      "balance_loss_mlp": 1.02359843,
      "epoch": 0.5372452353754584,
      "flos": 17968120104960.0,
      "grad_norm": 2.1555945184180154,
      "language_loss": 0.83020788,
      "learning_rate": 1.8560149135816659e-06,
      "loss": 0.85200584,
      "num_input_tokens_seen": 96473990,
      "step": 4468,
      "time_per_iteration": 2.6488723754882812
    },
    {
      "auxiliary_loss_clip": 0.01184613,
      "auxiliary_loss_mlp": 0.01030462,
      "balance_loss_clip": 1.05503464,
      "balance_loss_mlp": 1.02204633,
      "epoch": 0.5373654782660975,
      "flos": 15377021642880.0,
      "grad_norm": 2.344580480286949,
      "language_loss": 0.84367043,
      "learning_rate": 1.8552379741739873e-06,
      "loss": 0.86582118,
      "num_input_tokens_seen": 96491335,
      "step": 4469,
      "time_per_iteration": 3.6549232006073
    },
    {
      "auxiliary_loss_clip": 0.0107277,
      "auxiliary_loss_mlp": 0.01003074,
      "balance_loss_clip": 1.01066566,
      "balance_loss_mlp": 1.00203061,
      "epoch": 0.5374857211567367,
      "flos": 69000091574400.0,
      "grad_norm": 0.8929296163305921,
      "language_loss": 0.55640221,
      "learning_rate": 1.8544610567265935e-06,
      "loss": 0.5771606,
      "num_input_tokens_seen": 96545275,
      "step": 4470,
      "time_per_iteration": 3.2397069931030273
    },
    {
      "auxiliary_loss_clip": 0.01168916,
      "auxiliary_loss_mlp": 0.01032258,
      "balance_loss_clip": 1.05539739,
      "balance_loss_mlp": 1.024158,
      "epoch": 0.5376059640473757,
      "flos": 15085355207040.0,
      "grad_norm": 1.9573891495552493,
      "language_loss": 0.83774602,
      "learning_rate": 1.853684161357341e-06,
      "loss": 0.85975778,
      "num_input_tokens_seen": 96562935,
      "step": 4471,
      "time_per_iteration": 2.6128973960876465
    },
    {
      "auxiliary_loss_clip": 0.01170798,
      "auxiliary_loss_mlp": 0.01028742,
      "balance_loss_clip": 1.0575273,
      "balance_loss_mlp": 1.0206778,
      "epoch": 0.5377262069380148,
      "flos": 19792597570560.0,
      "grad_norm": 1.8360159474599171,
      "language_loss": 0.7716738,
      "learning_rate": 1.852907288184085e-06,
      "loss": 0.79366916,
      "num_input_tokens_seen": 96581820,
      "step": 4472,
      "time_per_iteration": 2.7480380535125732
    },
    {
      "auxiliary_loss_clip": 0.01172475,
      "auxiliary_loss_mlp": 0.01030183,
      "balance_loss_clip": 1.05412066,
      "balance_loss_mlp": 1.02124858,
      "epoch": 0.5378464498286539,
      "flos": 30003077640960.0,
      "grad_norm": 5.783901902472473,
      "language_loss": 0.70105147,
      "learning_rate": 1.8521304373246762e-06,
      "loss": 0.72307807,
      "num_input_tokens_seen": 96602865,
      "step": 4473,
      "time_per_iteration": 2.714970588684082
    },
    {
      "auxiliary_loss_clip": 0.01174641,
      "auxiliary_loss_mlp": 0.01029726,
      "balance_loss_clip": 1.05598283,
      "balance_loss_mlp": 1.02083945,
      "epoch": 0.537966692719293,
      "flos": 21251217058560.0,
      "grad_norm": 3.0880822417296256,
      "language_loss": 0.89222848,
      "learning_rate": 1.8513536088969626e-06,
      "loss": 0.91427207,
      "num_input_tokens_seen": 96620530,
      "step": 4474,
      "time_per_iteration": 2.715424060821533
    },
    {
      "auxiliary_loss_clip": 0.01189229,
      "auxiliary_loss_mlp": 0.01040492,
      "balance_loss_clip": 1.05726278,
      "balance_loss_mlp": 1.03200448,
      "epoch": 0.538086935609932,
      "flos": 21543170803200.0,
      "grad_norm": 2.3529190912676903,
      "language_loss": 0.80337906,
      "learning_rate": 1.8505768030187884e-06,
      "loss": 0.82567626,
      "num_input_tokens_seen": 96640660,
      "step": 4475,
      "time_per_iteration": 2.5880606174468994
    },
    {
      "auxiliary_loss_clip": 0.0116219,
      "auxiliary_loss_mlp": 0.01031614,
      "balance_loss_clip": 1.05302298,
      "balance_loss_mlp": 1.02311456,
      "epoch": 0.5382071785005712,
      "flos": 22747219626240.0,
      "grad_norm": 1.7039826235336935,
      "language_loss": 0.80385536,
      "learning_rate": 1.849800019807995e-06,
      "loss": 0.82579345,
      "num_input_tokens_seen": 96661885,
      "step": 4476,
      "time_per_iteration": 2.779714822769165
    },
    {
      "auxiliary_loss_clip": 0.01159562,
      "auxiliary_loss_mlp": 0.01030219,
      "balance_loss_clip": 1.05156648,
      "balance_loss_mlp": 1.02160025,
      "epoch": 0.5383274213912103,
      "flos": 24934574240640.0,
      "grad_norm": 2.2497779154582664,
      "language_loss": 0.71692127,
      "learning_rate": 1.8490232593824186e-06,
      "loss": 0.73881912,
      "num_input_tokens_seen": 96678340,
      "step": 4477,
      "time_per_iteration": 2.8848836421966553
    },
    {
      "auxiliary_loss_clip": 0.01162605,
      "auxiliary_loss_mlp": 0.0102774,
      "balance_loss_clip": 1.05219769,
      "balance_loss_mlp": 1.01950264,
      "epoch": 0.5384476642818493,
      "flos": 22310186849280.0,
      "grad_norm": 2.0751942803751255,
      "language_loss": 0.84974158,
      "learning_rate": 1.8482465218598935e-06,
      "loss": 0.87164509,
      "num_input_tokens_seen": 96698285,
      "step": 4478,
      "time_per_iteration": 2.6537201404571533
    },
    {
      "auxiliary_loss_clip": 0.01175051,
      "auxiliary_loss_mlp": 0.01029794,
      "balance_loss_clip": 1.05830121,
      "balance_loss_mlp": 1.02134812,
      "epoch": 0.5385679071724885,
      "flos": 22711021695360.0,
      "grad_norm": 1.8946920844129278,
      "language_loss": 0.83523411,
      "learning_rate": 1.8474698073582508e-06,
      "loss": 0.85728252,
      "num_input_tokens_seen": 96719655,
      "step": 4479,
      "time_per_iteration": 2.7142086029052734
    },
    {
      "auxiliary_loss_clip": 0.0117795,
      "auxiliary_loss_mlp": 0.01033016,
      "balance_loss_clip": 1.05806601,
      "balance_loss_mlp": 1.02422428,
      "epoch": 0.5386881500631275,
      "flos": 15953746412160.0,
      "grad_norm": 2.4399462330687323,
      "language_loss": 0.87145603,
      "learning_rate": 1.8466931159953166e-06,
      "loss": 0.89356565,
      "num_input_tokens_seen": 96736290,
      "step": 4480,
      "time_per_iteration": 2.624830722808838
    },
    {
      "auxiliary_loss_clip": 0.01176094,
      "auxiliary_loss_mlp": 0.01030153,
      "balance_loss_clip": 1.05836594,
      "balance_loss_mlp": 1.02171898,
      "epoch": 0.5388083929537666,
      "flos": 24060041809920.0,
      "grad_norm": 2.4288423190150343,
      "language_loss": 0.84514236,
      "learning_rate": 1.8459164478889158e-06,
      "loss": 0.86720484,
      "num_input_tokens_seen": 96757685,
      "step": 4481,
      "time_per_iteration": 2.7222208976745605
    },
    {
      "auxiliary_loss_clip": 0.01186409,
      "auxiliary_loss_mlp": 0.01031584,
      "balance_loss_clip": 1.05778146,
      "balance_loss_mlp": 1.02347517,
      "epoch": 0.5389286358444056,
      "flos": 22236893147520.0,
      "grad_norm": 1.707249890326874,
      "language_loss": 0.75817204,
      "learning_rate": 1.8451398031568663e-06,
      "loss": 0.780352,
      "num_input_tokens_seen": 96777310,
      "step": 4482,
      "time_per_iteration": 2.597100257873535
    },
    {
      "auxiliary_loss_clip": 0.01146697,
      "auxiliary_loss_mlp": 0.01024842,
      "balance_loss_clip": 1.05567622,
      "balance_loss_mlp": 1.01656914,
      "epoch": 0.5390488787350448,
      "flos": 24281718595200.0,
      "grad_norm": 1.8476483013745941,
      "language_loss": 0.75146157,
      "learning_rate": 1.844363181916986e-06,
      "loss": 0.77317697,
      "num_input_tokens_seen": 96798035,
      "step": 4483,
      "time_per_iteration": 2.902827739715576
    },
    {
      "auxiliary_loss_clip": 0.01182133,
      "auxiliary_loss_mlp": 0.0103132,
      "balance_loss_clip": 1.05748653,
      "balance_loss_mlp": 1.02284968,
      "epoch": 0.5391691216256839,
      "flos": 16581393688320.0,
      "grad_norm": 2.6073273117434494,
      "language_loss": 0.82934213,
      "learning_rate": 1.8435865842870868e-06,
      "loss": 0.85147661,
      "num_input_tokens_seen": 96815975,
      "step": 4484,
      "time_per_iteration": 3.6013612747192383
    },
    {
      "auxiliary_loss_clip": 0.01176407,
      "auxiliary_loss_mlp": 0.01032077,
      "balance_loss_clip": 1.05816793,
      "balance_loss_mlp": 1.02372646,
      "epoch": 0.5392893645163229,
      "flos": 23330049707520.0,
      "grad_norm": 2.339158502833298,
      "language_loss": 0.72144973,
      "learning_rate": 1.8428100103849787e-06,
      "loss": 0.74353451,
      "num_input_tokens_seen": 96835770,
      "step": 4485,
      "time_per_iteration": 3.6503071784973145
    },
    {
      "auxiliary_loss_clip": 0.011639,
      "auxiliary_loss_mlp": 0.01029575,
      "balance_loss_clip": 1.05471158,
      "balance_loss_mlp": 1.02143312,
      "epoch": 0.5394096074069621,
      "flos": 15669801400320.0,
      "grad_norm": 3.3043441387231365,
      "language_loss": 0.73710567,
      "learning_rate": 1.842033460328467e-06,
      "loss": 0.75904036,
      "num_input_tokens_seen": 96854490,
      "step": 4486,
      "time_per_iteration": 2.687183380126953
    },
    {
      "auxiliary_loss_clip": 0.01162031,
      "auxiliary_loss_mlp": 0.01029269,
      "balance_loss_clip": 1.05258918,
      "balance_loss_mlp": 1.0215143,
      "epoch": 0.5395298502976011,
      "flos": 22893447893760.0,
      "grad_norm": 1.7309426813549567,
      "language_loss": 0.75096583,
      "learning_rate": 1.8412569342353541e-06,
      "loss": 0.77287889,
      "num_input_tokens_seen": 96874645,
      "step": 4487,
      "time_per_iteration": 3.665029764175415
    },
    {
      "auxiliary_loss_clip": 0.01137428,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.05113971,
      "balance_loss_mlp": 1.01909459,
      "epoch": 0.5396500931882402,
      "flos": 23842135952640.0,
      "grad_norm": 2.183133594512857,
      "language_loss": 0.84927523,
      "learning_rate": 1.840480432223438e-06,
      "loss": 0.87092948,
      "num_input_tokens_seen": 96893650,
      "step": 4488,
      "time_per_iteration": 2.7331902980804443
    },
    {
      "auxiliary_loss_clip": 0.01157682,
      "auxiliary_loss_mlp": 0.0102818,
      "balance_loss_clip": 1.05473673,
      "balance_loss_mlp": 1.01984715,
      "epoch": 0.5397703360788794,
      "flos": 26322988596480.0,
      "grad_norm": 2.09349059069155,
      "language_loss": 0.77997863,
      "learning_rate": 1.8397039544105131e-06,
      "loss": 0.80183721,
      "num_input_tokens_seen": 96912735,
      "step": 4489,
      "time_per_iteration": 2.804117202758789
    },
    {
      "auxiliary_loss_clip": 0.01170553,
      "auxiliary_loss_mlp": 0.01033761,
      "balance_loss_clip": 1.05413389,
      "balance_loss_mlp": 1.02524316,
      "epoch": 0.5398905789695184,
      "flos": 21214588164480.0,
      "grad_norm": 2.269435890065144,
      "language_loss": 0.70226467,
      "learning_rate": 1.8389275009143711e-06,
      "loss": 0.72430778,
      "num_input_tokens_seen": 96932475,
      "step": 4490,
      "time_per_iteration": 2.6472132205963135
    },
    {
      "auxiliary_loss_clip": 0.0118245,
      "auxiliary_loss_mlp": 0.01026066,
      "balance_loss_clip": 1.05443275,
      "balance_loss_mlp": 1.01813889,
      "epoch": 0.5400108218601575,
      "flos": 25080335631360.0,
      "grad_norm": 1.9869730630576437,
      "language_loss": 0.73442113,
      "learning_rate": 1.8381510718527988e-06,
      "loss": 0.7565062,
      "num_input_tokens_seen": 96952085,
      "step": 4491,
      "time_per_iteration": 2.673447608947754
    },
    {
      "auxiliary_loss_clip": 0.01177118,
      "auxiliary_loss_mlp": 0.01031483,
      "balance_loss_clip": 1.0581522,
      "balance_loss_mlp": 1.02346039,
      "epoch": 0.5401310647507966,
      "flos": 26357498588160.0,
      "grad_norm": 1.882572504018517,
      "language_loss": 0.63944101,
      "learning_rate": 1.8373746673435812e-06,
      "loss": 0.66152704,
      "num_input_tokens_seen": 96973110,
      "step": 4492,
      "time_per_iteration": 2.683483600616455
    },
    {
      "auxiliary_loss_clip": 0.0118935,
      "auxiliary_loss_mlp": 0.01029873,
      "balance_loss_clip": 1.05899954,
      "balance_loss_mlp": 1.02170658,
      "epoch": 0.5402513076414357,
      "flos": 27855332749440.0,
      "grad_norm": 3.0231511813907277,
      "language_loss": 0.79404426,
      "learning_rate": 1.8365982875044964e-06,
      "loss": 0.81623644,
      "num_input_tokens_seen": 96993420,
      "step": 4493,
      "time_per_iteration": 2.710148572921753
    },
    {
      "auxiliary_loss_clip": 0.01177568,
      "auxiliary_loss_mlp": 0.01029978,
      "balance_loss_clip": 1.05847669,
      "balance_loss_mlp": 1.02084708,
      "epoch": 0.5403715505320748,
      "flos": 22893771116160.0,
      "grad_norm": 2.666216561744523,
      "language_loss": 0.7616874,
      "learning_rate": 1.8358219324533217e-06,
      "loss": 0.78376293,
      "num_input_tokens_seen": 97013685,
      "step": 4494,
      "time_per_iteration": 2.656820774078369
    },
    {
      "auxiliary_loss_clip": 0.01181812,
      "auxiliary_loss_mlp": 0.01028946,
      "balance_loss_clip": 1.05631948,
      "balance_loss_mlp": 1.02097058,
      "epoch": 0.5404917934227139,
      "flos": 30224143895040.0,
      "grad_norm": 1.6151791008511702,
      "language_loss": 0.70390725,
      "learning_rate": 1.8350456023078292e-06,
      "loss": 0.72601485,
      "num_input_tokens_seen": 97036060,
      "step": 4495,
      "time_per_iteration": 3.6590476036071777
    },
    {
      "auxiliary_loss_clip": 0.01178739,
      "auxiliary_loss_mlp": 0.01036277,
      "balance_loss_clip": 1.05522943,
      "balance_loss_mlp": 1.02725339,
      "epoch": 0.540612036313353,
      "flos": 19938502615680.0,
      "grad_norm": 2.239228619634623,
      "language_loss": 0.77953851,
      "learning_rate": 1.8342692971857874e-06,
      "loss": 0.80168867,
      "num_input_tokens_seen": 97055260,
      "step": 4496,
      "time_per_iteration": 2.6657702922821045
    },
    {
      "auxiliary_loss_clip": 0.01174833,
      "auxiliary_loss_mlp": 0.01028831,
      "balance_loss_clip": 1.05459046,
      "balance_loss_mlp": 1.02084994,
      "epoch": 0.540732279203992,
      "flos": 24279599692800.0,
      "grad_norm": 6.238716062072259,
      "language_loss": 0.71439338,
      "learning_rate": 1.833493017204962e-06,
      "loss": 0.73642999,
      "num_input_tokens_seen": 97075365,
      "step": 4497,
      "time_per_iteration": 2.653043031692505
    },
    {
      "auxiliary_loss_clip": 0.01146759,
      "auxiliary_loss_mlp": 0.01026096,
      "balance_loss_clip": 1.05216622,
      "balance_loss_mlp": 1.01797771,
      "epoch": 0.5408525220946312,
      "flos": 20193216935040.0,
      "grad_norm": 2.7181869519230855,
      "language_loss": 0.78316998,
      "learning_rate": 1.8327167624831134e-06,
      "loss": 0.8048985,
      "num_input_tokens_seen": 97093095,
      "step": 4498,
      "time_per_iteration": 2.6948323249816895
    },
    {
      "auxiliary_loss_clip": 0.01190747,
      "auxiliary_loss_mlp": 0.01030743,
      "balance_loss_clip": 1.05981326,
      "balance_loss_mlp": 1.02166533,
      "epoch": 0.5409727649852702,
      "flos": 24134448833280.0,
      "grad_norm": 1.9600847426058263,
      "language_loss": 0.7081303,
      "learning_rate": 1.831940533137999e-06,
      "loss": 0.73034525,
      "num_input_tokens_seen": 97112000,
      "step": 4499,
      "time_per_iteration": 2.6412482261657715
    },
    {
      "auxiliary_loss_clip": 0.01170725,
      "auxiliary_loss_mlp": 0.01031676,
      "balance_loss_clip": 1.05573702,
      "balance_loss_mlp": 1.02293801,
      "epoch": 0.5410930078759093,
      "flos": 23912700220800.0,
      "grad_norm": 1.8902119431637205,
      "language_loss": 0.72558224,
      "learning_rate": 1.8311643292873718e-06,
      "loss": 0.74760628,
      "num_input_tokens_seen": 97130820,
      "step": 4500,
      "time_per_iteration": 2.7162184715270996
    },
    {
      "auxiliary_loss_clip": 0.01164707,
      "auxiliary_loss_mlp": 0.01034304,
      "balance_loss_clip": 1.05420542,
      "balance_loss_mlp": 1.02603722,
      "epoch": 0.5412132507665485,
      "flos": 21105132445440.0,
      "grad_norm": 1.9590348029503024,
      "language_loss": 0.88005221,
      "learning_rate": 1.8303881510489818e-06,
      "loss": 0.90204227,
      "num_input_tokens_seen": 97149210,
      "step": 4501,
      "time_per_iteration": 2.7113842964172363
    },
    {
      "auxiliary_loss_clip": 0.01191934,
      "auxiliary_loss_mlp": 0.01031942,
      "balance_loss_clip": 1.05942857,
      "balance_loss_mlp": 1.02357388,
      "epoch": 0.5413334936571875,
      "flos": 30227340205440.0,
      "grad_norm": 2.2865631404739504,
      "language_loss": 0.69892842,
      "learning_rate": 1.829611998540574e-06,
      "loss": 0.72116715,
      "num_input_tokens_seen": 97170415,
      "step": 4502,
      "time_per_iteration": 2.714559555053711
    },
    {
      "auxiliary_loss_clip": 0.01186109,
      "auxiliary_loss_mlp": 0.01027725,
      "balance_loss_clip": 1.0543102,
      "balance_loss_mlp": 1.01911759,
      "epoch": 0.5414537365478266,
      "flos": 24279635606400.0,
      "grad_norm": 1.9737118403131761,
      "language_loss": 0.80050319,
      "learning_rate": 1.8288358718798914e-06,
      "loss": 0.82264149,
      "num_input_tokens_seen": 97189605,
      "step": 4503,
      "time_per_iteration": 2.6226601600646973
    },
    {
      "auxiliary_loss_clip": 0.01155381,
      "auxiliary_loss_mlp": 0.01032777,
      "balance_loss_clip": 1.05553806,
      "balance_loss_mlp": 1.0249809,
      "epoch": 0.5415739794384657,
      "flos": 16654543735680.0,
      "grad_norm": 3.4976351681861564,
      "language_loss": 0.72944999,
      "learning_rate": 1.8280597711846703e-06,
      "loss": 0.75133157,
      "num_input_tokens_seen": 97207845,
      "step": 4504,
      "time_per_iteration": 2.738022804260254
    },
    {
      "auxiliary_loss_clip": 0.01184128,
      "auxiliary_loss_mlp": 0.0103076,
      "balance_loss_clip": 1.05521154,
      "balance_loss_mlp": 1.02342319,
      "epoch": 0.5416942223291048,
      "flos": 23185724860800.0,
      "grad_norm": 2.3441337375258136,
      "language_loss": 0.83545804,
      "learning_rate": 1.8272836965726455e-06,
      "loss": 0.85760689,
      "num_input_tokens_seen": 97226780,
      "step": 4505,
      "time_per_iteration": 2.65932297706604
    },
    {
      "auxiliary_loss_clip": 0.01174489,
      "auxiliary_loss_mlp": 0.01034604,
      "balance_loss_clip": 1.05741286,
      "balance_loss_mlp": 1.02600312,
      "epoch": 0.5418144652197439,
      "flos": 20303247271680.0,
      "grad_norm": 1.7252810631616997,
      "language_loss": 0.78947794,
      "learning_rate": 1.8265076481615461e-06,
      "loss": 0.81156886,
      "num_input_tokens_seen": 97246695,
      "step": 4506,
      "time_per_iteration": 2.6746292114257812
    },
    {
      "auxiliary_loss_clip": 0.01177387,
      "auxiliary_loss_mlp": 0.0103255,
      "balance_loss_clip": 1.0577203,
      "balance_loss_mlp": 1.02397895,
      "epoch": 0.541934708110383,
      "flos": 12458633431680.0,
      "grad_norm": 3.0746843907092596,
      "language_loss": 0.87690783,
      "learning_rate": 1.8257316260690987e-06,
      "loss": 0.8990072,
      "num_input_tokens_seen": 97264480,
      "step": 4507,
      "time_per_iteration": 2.6984896659851074
    },
    {
      "auxiliary_loss_clip": 0.01183854,
      "auxiliary_loss_mlp": 0.01030852,
      "balance_loss_clip": 1.05600882,
      "balance_loss_mlp": 1.02291811,
      "epoch": 0.5420549510010221,
      "flos": 21253802837760.0,
      "grad_norm": 1.8825945034371685,
      "language_loss": 0.76291692,
      "learning_rate": 1.8249556304130254e-06,
      "loss": 0.78506398,
      "num_input_tokens_seen": 97285760,
      "step": 4508,
      "time_per_iteration": 2.6712799072265625
    },
    {
      "auxiliary_loss_clip": 0.0113164,
      "auxiliary_loss_mlp": 0.01030128,
      "balance_loss_clip": 1.05129755,
      "balance_loss_mlp": 1.02187848,
      "epoch": 0.5421751938916611,
      "flos": 29490524519040.0,
      "grad_norm": 2.531539075001217,
      "language_loss": 0.6892153,
      "learning_rate": 1.824179661311044e-06,
      "loss": 0.71083295,
      "num_input_tokens_seen": 97304510,
      "step": 4509,
      "time_per_iteration": 2.8229289054870605
    },
    {
      "auxiliary_loss_clip": 0.01166252,
      "auxiliary_loss_mlp": 0.01033468,
      "balance_loss_clip": 1.05431259,
      "balance_loss_mlp": 1.02504611,
      "epoch": 0.5422954367823003,
      "flos": 18734238311040.0,
      "grad_norm": 2.020897277026257,
      "language_loss": 0.80310708,
      "learning_rate": 1.823403718880868e-06,
      "loss": 0.82510424,
      "num_input_tokens_seen": 97323270,
      "step": 4510,
      "time_per_iteration": 3.6191093921661377
    },
    {
      "auxiliary_loss_clip": 0.01188303,
      "auxiliary_loss_mlp": 0.01029767,
      "balance_loss_clip": 1.05717111,
      "balance_loss_mlp": 1.02116656,
      "epoch": 0.5424156796729394,
      "flos": 39969006940800.0,
      "grad_norm": 2.328289568453737,
      "language_loss": 0.66877073,
      "learning_rate": 1.822627803240207e-06,
      "loss": 0.69095147,
      "num_input_tokens_seen": 97345600,
      "step": 4511,
      "time_per_iteration": 3.612290143966675
    },
    {
      "auxiliary_loss_clip": 0.01176387,
      "auxiliary_loss_mlp": 0.01028008,
      "balance_loss_clip": 1.05610633,
      "balance_loss_mlp": 1.01976478,
      "epoch": 0.5425359225635784,
      "flos": 11546538353280.0,
      "grad_norm": 2.464483662846973,
      "language_loss": 0.85513008,
      "learning_rate": 1.8218519145067675e-06,
      "loss": 0.87717402,
      "num_input_tokens_seen": 97361220,
      "step": 4512,
      "time_per_iteration": 2.703829526901245
    },
    {
      "auxiliary_loss_clip": 0.01188395,
      "auxiliary_loss_mlp": 0.01022071,
      "balance_loss_clip": 1.05650151,
      "balance_loss_mlp": 1.01402402,
      "epoch": 0.5426561654542175,
      "flos": 20229702174720.0,
      "grad_norm": 2.5096633468380167,
      "language_loss": 0.89717621,
      "learning_rate": 1.8210760527982508e-06,
      "loss": 0.91928089,
      "num_input_tokens_seen": 97381505,
      "step": 4513,
      "time_per_iteration": 3.4896793365478516
    },
    {
      "auxiliary_loss_clip": 0.01175313,
      "auxiliary_loss_mlp": 0.01029267,
      "balance_loss_clip": 1.05499148,
      "balance_loss_mlp": 1.0206418,
      "epoch": 0.5427764083448566,
      "flos": 21871681614720.0,
      "grad_norm": 2.168439389408905,
      "language_loss": 0.75325286,
      "learning_rate": 1.8203002182323552e-06,
      "loss": 0.77529866,
      "num_input_tokens_seen": 97399060,
      "step": 4514,
      "time_per_iteration": 2.7118659019470215
    },
    {
      "auxiliary_loss_clip": 0.01173099,
      "auxiliary_loss_mlp": 0.01028926,
      "balance_loss_clip": 1.05569983,
      "balance_loss_mlp": 1.01963341,
      "epoch": 0.5428966512354957,
      "flos": 19640946349440.0,
      "grad_norm": 2.0247665157234276,
      "language_loss": 0.75713611,
      "learning_rate": 1.819524410926773e-06,
      "loss": 0.77915633,
      "num_input_tokens_seen": 97416740,
      "step": 4515,
      "time_per_iteration": 2.6772046089172363
    },
    {
      "auxiliary_loss_clip": 0.01189777,
      "auxiliary_loss_mlp": 0.01031168,
      "balance_loss_clip": 1.05915833,
      "balance_loss_mlp": 1.02286482,
      "epoch": 0.5430168941261347,
      "flos": 22382187661440.0,
      "grad_norm": 2.2505273301908253,
      "language_loss": 0.77115786,
      "learning_rate": 1.8187486309991944e-06,
      "loss": 0.79336727,
      "num_input_tokens_seen": 97437620,
      "step": 4516,
      "time_per_iteration": 2.6900675296783447
    },
    {
      "auxiliary_loss_clip": 0.01187964,
      "auxiliary_loss_mlp": 0.01034004,
      "balance_loss_clip": 1.05744958,
      "balance_loss_mlp": 1.02540874,
      "epoch": 0.5431371370167739,
      "flos": 18764187275520.0,
      "grad_norm": 2.0198894292234106,
      "language_loss": 0.77739114,
      "learning_rate": 1.817972878567304e-06,
      "loss": 0.79961085,
      "num_input_tokens_seen": 97456275,
      "step": 4517,
      "time_per_iteration": 2.66072416305542
    },
    {
      "auxiliary_loss_clip": 0.01164492,
      "auxiliary_loss_mlp": 0.01035899,
      "balance_loss_clip": 1.05478644,
      "balance_loss_mlp": 1.02779889,
      "epoch": 0.543257379907413,
      "flos": 18806023641600.0,
      "grad_norm": 1.817636373960645,
      "language_loss": 0.76592469,
      "learning_rate": 1.8171971537487834e-06,
      "loss": 0.78792858,
      "num_input_tokens_seen": 97474925,
      "step": 4518,
      "time_per_iteration": 2.728332042694092
    },
    {
      "auxiliary_loss_clip": 0.01178622,
      "auxiliary_loss_mlp": 0.01031575,
      "balance_loss_clip": 1.05522954,
      "balance_loss_mlp": 1.02301586,
      "epoch": 0.543377622798052,
      "flos": 17493381025920.0,
      "grad_norm": 2.144612052710882,
      "language_loss": 0.80779988,
      "learning_rate": 1.8164214566613093e-06,
      "loss": 0.82990193,
      "num_input_tokens_seen": 97493550,
      "step": 4519,
      "time_per_iteration": 2.6760995388031006
    },
    {
      "auxiliary_loss_clip": 0.01186859,
      "auxiliary_loss_mlp": 0.01025832,
      "balance_loss_clip": 1.05643582,
      "balance_loss_mlp": 1.01725483,
      "epoch": 0.5434978656886912,
      "flos": 18989311766400.0,
      "grad_norm": 2.821201161199854,
      "language_loss": 0.66213673,
      "learning_rate": 1.8156457874225547e-06,
      "loss": 0.68426371,
      "num_input_tokens_seen": 97512010,
      "step": 4520,
      "time_per_iteration": 2.7053613662719727
    },
    {
      "auxiliary_loss_clip": 0.01161658,
      "auxiliary_loss_mlp": 0.01034226,
      "balance_loss_clip": 1.0517081,
      "balance_loss_mlp": 1.02563715,
      "epoch": 0.5436181085793302,
      "flos": 17274936464640.0,
      "grad_norm": 1.8490156365357968,
      "language_loss": 0.80775428,
      "learning_rate": 1.814870146150187e-06,
      "loss": 0.82971311,
      "num_input_tokens_seen": 97530120,
      "step": 4521,
      "time_per_iteration": 2.6647346019744873
    },
    {
      "auxiliary_loss_clip": 0.01190013,
      "auxiliary_loss_mlp": 0.01027273,
      "balance_loss_clip": 1.05753541,
      "balance_loss_mlp": 1.01792121,
      "epoch": 0.5437383514699693,
      "flos": 19098587917440.0,
      "grad_norm": 1.9650125514962724,
      "language_loss": 0.78496891,
      "learning_rate": 1.814094532961871e-06,
      "loss": 0.80714178,
      "num_input_tokens_seen": 97548695,
      "step": 4522,
      "time_per_iteration": 3.5597524642944336
    },
    {
      "auxiliary_loss_clip": 0.0118036,
      "auxiliary_loss_mlp": 0.01030144,
      "balance_loss_clip": 1.05639696,
      "balance_loss_mlp": 1.021734,
      "epoch": 0.5438585943606085,
      "flos": 22602715211520.0,
      "grad_norm": 2.288294511381094,
      "language_loss": 0.83165735,
      "learning_rate": 1.8133189479752666e-06,
      "loss": 0.85376239,
      "num_input_tokens_seen": 97567625,
      "step": 4523,
      "time_per_iteration": 2.716317653656006
    },
    {
      "auxiliary_loss_clip": 0.01175633,
      "auxiliary_loss_mlp": 0.01034412,
      "balance_loss_clip": 1.05542302,
      "balance_loss_mlp": 1.02577567,
      "epoch": 0.5439788372512475,
      "flos": 21798495653760.0,
      "grad_norm": 2.52822713513095,
      "language_loss": 0.82294059,
      "learning_rate": 1.8125433913080292e-06,
      "loss": 0.84504104,
      "num_input_tokens_seen": 97585325,
      "step": 4524,
      "time_per_iteration": 2.6161577701568604
    },
    {
      "auxiliary_loss_clip": 0.01184931,
      "auxiliary_loss_mlp": 0.01026826,
      "balance_loss_clip": 1.05660915,
      "balance_loss_mlp": 1.0193156,
      "epoch": 0.5440990801418866,
      "flos": 16399362539520.0,
      "grad_norm": 2.1949718872684834,
      "language_loss": 0.83225608,
      "learning_rate": 1.811767863077811e-06,
      "loss": 0.85437357,
      "num_input_tokens_seen": 97604275,
      "step": 4525,
      "time_per_iteration": 2.652965784072876
    },
    {
      "auxiliary_loss_clip": 0.01162076,
      "auxiliary_loss_mlp": 0.01031081,
      "balance_loss_clip": 1.05065167,
      "balance_loss_mlp": 1.02284968,
      "epoch": 0.5442193230325257,
      "flos": 21615638492160.0,
      "grad_norm": 2.007145696839095,
      "language_loss": 0.78529847,
      "learning_rate": 1.8109923634022577e-06,
      "loss": 0.80723011,
      "num_input_tokens_seen": 97624300,
      "step": 4526,
      "time_per_iteration": 2.610323905944824
    },
    {
      "auxiliary_loss_clip": 0.01174458,
      "auxiliary_loss_mlp": 0.01028882,
      "balance_loss_clip": 1.05286586,
      "balance_loss_mlp": 1.02062058,
      "epoch": 0.5443395659231648,
      "flos": 15481198062720.0,
      "grad_norm": 2.4430899619163307,
      "language_loss": 0.86622566,
      "learning_rate": 1.8102168923990128e-06,
      "loss": 0.88825905,
      "num_input_tokens_seen": 97637845,
      "step": 4527,
      "time_per_iteration": 2.6050174236297607
    },
    {
      "auxiliary_loss_clip": 0.01188708,
      "auxiliary_loss_mlp": 0.00762497,
      "balance_loss_clip": 1.05838621,
      "balance_loss_mlp": 1.00175619,
      "epoch": 0.5444598088138038,
      "flos": 18770436241920.0,
      "grad_norm": 2.521985562841782,
      "language_loss": 0.80132926,
      "learning_rate": 1.809441450185714e-06,
      "loss": 0.82084125,
      "num_input_tokens_seen": 97656330,
      "step": 4528,
      "time_per_iteration": 2.6252694129943848
    },
    {
      "auxiliary_loss_clip": 0.01187383,
      "auxiliary_loss_mlp": 0.01025745,
      "balance_loss_clip": 1.05660427,
      "balance_loss_mlp": 1.01753795,
      "epoch": 0.544580051704443,
      "flos": 21142335957120.0,
      "grad_norm": 4.071378942409507,
      "language_loss": 0.73396504,
      "learning_rate": 1.8086660368799958e-06,
      "loss": 0.75609636,
      "num_input_tokens_seen": 97674380,
      "step": 4529,
      "time_per_iteration": 2.6343181133270264
    },
    {
      "auxiliary_loss_clip": 0.01190485,
      "auxiliary_loss_mlp": 0.01031065,
      "balance_loss_clip": 1.0568471,
      "balance_loss_mlp": 1.02182627,
      "epoch": 0.5447002945950821,
      "flos": 32491508054400.0,
      "grad_norm": 2.7746003883553243,
      "language_loss": 0.77624393,
      "learning_rate": 1.807890652599488e-06,
      "loss": 0.79845935,
      "num_input_tokens_seen": 97698765,
      "step": 4530,
      "time_per_iteration": 2.7523810863494873
    },
    {
      "auxiliary_loss_clip": 0.01186389,
      "auxiliary_loss_mlp": 0.01030802,
      "balance_loss_clip": 1.05823445,
      "balance_loss_mlp": 1.02324736,
      "epoch": 0.5448205374857211,
      "flos": 11798307757440.0,
      "grad_norm": 2.021398254196531,
      "language_loss": 0.8269583,
      "learning_rate": 1.8071152974618156e-06,
      "loss": 0.84913021,
      "num_input_tokens_seen": 97716565,
      "step": 4531,
      "time_per_iteration": 2.570491313934326
    },
    {
      "auxiliary_loss_clip": 0.01185029,
      "auxiliary_loss_mlp": 0.01027257,
      "balance_loss_clip": 1.05628073,
      "balance_loss_mlp": 1.01894236,
      "epoch": 0.5449407803763603,
      "flos": 24133766474880.0,
      "grad_norm": 3.520608077810108,
      "language_loss": 0.79137814,
      "learning_rate": 1.806339971584599e-06,
      "loss": 0.813501,
      "num_input_tokens_seen": 97733225,
      "step": 4532,
      "time_per_iteration": 2.6725387573242188
    },
    {
      "auxiliary_loss_clip": 0.01150206,
      "auxiliary_loss_mlp": 0.01028773,
      "balance_loss_clip": 1.05374169,
      "balance_loss_mlp": 1.02058363,
      "epoch": 0.5450610232669993,
      "flos": 23258551685760.0,
      "grad_norm": 3.5880656492715794,
      "language_loss": 0.85205287,
      "learning_rate": 1.8055646750854546e-06,
      "loss": 0.87384272,
      "num_input_tokens_seen": 97752735,
      "step": 4533,
      "time_per_iteration": 2.774946451187134
    },
    {
      "auxiliary_loss_clip": 0.01173865,
      "auxiliary_loss_mlp": 0.01042367,
      "balance_loss_clip": 1.05753684,
      "balance_loss_mlp": 1.03374791,
      "epoch": 0.5451812661576384,
      "flos": 17785083375360.0,
      "grad_norm": 2.731273270914073,
      "language_loss": 0.81854463,
      "learning_rate": 1.8047894080819945e-06,
      "loss": 0.84070694,
      "num_input_tokens_seen": 97769985,
      "step": 4534,
      "time_per_iteration": 2.643080949783325
    },
    {
      "auxiliary_loss_clip": 0.01074995,
      "auxiliary_loss_mlp": 0.01004618,
      "balance_loss_clip": 1.01297545,
      "balance_loss_mlp": 1.00365245,
      "epoch": 0.5453015090482776,
      "flos": 71062586513280.0,
      "grad_norm": 0.7199356608652576,
      "language_loss": 0.63147199,
      "learning_rate": 1.8040141706918258e-06,
      "loss": 0.65226811,
      "num_input_tokens_seen": 97831225,
      "step": 4535,
      "time_per_iteration": 3.350853443145752
    },
    {
      "auxiliary_loss_clip": 0.01188816,
      "auxiliary_loss_mlp": 0.01027398,
      "balance_loss_clip": 1.0594089,
      "balance_loss_mlp": 1.01969147,
      "epoch": 0.5454217519389166,
      "flos": 25552201622400.0,
      "grad_norm": 2.8071311698180104,
      "language_loss": 0.77101636,
      "learning_rate": 1.8032389630325525e-06,
      "loss": 0.79317856,
      "num_input_tokens_seen": 97849975,
      "step": 4536,
      "time_per_iteration": 3.6242098808288574
    },
    {
      "auxiliary_loss_clip": 0.01184798,
      "auxiliary_loss_mlp": 0.01031165,
      "balance_loss_clip": 1.05520177,
      "balance_loss_mlp": 1.02314186,
      "epoch": 0.5455419948295557,
      "flos": 23658345037440.0,
      "grad_norm": 1.7518138726253158,
      "language_loss": 0.75507134,
      "learning_rate": 1.8024637852217707e-06,
      "loss": 0.77723098,
      "num_input_tokens_seen": 97869700,
      "step": 4537,
      "time_per_iteration": 3.614280939102173
    },
    {
      "auxiliary_loss_clip": 0.01154817,
      "auxiliary_loss_mlp": 0.0102829,
      "balance_loss_clip": 1.05588937,
      "balance_loss_mlp": 1.02035642,
      "epoch": 0.5456622377201948,
      "flos": 23403989854080.0,
      "grad_norm": 2.0947468413200943,
      "language_loss": 0.84769571,
      "learning_rate": 1.8016886373770766e-06,
      "loss": 0.8695268,
      "num_input_tokens_seen": 97888215,
      "step": 4538,
      "time_per_iteration": 2.7068278789520264
    },
    {
      "auxiliary_loss_clip": 0.01189463,
      "auxiliary_loss_mlp": 0.01035489,
      "balance_loss_clip": 1.05892622,
      "balance_loss_mlp": 1.0269115,
      "epoch": 0.5457824806108339,
      "flos": 23988040997760.0,
      "grad_norm": 1.6430992529352706,
      "language_loss": 0.78785515,
      "learning_rate": 1.8009135196160579e-06,
      "loss": 0.81010461,
      "num_input_tokens_seen": 97907090,
      "step": 4539,
      "time_per_iteration": 2.718641996383667
    },
    {
      "auxiliary_loss_clip": 0.01151373,
      "auxiliary_loss_mlp": 0.01030086,
      "balance_loss_clip": 1.05215466,
      "balance_loss_mlp": 1.02252853,
      "epoch": 0.545902723501473,
      "flos": 22565870835840.0,
      "grad_norm": 1.8099778099938997,
      "language_loss": 0.84200245,
      "learning_rate": 1.8001384320563e-06,
      "loss": 0.8638171,
      "num_input_tokens_seen": 97927345,
      "step": 4540,
      "time_per_iteration": 3.763136863708496
    },
    {
      "auxiliary_loss_clip": 0.01074695,
      "auxiliary_loss_mlp": 0.01000826,
      "balance_loss_clip": 1.01249218,
      "balance_loss_mlp": 0.99983644,
      "epoch": 0.5460229663921121,
      "flos": 55198399685760.0,
      "grad_norm": 0.769119163191617,
      "language_loss": 0.57748902,
      "learning_rate": 1.7993633748153833e-06,
      "loss": 0.59824425,
      "num_input_tokens_seen": 97981950,
      "step": 4541,
      "time_per_iteration": 3.0705790519714355
    },
    {
      "auxiliary_loss_clip": 0.01174683,
      "auxiliary_loss_mlp": 0.01026886,
      "balance_loss_clip": 1.05425942,
      "balance_loss_mlp": 1.01830888,
      "epoch": 0.5461432092827512,
      "flos": 15413866018560.0,
      "grad_norm": 1.9243947959541088,
      "language_loss": 0.73533869,
      "learning_rate": 1.7985883480108834e-06,
      "loss": 0.75735438,
      "num_input_tokens_seen": 97999585,
      "step": 4542,
      "time_per_iteration": 2.6936886310577393
    },
    {
      "auxiliary_loss_clip": 0.01174271,
      "auxiliary_loss_mlp": 0.01030526,
      "balance_loss_clip": 1.05641162,
      "balance_loss_mlp": 1.0221101,
      "epoch": 0.5462634521733902,
      "flos": 24024921287040.0,
      "grad_norm": 2.4583715283374095,
      "language_loss": 0.72273952,
      "learning_rate": 1.797813351760371e-06,
      "loss": 0.74478751,
      "num_input_tokens_seen": 98021290,
      "step": 4543,
      "time_per_iteration": 2.705632448196411
    },
    {
      "auxiliary_loss_clip": 0.01170468,
      "auxiliary_loss_mlp": 0.01030468,
      "balance_loss_clip": 1.05442691,
      "balance_loss_mlp": 1.02230263,
      "epoch": 0.5463836950640293,
      "flos": 22820944291200.0,
      "grad_norm": 1.987339112381887,
      "language_loss": 0.78377438,
      "learning_rate": 1.7970383861814116e-06,
      "loss": 0.80578375,
      "num_input_tokens_seen": 98041060,
      "step": 4544,
      "time_per_iteration": 2.667691946029663
    },
    {
      "auxiliary_loss_clip": 0.01160825,
      "auxiliary_loss_mlp": 0.01033377,
      "balance_loss_clip": 1.05420876,
      "balance_loss_mlp": 1.02415013,
      "epoch": 0.5465039379546685,
      "flos": 20448290390400.0,
      "grad_norm": 2.6590209694811078,
      "language_loss": 0.73964381,
      "learning_rate": 1.7962634513915684e-06,
      "loss": 0.76158583,
      "num_input_tokens_seen": 98058410,
      "step": 4545,
      "time_per_iteration": 2.732707977294922
    },
    {
      "auxiliary_loss_clip": 0.01175288,
      "auxiliary_loss_mlp": 0.01031459,
      "balance_loss_clip": 1.05544472,
      "balance_loss_mlp": 1.02342701,
      "epoch": 0.5466241808453075,
      "flos": 17343310003200.0,
      "grad_norm": 1.7580009405084007,
      "language_loss": 0.79479957,
      "learning_rate": 1.7954885475083969e-06,
      "loss": 0.81686705,
      "num_input_tokens_seen": 98076080,
      "step": 4546,
      "time_per_iteration": 2.684492826461792
    },
    {
      "auxiliary_loss_clip": 0.01180383,
      "auxiliary_loss_mlp": 0.01031539,
      "balance_loss_clip": 1.05737603,
      "balance_loss_mlp": 1.02269948,
      "epoch": 0.5467444237359466,
      "flos": 21617039122560.0,
      "grad_norm": 2.8546539677051013,
      "language_loss": 0.73189211,
      "learning_rate": 1.7947136746494513e-06,
      "loss": 0.75401133,
      "num_input_tokens_seen": 98096995,
      "step": 4547,
      "time_per_iteration": 2.734624147415161
    },
    {
      "auxiliary_loss_clip": 0.01187443,
      "auxiliary_loss_mlp": 0.01027895,
      "balance_loss_clip": 1.05755091,
      "balance_loss_mlp": 1.01901996,
      "epoch": 0.5468646666265857,
      "flos": 24170467196160.0,
      "grad_norm": 2.1492724743142473,
      "language_loss": 0.88002825,
      "learning_rate": 1.793938832932277e-06,
      "loss": 0.90218163,
      "num_input_tokens_seen": 98115105,
      "step": 4548,
      "time_per_iteration": 3.547907590866089
    },
    {
      "auxiliary_loss_clip": 0.01159632,
      "auxiliary_loss_mlp": 0.01028071,
      "balance_loss_clip": 1.05397832,
      "balance_loss_mlp": 1.01991677,
      "epoch": 0.5469849095172248,
      "flos": 27527001505920.0,
      "grad_norm": 2.310009296970489,
      "language_loss": 0.70849919,
      "learning_rate": 1.7931640224744185e-06,
      "loss": 0.73037618,
      "num_input_tokens_seen": 98135655,
      "step": 4549,
      "time_per_iteration": 2.7352168560028076
    },
    {
      "auxiliary_loss_clip": 0.01171295,
      "auxiliary_loss_mlp": 0.01027892,
      "balance_loss_clip": 1.05131972,
      "balance_loss_mlp": 1.01951194,
      "epoch": 0.5471051524078638,
      "flos": 27964680727680.0,
      "grad_norm": 2.2911125649302844,
      "language_loss": 0.73462629,
      "learning_rate": 1.7923892433934127e-06,
      "loss": 0.75661814,
      "num_input_tokens_seen": 98156730,
      "step": 4550,
      "time_per_iteration": 2.6906259059906006
    },
    {
      "auxiliary_loss_clip": 0.011892,
      "auxiliary_loss_mlp": 0.01029277,
      "balance_loss_clip": 1.05911124,
      "balance_loss_mlp": 1.02092695,
      "epoch": 0.547225395298503,
      "flos": 18150510389760.0,
      "grad_norm": 2.1253738705422935,
      "language_loss": 0.79014397,
      "learning_rate": 1.7916144958067939e-06,
      "loss": 0.81232876,
      "num_input_tokens_seen": 98174590,
      "step": 4551,
      "time_per_iteration": 2.6668686866760254
    },
    {
      "auxiliary_loss_clip": 0.01186358,
      "auxiliary_loss_mlp": 0.010335,
      "balance_loss_clip": 1.05575919,
      "balance_loss_mlp": 1.02567971,
      "epoch": 0.5473456381891421,
      "flos": 21361498790400.0,
      "grad_norm": 1.7388002278539487,
      "language_loss": 0.79075474,
      "learning_rate": 1.7908397798320905e-06,
      "loss": 0.81295335,
      "num_input_tokens_seen": 98194325,
      "step": 4552,
      "time_per_iteration": 2.588613510131836
    },
    {
      "auxiliary_loss_clip": 0.01179258,
      "auxiliary_loss_mlp": 0.0103034,
      "balance_loss_clip": 1.05978012,
      "balance_loss_mlp": 1.02181673,
      "epoch": 0.5474658810797811,
      "flos": 19932145908480.0,
      "grad_norm": 1.7461948639361342,
      "language_loss": 0.7484436,
      "learning_rate": 1.7900650955868265e-06,
      "loss": 0.77053958,
      "num_input_tokens_seen": 98213970,
      "step": 4553,
      "time_per_iteration": 2.6983096599578857
    },
    {
      "auxiliary_loss_clip": 0.01171623,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 1.05849028,
      "balance_loss_mlp": 1.02268577,
      "epoch": 0.5475861239704203,
      "flos": 50476217264640.0,
      "grad_norm": 1.4211433700959284,
      "language_loss": 0.76897138,
      "learning_rate": 1.7892904431885202e-06,
      "loss": 0.79099596,
      "num_input_tokens_seen": 98241145,
      "step": 4554,
      "time_per_iteration": 2.933889865875244
    },
    {
      "auxiliary_loss_clip": 0.011818,
      "auxiliary_loss_mlp": 0.01031695,
      "balance_loss_clip": 1.05524528,
      "balance_loss_mlp": 1.02369583,
      "epoch": 0.5477063668610593,
      "flos": 20705123612160.0,
      "grad_norm": 1.8838045508984682,
      "language_loss": 0.75432533,
      "learning_rate": 1.788515822754686e-06,
      "loss": 0.77646029,
      "num_input_tokens_seen": 98261565,
      "step": 4555,
      "time_per_iteration": 2.6515450477600098
    },
    {
      "auxiliary_loss_clip": 0.01174442,
      "auxiliary_loss_mlp": 0.01034009,
      "balance_loss_clip": 1.05554295,
      "balance_loss_mlp": 1.02529502,
      "epoch": 0.5478266097516984,
      "flos": 19609740408960.0,
      "grad_norm": 2.5131353832028345,
      "language_loss": 0.78459799,
      "learning_rate": 1.7877412344028335e-06,
      "loss": 0.80668253,
      "num_input_tokens_seen": 98281370,
      "step": 4556,
      "time_per_iteration": 2.7307209968566895
    },
    {
      "auxiliary_loss_clip": 0.01177925,
      "auxiliary_loss_mlp": 0.01030583,
      "balance_loss_clip": 1.05620754,
      "balance_loss_mlp": 1.02182698,
      "epoch": 0.5479468526423376,
      "flos": 12896599962240.0,
      "grad_norm": 2.8237881137288707,
      "language_loss": 0.77686298,
      "learning_rate": 1.7869666782504668e-06,
      "loss": 0.79894805,
      "num_input_tokens_seen": 98297950,
      "step": 4557,
      "time_per_iteration": 2.6135363578796387
    },
    {
      "auxiliary_loss_clip": 0.0115365,
      "auxiliary_loss_mlp": 0.01026831,
      "balance_loss_clip": 1.04767632,
      "balance_loss_mlp": 1.01849854,
      "epoch": 0.5480670955329766,
      "flos": 18588800142720.0,
      "grad_norm": 2.3478033558686078,
      "language_loss": 0.69224995,
      "learning_rate": 1.7861921544150867e-06,
      "loss": 0.71405482,
      "num_input_tokens_seen": 98316800,
      "step": 4558,
      "time_per_iteration": 2.6974995136260986
    },
    {
      "auxiliary_loss_clip": 0.01174264,
      "auxiliary_loss_mlp": 0.01037988,
      "balance_loss_clip": 1.0547533,
      "balance_loss_mlp": 1.02967346,
      "epoch": 0.5481873384236157,
      "flos": 15954608338560.0,
      "grad_norm": 2.622803141161719,
      "language_loss": 0.76708758,
      "learning_rate": 1.7854176630141856e-06,
      "loss": 0.78921014,
      "num_input_tokens_seen": 98333935,
      "step": 4559,
      "time_per_iteration": 2.6508615016937256
    },
    {
      "auxiliary_loss_clip": 0.01188491,
      "auxiliary_loss_mlp": 0.01034575,
      "balance_loss_clip": 1.05680382,
      "balance_loss_mlp": 1.02553916,
      "epoch": 0.5483075813142548,
      "flos": 22783812606720.0,
      "grad_norm": 3.6041331278627307,
      "language_loss": 0.84550458,
      "learning_rate": 1.784643204165255e-06,
      "loss": 0.86773527,
      "num_input_tokens_seen": 98353255,
      "step": 4560,
      "time_per_iteration": 2.6715664863586426
    },
    {
      "auxiliary_loss_clip": 0.01185139,
      "auxiliary_loss_mlp": 0.01029546,
      "balance_loss_clip": 1.05689692,
      "balance_loss_mlp": 1.02118897,
      "epoch": 0.5484278242048939,
      "flos": 19317212046720.0,
      "grad_norm": 2.6407755744079116,
      "language_loss": 0.77163857,
      "learning_rate": 1.7838687779857783e-06,
      "loss": 0.79378539,
      "num_input_tokens_seen": 98371130,
      "step": 4561,
      "time_per_iteration": 2.6344962120056152
    },
    {
      "auxiliary_loss_clip": 0.01152924,
      "auxiliary_loss_mlp": 0.01031489,
      "balance_loss_clip": 1.05484509,
      "balance_loss_mlp": 1.02312684,
      "epoch": 0.5485480670955329,
      "flos": 22816024128000.0,
      "grad_norm": 2.061858389387995,
      "language_loss": 0.6393522,
      "learning_rate": 1.7830943845932366e-06,
      "loss": 0.66119635,
      "num_input_tokens_seen": 98390455,
      "step": 4562,
      "time_per_iteration": 2.801018714904785
    },
    {
      "auxiliary_loss_clip": 0.01155879,
      "auxiliary_loss_mlp": 0.01030062,
      "balance_loss_clip": 1.05310369,
      "balance_loss_mlp": 1.02192247,
      "epoch": 0.5486683099861721,
      "flos": 22671304231680.0,
      "grad_norm": 2.542069332963762,
      "language_loss": 0.74873704,
      "learning_rate": 1.7823200241051044e-06,
      "loss": 0.77059639,
      "num_input_tokens_seen": 98409370,
      "step": 4563,
      "time_per_iteration": 3.619872808456421
    },
    {
      "auxiliary_loss_clip": 0.01173608,
      "auxiliary_loss_mlp": 0.01032759,
      "balance_loss_clip": 1.05628133,
      "balance_loss_mlp": 1.02465248,
      "epoch": 0.5487885528768112,
      "flos": 23149383275520.0,
      "grad_norm": 2.1177858541529373,
      "language_loss": 0.80385447,
      "learning_rate": 1.7815456966388513e-06,
      "loss": 0.82591814,
      "num_input_tokens_seen": 98428465,
      "step": 4564,
      "time_per_iteration": 3.834592819213867
    },
    {
      "auxiliary_loss_clip": 0.01144781,
      "auxiliary_loss_mlp": 0.01026485,
      "balance_loss_clip": 1.04971802,
      "balance_loss_mlp": 1.01839709,
      "epoch": 0.5489087957674502,
      "flos": 22053928245120.0,
      "grad_norm": 2.6065285213586993,
      "language_loss": 0.81115127,
      "learning_rate": 1.780771402311943e-06,
      "loss": 0.83286393,
      "num_input_tokens_seen": 98447300,
      "step": 4565,
      "time_per_iteration": 3.8323445320129395
    },
    {
      "auxiliary_loss_clip": 0.011655,
      "auxiliary_loss_mlp": 0.01035571,
      "balance_loss_clip": 1.05369282,
      "balance_loss_mlp": 1.02698851,
      "epoch": 0.5490290386580894,
      "flos": 24315977191680.0,
      "grad_norm": 1.6733612040255215,
      "language_loss": 0.7869764,
      "learning_rate": 1.7799971412418374e-06,
      "loss": 0.80898714,
      "num_input_tokens_seen": 98468695,
      "step": 4566,
      "time_per_iteration": 2.7366771697998047
    },
    {
      "auxiliary_loss_clip": 0.0117365,
      "auxiliary_loss_mlp": 0.01033797,
      "balance_loss_clip": 1.0551995,
      "balance_loss_mlp": 1.02488303,
      "epoch": 0.5491492815487284,
      "flos": 18294942977280.0,
      "grad_norm": 2.255374940167123,
      "language_loss": 0.73935705,
      "learning_rate": 1.7792229135459918e-06,
      "loss": 0.76143157,
      "num_input_tokens_seen": 98485345,
      "step": 4567,
      "time_per_iteration": 2.6369614601135254
    },
    {
      "auxiliary_loss_clip": 0.01070949,
      "auxiliary_loss_mlp": 0.01007168,
      "balance_loss_clip": 1.0094564,
      "balance_loss_mlp": 1.00619686,
      "epoch": 0.5492695244393675,
      "flos": 64550257050240.0,
      "grad_norm": 0.7421505551917356,
      "language_loss": 0.61517358,
      "learning_rate": 1.7784487193418538e-06,
      "loss": 0.63595474,
      "num_input_tokens_seen": 98543195,
      "step": 4568,
      "time_per_iteration": 3.132174015045166
    },
    {
      "auxiliary_loss_clip": 0.01173135,
      "auxiliary_loss_mlp": 0.01029719,
      "balance_loss_clip": 1.05344105,
      "balance_loss_mlp": 1.02114785,
      "epoch": 0.5493897673300067,
      "flos": 17379579761280.0,
      "grad_norm": 2.944090896931527,
      "language_loss": 0.61671889,
      "learning_rate": 1.7776745587468698e-06,
      "loss": 0.63874733,
      "num_input_tokens_seen": 98560620,
      "step": 4569,
      "time_per_iteration": 2.633326530456543
    },
    {
      "auxiliary_loss_clip": 0.01183935,
      "auxiliary_loss_mlp": 0.01024642,
      "balance_loss_clip": 1.05465889,
      "balance_loss_mlp": 1.01655936,
      "epoch": 0.5495100102206457,
      "flos": 19901765980800.0,
      "grad_norm": 3.0558739875874834,
      "language_loss": 0.82075727,
      "learning_rate": 1.7769004318784776e-06,
      "loss": 0.842843,
      "num_input_tokens_seen": 98578265,
      "step": 4570,
      "time_per_iteration": 2.660600185394287
    },
    {
      "auxiliary_loss_clip": 0.01154537,
      "auxiliary_loss_mlp": 0.01028451,
      "balance_loss_clip": 1.04901063,
      "balance_loss_mlp": 1.02000761,
      "epoch": 0.5496302531112848,
      "flos": 16727190992640.0,
      "grad_norm": 1.6375764914788002,
      "language_loss": 0.80752087,
      "learning_rate": 1.776126338854113e-06,
      "loss": 0.82935077,
      "num_input_tokens_seen": 98596055,
      "step": 4571,
      "time_per_iteration": 2.6878130435943604
    },
    {
      "auxiliary_loss_clip": 0.01167325,
      "auxiliary_loss_mlp": 0.01030576,
      "balance_loss_clip": 1.05292106,
      "balance_loss_mlp": 1.02312541,
      "epoch": 0.5497504960019239,
      "flos": 24572343536640.0,
      "grad_norm": 5.098113270389135,
      "language_loss": 0.84868747,
      "learning_rate": 1.7753522797912044e-06,
      "loss": 0.8706665,
      "num_input_tokens_seen": 98616140,
      "step": 4572,
      "time_per_iteration": 2.700784683227539
    },
    {
      "auxiliary_loss_clip": 0.0117137,
      "auxiliary_loss_mlp": 0.01028504,
      "balance_loss_clip": 1.05412579,
      "balance_loss_mlp": 1.02026677,
      "epoch": 0.549870738892563,
      "flos": 15450494912640.0,
      "grad_norm": 2.0877355761705427,
      "language_loss": 0.70037389,
      "learning_rate": 1.7745782548071765e-06,
      "loss": 0.72237265,
      "num_input_tokens_seen": 98633035,
      "step": 4573,
      "time_per_iteration": 2.608778476715088
    },
    {
      "auxiliary_loss_clip": 0.01184667,
      "auxiliary_loss_mlp": 0.01035122,
      "balance_loss_clip": 1.05692172,
      "balance_loss_mlp": 1.02700996,
      "epoch": 0.549990981783202,
      "flos": 21069114082560.0,
      "grad_norm": 2.074082916753332,
      "language_loss": 0.74330652,
      "learning_rate": 1.7738042640194482e-06,
      "loss": 0.76550442,
      "num_input_tokens_seen": 98652700,
      "step": 4574,
      "time_per_iteration": 3.5841495990753174
    },
    {
      "auxiliary_loss_clip": 0.01146135,
      "auxiliary_loss_mlp": 0.01031883,
      "balance_loss_clip": 1.04854107,
      "balance_loss_mlp": 1.02334154,
      "epoch": 0.5501112246738411,
      "flos": 21395901041280.0,
      "grad_norm": 2.048639308425984,
      "language_loss": 0.70544112,
      "learning_rate": 1.7730303075454335e-06,
      "loss": 0.72722125,
      "num_input_tokens_seen": 98671590,
      "step": 4575,
      "time_per_iteration": 2.7600865364074707
    },
    {
      "auxiliary_loss_clip": 0.01188944,
      "auxiliary_loss_mlp": 0.0103482,
      "balance_loss_clip": 1.05863333,
      "balance_loss_mlp": 1.02612412,
      "epoch": 0.5502314675644803,
      "flos": 17456931699840.0,
      "grad_norm": 1.9920106194158693,
      "language_loss": 0.84600186,
      "learning_rate": 1.7722563855025402e-06,
      "loss": 0.86823952,
      "num_input_tokens_seen": 98689620,
      "step": 4576,
      "time_per_iteration": 2.5253026485443115
    },
    {
      "auxiliary_loss_clip": 0.01172826,
      "auxiliary_loss_mlp": 0.01030416,
      "balance_loss_clip": 1.05481458,
      "balance_loss_mlp": 1.02205312,
      "epoch": 0.5503517104551193,
      "flos": 24310410583680.0,
      "grad_norm": 2.3545306868075286,
      "language_loss": 0.70866358,
      "learning_rate": 1.7714824980081721e-06,
      "loss": 0.73069596,
      "num_input_tokens_seen": 98708915,
      "step": 4577,
      "time_per_iteration": 2.6840333938598633
    },
    {
      "auxiliary_loss_clip": 0.01117857,
      "auxiliary_loss_mlp": 0.01026785,
      "balance_loss_clip": 1.04750431,
      "balance_loss_mlp": 1.01877999,
      "epoch": 0.5504719533457584,
      "flos": 22419427086720.0,
      "grad_norm": 1.8755533083470501,
      "language_loss": 0.73898399,
      "learning_rate": 1.7707086451797276e-06,
      "loss": 0.7604304,
      "num_input_tokens_seen": 98729790,
      "step": 4578,
      "time_per_iteration": 2.697046995162964
    },
    {
      "auxiliary_loss_clip": 0.01061942,
      "auxiliary_loss_mlp": 0.0100386,
      "balance_loss_clip": 1.00902796,
      "balance_loss_mlp": 1.00282335,
      "epoch": 0.5505921962363975,
      "flos": 67294155968640.0,
      "grad_norm": 0.700669887619833,
      "language_loss": 0.52246839,
      "learning_rate": 1.7699348271345993e-06,
      "loss": 0.5431264,
      "num_input_tokens_seen": 98792415,
      "step": 4579,
      "time_per_iteration": 3.1350998878479004
    },
    {
      "auxiliary_loss_clip": 0.0105873,
      "auxiliary_loss_mlp": 0.01000104,
      "balance_loss_clip": 1.00902569,
      "balance_loss_mlp": 0.99916232,
      "epoch": 0.5507124391270366,
      "flos": 45685125578880.0,
      "grad_norm": 0.7725424069610678,
      "language_loss": 0.54378432,
      "learning_rate": 1.7691610439901753e-06,
      "loss": 0.56437266,
      "num_input_tokens_seen": 98855350,
      "step": 4580,
      "time_per_iteration": 3.2095673084259033
    },
    {
      "auxiliary_loss_clip": 0.01171257,
      "auxiliary_loss_mlp": 0.01029488,
      "balance_loss_clip": 1.05540788,
      "balance_loss_mlp": 1.02125621,
      "epoch": 0.5508326820176757,
      "flos": 22273845264000.0,
      "grad_norm": 1.8258173160862423,
      "language_loss": 0.76149446,
      "learning_rate": 1.7683872958638367e-06,
      "loss": 0.78350192,
      "num_input_tokens_seen": 98874230,
      "step": 4581,
      "time_per_iteration": 2.6523025035858154
    },
    {
      "auxiliary_loss_clip": 0.01184693,
      "auxiliary_loss_mlp": 0.01027148,
      "balance_loss_clip": 1.05497026,
      "balance_loss_mlp": 1.01930404,
      "epoch": 0.5509529249083148,
      "flos": 20012442762240.0,
      "grad_norm": 2.8091788755382563,
      "language_loss": 0.84543765,
      "learning_rate": 1.7676135828729614e-06,
      "loss": 0.86755598,
      "num_input_tokens_seen": 98893940,
      "step": 4582,
      "time_per_iteration": 2.60081148147583
    },
    {
      "auxiliary_loss_clip": 0.01166799,
      "auxiliary_loss_mlp": 0.01031339,
      "balance_loss_clip": 1.05508685,
      "balance_loss_mlp": 1.02376008,
      "epoch": 0.5510731677989539,
      "flos": 21834801325440.0,
      "grad_norm": 2.3521949870852468,
      "language_loss": 0.82909,
      "learning_rate": 1.7668399051349205e-06,
      "loss": 0.85107142,
      "num_input_tokens_seen": 98913620,
      "step": 4583,
      "time_per_iteration": 2.670168161392212
    },
    {
      "auxiliary_loss_clip": 0.01136057,
      "auxiliary_loss_mlp": 0.01026594,
      "balance_loss_clip": 1.04686093,
      "balance_loss_mlp": 1.01779008,
      "epoch": 0.5511934106895929,
      "flos": 21467901853440.0,
      "grad_norm": 2.2662222806887455,
      "language_loss": 0.83343267,
      "learning_rate": 1.766066262767081e-06,
      "loss": 0.85505927,
      "num_input_tokens_seen": 98931460,
      "step": 4584,
      "time_per_iteration": 2.7138543128967285
    },
    {
      "auxiliary_loss_clip": 0.01183062,
      "auxiliary_loss_mlp": 0.01029166,
      "balance_loss_clip": 1.05531454,
      "balance_loss_mlp": 1.0211463,
      "epoch": 0.5513136535802321,
      "flos": 21068934514560.0,
      "grad_norm": 3.9232645318969586,
      "language_loss": 0.77160585,
      "learning_rate": 1.765292655886803e-06,
      "loss": 0.79372817,
      "num_input_tokens_seen": 98950105,
      "step": 4585,
      "time_per_iteration": 2.5599520206451416
    },
    {
      "auxiliary_loss_clip": 0.01169845,
      "auxiliary_loss_mlp": 0.0103254,
      "balance_loss_clip": 1.05458355,
      "balance_loss_mlp": 1.02386785,
      "epoch": 0.5514338964708712,
      "flos": 27815004754560.0,
      "grad_norm": 4.849322459309791,
      "language_loss": 0.71065748,
      "learning_rate": 1.764519084611443e-06,
      "loss": 0.73268133,
      "num_input_tokens_seen": 98970560,
      "step": 4586,
      "time_per_iteration": 2.7170443534851074
    },
    {
      "auxiliary_loss_clip": 0.01138744,
      "auxiliary_loss_mlp": 0.01028959,
      "balance_loss_clip": 1.05081177,
      "balance_loss_mlp": 1.01997089,
      "epoch": 0.5515541393615102,
      "flos": 21908525990400.0,
      "grad_norm": 11.472999268165108,
      "language_loss": 0.77588129,
      "learning_rate": 1.7637455490583505e-06,
      "loss": 0.79755831,
      "num_input_tokens_seen": 98989885,
      "step": 4587,
      "time_per_iteration": 2.6675961017608643
    },
    {
      "auxiliary_loss_clip": 0.01185646,
      "auxiliary_loss_mlp": 0.0103698,
      "balance_loss_clip": 1.05722916,
      "balance_loss_mlp": 1.02928543,
      "epoch": 0.5516743822521494,
      "flos": 20485422074880.0,
      "grad_norm": 1.9340678746590694,
      "language_loss": 0.77649188,
      "learning_rate": 1.7629720493448701e-06,
      "loss": 0.79871809,
      "num_input_tokens_seen": 99007180,
      "step": 4588,
      "time_per_iteration": 2.6995081901550293
    },
    {
      "auxiliary_loss_clip": 0.01188544,
      "auxiliary_loss_mlp": 0.01039053,
      "balance_loss_clip": 1.05800152,
      "balance_loss_mlp": 1.03012407,
      "epoch": 0.5517946251427884,
      "flos": 14940383915520.0,
      "grad_norm": 2.3747782948325233,
      "language_loss": 0.8539046,
      "learning_rate": 1.7621985855883418e-06,
      "loss": 0.87618053,
      "num_input_tokens_seen": 99023880,
      "step": 4589,
      "time_per_iteration": 3.522578001022339
    },
    {
      "auxiliary_loss_clip": 0.01116889,
      "auxiliary_loss_mlp": 0.01031519,
      "balance_loss_clip": 1.04888749,
      "balance_loss_mlp": 1.02352631,
      "epoch": 0.5519148680334275,
      "flos": 18404865573120.0,
      "grad_norm": 2.4820656734966757,
      "language_loss": 0.72584999,
      "learning_rate": 1.7614251579060983e-06,
      "loss": 0.747334,
      "num_input_tokens_seen": 99042475,
      "step": 4590,
      "time_per_iteration": 3.7000606060028076
    },
    {
      "auxiliary_loss_clip": 0.01177261,
      "auxiliary_loss_mlp": 0.01031342,
      "balance_loss_clip": 1.05409276,
      "balance_loss_mlp": 1.02308059,
      "epoch": 0.5520351109240667,
      "flos": 25113337251840.0,
      "grad_norm": 1.871829961924277,
      "language_loss": 0.8488704,
      "learning_rate": 1.76065176641547e-06,
      "loss": 0.87095642,
      "num_input_tokens_seen": 99065185,
      "step": 4591,
      "time_per_iteration": 2.7916100025177
    },
    {
      "auxiliary_loss_clip": 0.01167106,
      "auxiliary_loss_mlp": 0.01029379,
      "balance_loss_clip": 1.0531106,
      "balance_loss_mlp": 1.02126741,
      "epoch": 0.5521553538147057,
      "flos": 21069545045760.0,
      "grad_norm": 1.6953840092975534,
      "language_loss": 0.77915525,
      "learning_rate": 1.759878411233777e-06,
      "loss": 0.80112004,
      "num_input_tokens_seen": 99083645,
      "step": 4592,
      "time_per_iteration": 3.625448703765869
    },
    {
      "auxiliary_loss_clip": 0.01146434,
      "auxiliary_loss_mlp": 0.0103103,
      "balance_loss_clip": 1.05201817,
      "balance_loss_mlp": 1.02231574,
      "epoch": 0.5522755967053448,
      "flos": 18879999701760.0,
      "grad_norm": 8.493717081413276,
      "language_loss": 0.75824457,
      "learning_rate": 1.7591050924783388e-06,
      "loss": 0.78001922,
      "num_input_tokens_seen": 99100835,
      "step": 4593,
      "time_per_iteration": 0.024780750274658203
    },
    {
      "auxiliary_loss_clip": 0.01071449,
      "auxiliary_loss_mlp": 0.01000524,
      "balance_loss_clip": 1.00967646,
      "balance_loss_mlp": 0.99959433,
      "epoch": 0.5523958395959839,
      "flos": 64675622494080.0,
      "grad_norm": 0.8358009497100678,
      "language_loss": 0.5786109,
      "learning_rate": 1.7583318102664661e-06,
      "loss": 0.5993306,
      "num_input_tokens_seen": 99168400,
      "step": 4594,
      "time_per_iteration": 3.1871490478515625
    },
    {
      "auxiliary_loss_clip": 0.01183372,
      "auxiliary_loss_mlp": 0.01031135,
      "balance_loss_clip": 1.05234885,
      "balance_loss_mlp": 1.02281451,
      "epoch": 0.552516082486623,
      "flos": 10889732211840.0,
      "grad_norm": 1.9364978223429654,
      "language_loss": 0.78994185,
      "learning_rate": 1.757558564715466e-06,
      "loss": 0.81208682,
      "num_input_tokens_seen": 99186475,
      "step": 4595,
      "time_per_iteration": 2.605529546737671
    },
    {
      "auxiliary_loss_clip": 0.01189119,
      "auxiliary_loss_mlp": 0.01031918,
      "balance_loss_clip": 1.05570364,
      "balance_loss_mlp": 1.02337074,
      "epoch": 0.552636325377262,
      "flos": 22199797376640.0,
      "grad_norm": 3.278751216614144,
      "language_loss": 0.74343753,
      "learning_rate": 1.7567853559426386e-06,
      "loss": 0.76564789,
      "num_input_tokens_seen": 99203525,
      "step": 4596,
      "time_per_iteration": 2.635599136352539
    },
    {
      "auxiliary_loss_clip": 0.01162665,
      "auxiliary_loss_mlp": 0.01031024,
      "balance_loss_clip": 1.05112362,
      "balance_loss_mlp": 1.02244675,
      "epoch": 0.5527565682679012,
      "flos": 23988184652160.0,
      "grad_norm": 1.9977075776712936,
      "language_loss": 0.75390637,
      "learning_rate": 1.7560121840652797e-06,
      "loss": 0.77584314,
      "num_input_tokens_seen": 99222910,
      "step": 4597,
      "time_per_iteration": 2.642322301864624
    },
    {
      "auxiliary_loss_clip": 0.01142538,
      "auxiliary_loss_mlp": 0.010278,
      "balance_loss_clip": 1.05426514,
      "balance_loss_mlp": 1.0197053,
      "epoch": 0.5528768111585403,
      "flos": 19719267955200.0,
      "grad_norm": 1.9327934711037094,
      "language_loss": 0.69174361,
      "learning_rate": 1.7552390492006782e-06,
      "loss": 0.71344692,
      "num_input_tokens_seen": 99241230,
      "step": 4598,
      "time_per_iteration": 2.777571439743042
    },
    {
      "auxiliary_loss_clip": 0.01142577,
      "auxiliary_loss_mlp": 0.0103673,
      "balance_loss_clip": 1.05348802,
      "balance_loss_mlp": 1.0268594,
      "epoch": 0.5529970540491793,
      "flos": 26215975002240.0,
      "grad_norm": 2.842232809625961,
      "language_loss": 0.65438181,
      "learning_rate": 1.7544659514661184e-06,
      "loss": 0.67617488,
      "num_input_tokens_seen": 99264320,
      "step": 4599,
      "time_per_iteration": 2.75028133392334
    },
    {
      "auxiliary_loss_clip": 0.0117116,
      "auxiliary_loss_mlp": 0.01027757,
      "balance_loss_clip": 1.05508459,
      "balance_loss_mlp": 1.01906085,
      "epoch": 0.5531172969398185,
      "flos": 24425971614720.0,
      "grad_norm": 2.8382163482274207,
      "language_loss": 0.79926467,
      "learning_rate": 1.7536928909788786e-06,
      "loss": 0.8212539,
      "num_input_tokens_seen": 99283625,
      "step": 4600,
      "time_per_iteration": 3.6473305225372314
    },
    {
      "auxiliary_loss_clip": 0.01071892,
      "auxiliary_loss_mlp": 0.00752738,
      "balance_loss_clip": 1.01015615,
      "balance_loss_mlp": 1.00110948,
      "epoch": 0.5532375398304575,
      "flos": 64907316195840.0,
      "grad_norm": 0.8741115674365986,
      "language_loss": 0.61865687,
      "learning_rate": 1.752919867856231e-06,
      "loss": 0.63690317,
      "num_input_tokens_seen": 99335270,
      "step": 4601,
      "time_per_iteration": 3.0670132637023926
    },
    {
      "auxiliary_loss_clip": 0.01140738,
      "auxiliary_loss_mlp": 0.00762415,
      "balance_loss_clip": 1.04928422,
      "balance_loss_mlp": 1.00138617,
      "epoch": 0.5533577827210966,
      "flos": 19683105937920.0,
      "grad_norm": 1.722968098492528,
      "language_loss": 0.788001,
      "learning_rate": 1.7521468822154436e-06,
      "loss": 0.80703259,
      "num_input_tokens_seen": 99354185,
      "step": 4602,
      "time_per_iteration": 2.7298285961151123
    },
    {
      "auxiliary_loss_clip": 0.01166621,
      "auxiliary_loss_mlp": 0.01026898,
      "balance_loss_clip": 1.05298579,
      "balance_loss_mlp": 1.01895261,
      "epoch": 0.5534780256117358,
      "flos": 32306496076800.0,
      "grad_norm": 1.982683033482708,
      "language_loss": 0.75520575,
      "learning_rate": 1.751373934173777e-06,
      "loss": 0.77714092,
      "num_input_tokens_seen": 99376930,
      "step": 4603,
      "time_per_iteration": 2.769911527633667
    },
    {
      "auxiliary_loss_clip": 0.01189092,
      "auxiliary_loss_mlp": 0.01030177,
      "balance_loss_clip": 1.05595613,
      "balance_loss_mlp": 1.02170086,
      "epoch": 0.5535982685023748,
      "flos": 23222425582080.0,
      "grad_norm": 2.5156196026230537,
      "language_loss": 0.73351741,
      "learning_rate": 1.750601023848487e-06,
      "loss": 0.75571007,
      "num_input_tokens_seen": 99397655,
      "step": 4604,
      "time_per_iteration": 2.592113971710205
    },
    {
      "auxiliary_loss_clip": 0.0117122,
      "auxiliary_loss_mlp": 0.01028668,
      "balance_loss_clip": 1.05471134,
      "balance_loss_mlp": 1.0206089,
      "epoch": 0.5537185113930139,
      "flos": 24352534258560.0,
      "grad_norm": 1.881048707342012,
      "language_loss": 0.73978162,
      "learning_rate": 1.749828151356823e-06,
      "loss": 0.7617805,
      "num_input_tokens_seen": 99417850,
      "step": 4605,
      "time_per_iteration": 2.727707862854004
    },
    {
      "auxiliary_loss_clip": 0.01168652,
      "auxiliary_loss_mlp": 0.01031571,
      "balance_loss_clip": 1.05457199,
      "balance_loss_mlp": 1.02372098,
      "epoch": 0.553838754283653,
      "flos": 23549068886400.0,
      "grad_norm": 1.9033405511291128,
      "language_loss": 0.75885463,
      "learning_rate": 1.7490553168160297e-06,
      "loss": 0.78085685,
      "num_input_tokens_seen": 99438920,
      "step": 4606,
      "time_per_iteration": 2.7258293628692627
    },
    {
      "auxiliary_loss_clip": 0.01187581,
      "auxiliary_loss_mlp": 0.01028847,
      "balance_loss_clip": 1.05646276,
      "balance_loss_mlp": 1.02058578,
      "epoch": 0.5539589971742921,
      "flos": 17275044205440.0,
      "grad_norm": 2.7187160792129315,
      "language_loss": 0.76373577,
      "learning_rate": 1.748282520343345e-06,
      "loss": 0.7859,
      "num_input_tokens_seen": 99457950,
      "step": 4607,
      "time_per_iteration": 2.6797964572906494
    },
    {
      "auxiliary_loss_clip": 0.01164982,
      "auxiliary_loss_mlp": 0.01030533,
      "balance_loss_clip": 1.05473042,
      "balance_loss_mlp": 1.02129984,
      "epoch": 0.5540792400649311,
      "flos": 27564169104000.0,
      "grad_norm": 1.863339874259222,
      "language_loss": 0.78640413,
      "learning_rate": 1.7475097620560023e-06,
      "loss": 0.80835927,
      "num_input_tokens_seen": 99478015,
      "step": 4608,
      "time_per_iteration": 2.7242894172668457
    },
    {
      "auxiliary_loss_clip": 0.01171289,
      "auxiliary_loss_mlp": 0.0103216,
      "balance_loss_clip": 1.05216312,
      "balance_loss_mlp": 1.0236783,
      "epoch": 0.5541994829555702,
      "flos": 23878657105920.0,
      "grad_norm": 1.7966054451763254,
      "language_loss": 0.71204501,
      "learning_rate": 1.746737042071228e-06,
      "loss": 0.73407948,
      "num_input_tokens_seen": 99496520,
      "step": 4609,
      "time_per_iteration": 2.6597065925598145
    },
    {
      "auxiliary_loss_clip": 0.01152882,
      "auxiliary_loss_mlp": 0.0102756,
      "balance_loss_clip": 1.05313706,
      "balance_loss_mlp": 1.01919723,
      "epoch": 0.5543197258462094,
      "flos": 20115721342080.0,
      "grad_norm": 1.7889057530399959,
      "language_loss": 0.79166508,
      "learning_rate": 1.7459643605062424e-06,
      "loss": 0.81346947,
      "num_input_tokens_seen": 99513780,
      "step": 4610,
      "time_per_iteration": 2.696200132369995
    },
    {
      "auxiliary_loss_clip": 0.0117225,
      "auxiliary_loss_mlp": 0.01026921,
      "balance_loss_clip": 1.05442953,
      "balance_loss_mlp": 1.01858258,
      "epoch": 0.5544399687368484,
      "flos": 20916565021440.0,
      "grad_norm": 2.1721032826425786,
      "language_loss": 0.80980468,
      "learning_rate": 1.745191717478262e-06,
      "loss": 0.83179641,
      "num_input_tokens_seen": 99532360,
      "step": 4611,
      "time_per_iteration": 2.6625471115112305
    },
    {
      "auxiliary_loss_clip": 0.01161706,
      "auxiliary_loss_mlp": 0.01026334,
      "balance_loss_clip": 1.05088842,
      "balance_loss_mlp": 1.01831698,
      "epoch": 0.5545602116274875,
      "flos": 25518661297920.0,
      "grad_norm": 1.7642309341242837,
      "language_loss": 0.79609519,
      "learning_rate": 1.7444191131044948e-06,
      "loss": 0.81797558,
      "num_input_tokens_seen": 99552635,
      "step": 4612,
      "time_per_iteration": 2.779717206954956
    },
    {
      "auxiliary_loss_clip": 0.01145353,
      "auxiliary_loss_mlp": 0.01034018,
      "balance_loss_clip": 1.04973483,
      "balance_loss_mlp": 1.02520227,
      "epoch": 0.5546804545181266,
      "flos": 20995568985600.0,
      "grad_norm": 2.061214629602627,
      "language_loss": 0.73409331,
      "learning_rate": 1.7436465475021456e-06,
      "loss": 0.75588703,
      "num_input_tokens_seen": 99572685,
      "step": 4613,
      "time_per_iteration": 2.677849292755127
    },
    {
      "auxiliary_loss_clip": 0.01168836,
      "auxiliary_loss_mlp": 0.01028784,
      "balance_loss_clip": 1.05430639,
      "balance_loss_mlp": 1.02027857,
      "epoch": 0.5548006974087657,
      "flos": 26833638297600.0,
      "grad_norm": 2.097048359808914,
      "language_loss": 0.71807235,
      "learning_rate": 1.7428740207884111e-06,
      "loss": 0.74004853,
      "num_input_tokens_seen": 99593565,
      "step": 4614,
      "time_per_iteration": 2.7380592823028564
    },
    {
      "auxiliary_loss_clip": 0.01185545,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.05501437,
      "balance_loss_mlp": 1.02004707,
      "epoch": 0.5549209402994048,
      "flos": 33656414031360.0,
      "grad_norm": 1.7336777096582185,
      "language_loss": 0.6113584,
      "learning_rate": 1.7421015330804833e-06,
      "loss": 0.63350117,
      "num_input_tokens_seen": 99613485,
      "step": 4615,
      "time_per_iteration": 3.682499647140503
    },
    {
      "auxiliary_loss_clip": 0.01130731,
      "auxiliary_loss_mlp": 0.01028804,
      "balance_loss_clip": 1.04894447,
      "balance_loss_mlp": 1.02032876,
      "epoch": 0.5550411831900439,
      "flos": 23769524609280.0,
      "grad_norm": 2.036803195242137,
      "language_loss": 0.72530735,
      "learning_rate": 1.7413290844955475e-06,
      "loss": 0.7469027,
      "num_input_tokens_seen": 99633515,
      "step": 4616,
      "time_per_iteration": 2.736964464187622
    },
    {
      "auxiliary_loss_clip": 0.01187395,
      "auxiliary_loss_mlp": 0.01027927,
      "balance_loss_clip": 1.05822468,
      "balance_loss_mlp": 1.01946318,
      "epoch": 0.555161426080683,
      "flos": 21651189978240.0,
      "grad_norm": 1.9234257855541954,
      "language_loss": 0.78595209,
      "learning_rate": 1.7405566751507843e-06,
      "loss": 0.80810535,
      "num_input_tokens_seen": 99651560,
      "step": 4617,
      "time_per_iteration": 2.603182792663574
    },
    {
      "auxiliary_loss_clip": 0.01167627,
      "auxiliary_loss_mlp": 0.01032286,
      "balance_loss_clip": 1.0531801,
      "balance_loss_mlp": 1.02355373,
      "epoch": 0.555281668971322,
      "flos": 49563116605440.0,
      "grad_norm": 1.9854382845953247,
      "language_loss": 0.67969608,
      "learning_rate": 1.7397843051633668e-06,
      "loss": 0.7016952,
      "num_input_tokens_seen": 99674255,
      "step": 4618,
      "time_per_iteration": 3.7887704372406006
    },
    {
      "auxiliary_loss_clip": 0.01170572,
      "auxiliary_loss_mlp": 0.01027838,
      "balance_loss_clip": 1.055004,
      "balance_loss_mlp": 1.01961255,
      "epoch": 0.5554019118619612,
      "flos": 20741608851840.0,
      "grad_norm": 2.1716440076651744,
      "language_loss": 0.71808994,
      "learning_rate": 1.739011974650464e-06,
      "loss": 0.74007404,
      "num_input_tokens_seen": 99693585,
      "step": 4619,
      "time_per_iteration": 2.705832004547119
    },
    {
      "auxiliary_loss_clip": 0.01190057,
      "auxiliary_loss_mlp": 0.01032474,
      "balance_loss_clip": 1.05498266,
      "balance_loss_mlp": 1.02295494,
      "epoch": 0.5555221547526003,
      "flos": 25483217552640.0,
      "grad_norm": 2.344718399256129,
      "language_loss": 0.76669818,
      "learning_rate": 1.7382396837292365e-06,
      "loss": 0.7889235,
      "num_input_tokens_seen": 99714045,
      "step": 4620,
      "time_per_iteration": 2.6197712421417236
    },
    {
      "auxiliary_loss_clip": 0.01191337,
      "auxiliary_loss_mlp": 0.01034737,
      "balance_loss_clip": 1.05857158,
      "balance_loss_mlp": 1.02581394,
      "epoch": 0.5556423976432393,
      "flos": 21762513204480.0,
      "grad_norm": 1.8556219641845384,
      "language_loss": 0.73738301,
      "learning_rate": 1.737467432516841e-06,
      "loss": 0.75964373,
      "num_input_tokens_seen": 99734145,
      "step": 4621,
      "time_per_iteration": 2.662562370300293
    },
    {
      "auxiliary_loss_clip": 0.01164875,
      "auxiliary_loss_mlp": 0.01028084,
      "balance_loss_clip": 1.05384314,
      "balance_loss_mlp": 1.01979303,
      "epoch": 0.5557626405338785,
      "flos": 24900171989760.0,
      "grad_norm": 3.7520253914258355,
      "language_loss": 0.74709785,
      "learning_rate": 1.7366952211304274e-06,
      "loss": 0.76902747,
      "num_input_tokens_seen": 99751990,
      "step": 4622,
      "time_per_iteration": 2.643336534500122
    },
    {
      "auxiliary_loss_clip": 0.01169016,
      "auxiliary_loss_mlp": 0.01029753,
      "balance_loss_clip": 1.05470264,
      "balance_loss_mlp": 1.02181983,
      "epoch": 0.5558828834245175,
      "flos": 18697501676160.0,
      "grad_norm": 2.243072029265642,
      "language_loss": 0.83677542,
      "learning_rate": 1.735923049687139e-06,
      "loss": 0.8587631,
      "num_input_tokens_seen": 99768565,
      "step": 4623,
      "time_per_iteration": 2.678347110748291
    },
    {
      "auxiliary_loss_clip": 0.01188849,
      "auxiliary_loss_mlp": 0.01028604,
      "balance_loss_clip": 1.05900931,
      "balance_loss_mlp": 1.0202235,
      "epoch": 0.5560031263151566,
      "flos": 27272179445760.0,
      "grad_norm": 2.363555328356222,
      "language_loss": 0.74106902,
      "learning_rate": 1.7351509183041144e-06,
      "loss": 0.76324356,
      "num_input_tokens_seen": 99788895,
      "step": 4624,
      "time_per_iteration": 2.6494638919830322
    },
    {
      "auxiliary_loss_clip": 0.01167699,
      "auxiliary_loss_mlp": 0.01029114,
      "balance_loss_clip": 1.05546069,
      "balance_loss_mlp": 1.02078712,
      "epoch": 0.5561233692057957,
      "flos": 23403738458880.0,
      "grad_norm": 2.3532285475440466,
      "language_loss": 0.71820533,
      "learning_rate": 1.7343788270984852e-06,
      "loss": 0.74017352,
      "num_input_tokens_seen": 99808035,
      "step": 4625,
      "time_per_iteration": 2.7229866981506348
    },
    {
      "auxiliary_loss_clip": 0.01178908,
      "auxiliary_loss_mlp": 0.01033466,
      "balance_loss_clip": 1.05762374,
      "balance_loss_mlp": 1.02476335,
      "epoch": 0.5562436120964348,
      "flos": 37670867804160.0,
      "grad_norm": 2.003538227487944,
      "language_loss": 0.74786991,
      "learning_rate": 1.7336067761873764e-06,
      "loss": 0.76999366,
      "num_input_tokens_seen": 99830460,
      "step": 4626,
      "time_per_iteration": 3.6903200149536133
    },
    {
      "auxiliary_loss_clip": 0.01164485,
      "auxiliary_loss_mlp": 0.0103305,
      "balance_loss_clip": 1.05535638,
      "balance_loss_mlp": 1.02376997,
      "epoch": 0.5563638549870739,
      "flos": 25155245445120.0,
      "grad_norm": 2.0898875944164947,
      "language_loss": 0.76803219,
      "learning_rate": 1.7328347656879076e-06,
      "loss": 0.79000753,
      "num_input_tokens_seen": 99850320,
      "step": 4627,
      "time_per_iteration": 2.677198648452759
    },
    {
      "auxiliary_loss_clip": 0.01190081,
      "auxiliary_loss_mlp": 0.0103114,
      "balance_loss_clip": 1.05881071,
      "balance_loss_mlp": 1.02218723,
      "epoch": 0.556484097877713,
      "flos": 13581810783360.0,
      "grad_norm": 2.4217414546780196,
      "language_loss": 0.68346953,
      "learning_rate": 1.7320627957171927e-06,
      "loss": 0.70568174,
      "num_input_tokens_seen": 99864980,
      "step": 4628,
      "time_per_iteration": 2.5997626781463623
    },
    {
      "auxiliary_loss_clip": 0.01171272,
      "auxiliary_loss_mlp": 0.01035241,
      "balance_loss_clip": 1.05733967,
      "balance_loss_mlp": 1.0267235,
      "epoch": 0.5566043407683521,
      "flos": 24681368292480.0,
      "grad_norm": 1.7452829558308731,
      "language_loss": 0.81658071,
      "learning_rate": 1.7312908663923382e-06,
      "loss": 0.83864582,
      "num_input_tokens_seen": 99881155,
      "step": 4629,
      "time_per_iteration": 2.6355817317962646
    },
    {
      "auxiliary_loss_clip": 0.01166486,
      "auxiliary_loss_mlp": 0.00763449,
      "balance_loss_clip": 1.05401611,
      "balance_loss_mlp": 1.00133348,
      "epoch": 0.5567245836589911,
      "flos": 20588161950720.0,
      "grad_norm": 1.996821191325986,
      "language_loss": 0.67622614,
      "learning_rate": 1.7305189778304463e-06,
      "loss": 0.69552553,
      "num_input_tokens_seen": 99899330,
      "step": 4630,
      "time_per_iteration": 2.708195447921753
    },
    {
      "auxiliary_loss_clip": 0.01189813,
      "auxiliary_loss_mlp": 0.01025439,
      "balance_loss_clip": 1.0599544,
      "balance_loss_mlp": 1.01693368,
      "epoch": 0.5568448265496303,
      "flos": 20704189858560.0,
      "grad_norm": 1.8243774272810456,
      "language_loss": 0.80227268,
      "learning_rate": 1.729747130148611e-06,
      "loss": 0.82442522,
      "num_input_tokens_seen": 99918525,
      "step": 4631,
      "time_per_iteration": 2.5762815475463867
    },
    {
      "auxiliary_loss_clip": 0.01180211,
      "auxiliary_loss_mlp": 0.01033207,
      "balance_loss_clip": 1.05692768,
      "balance_loss_mlp": 1.02350986,
      "epoch": 0.5569650694402694,
      "flos": 25302910256640.0,
      "grad_norm": 2.3748136402272406,
      "language_loss": 0.76935816,
      "learning_rate": 1.7289753234639208e-06,
      "loss": 0.79149234,
      "num_input_tokens_seen": 99937500,
      "step": 4632,
      "time_per_iteration": 2.66495418548584
    },
    {
      "auxiliary_loss_clip": 0.01190543,
      "auxiliary_loss_mlp": 0.0102887,
      "balance_loss_clip": 1.05913341,
      "balance_loss_mlp": 1.02025127,
      "epoch": 0.5570853123309084,
      "flos": 19712623939200.0,
      "grad_norm": 2.2701877232419836,
      "language_loss": 0.76695597,
      "learning_rate": 1.7282035578934592e-06,
      "loss": 0.78915006,
      "num_input_tokens_seen": 99955665,
      "step": 4633,
      "time_per_iteration": 2.619276523590088
    },
    {
      "auxiliary_loss_clip": 0.01140687,
      "auxiliary_loss_mlp": 0.01032489,
      "balance_loss_clip": 1.0510745,
      "balance_loss_mlp": 1.0247221,
      "epoch": 0.5572055552215476,
      "flos": 16108091153280.0,
      "grad_norm": 1.8021429946775824,
      "language_loss": 0.78798068,
      "learning_rate": 1.727431833554301e-06,
      "loss": 0.80971241,
      "num_input_tokens_seen": 99974140,
      "step": 4634,
      "time_per_iteration": 2.7644505500793457
    },
    {
      "auxiliary_loss_clip": 0.01189193,
      "auxiliary_loss_mlp": 0.0076312,
      "balance_loss_clip": 1.05812585,
      "balance_loss_mlp": 1.00135326,
      "epoch": 0.5573257981121866,
      "flos": 17128815937920.0,
      "grad_norm": 1.7318076878823,
      "language_loss": 0.77402234,
      "learning_rate": 1.7266601505635175e-06,
      "loss": 0.79354548,
      "num_input_tokens_seen": 99991480,
      "step": 4635,
      "time_per_iteration": 2.57612943649292
    },
    {
      "auxiliary_loss_clip": 0.01188841,
      "auxiliary_loss_mlp": 0.01025617,
      "balance_loss_clip": 1.05854559,
      "balance_loss_mlp": 1.01705158,
      "epoch": 0.5574460410028257,
      "flos": 18807029222400.0,
      "grad_norm": 3.206366203360686,
      "language_loss": 0.76418597,
      "learning_rate": 1.7258885090381717e-06,
      "loss": 0.78633058,
      "num_input_tokens_seen": 100009520,
      "step": 4636,
      "time_per_iteration": 2.6094117164611816
    },
    {
      "auxiliary_loss_clip": 0.01182686,
      "auxiliary_loss_mlp": 0.01027435,
      "balance_loss_clip": 1.05500472,
      "balance_loss_mlp": 1.01956081,
      "epoch": 0.5575662838934649,
      "flos": 29642678530560.0,
      "grad_norm": 2.3416667948620837,
      "language_loss": 0.78306568,
      "learning_rate": 1.7251169090953213e-06,
      "loss": 0.80516684,
      "num_input_tokens_seen": 100029995,
      "step": 4637,
      "time_per_iteration": 2.6977972984313965
    },
    {
      "auxiliary_loss_clip": 0.01187021,
      "auxiliary_loss_mlp": 0.01036183,
      "balance_loss_clip": 1.05572891,
      "balance_loss_mlp": 1.02728391,
      "epoch": 0.5576865267841039,
      "flos": 22054466949120.0,
      "grad_norm": 4.152989829754079,
      "language_loss": 0.76297069,
      "learning_rate": 1.7243453508520168e-06,
      "loss": 0.78520274,
      "num_input_tokens_seen": 100046980,
      "step": 4638,
      "time_per_iteration": 2.6076552867889404
    },
    {
      "auxiliary_loss_clip": 0.01189806,
      "auxiliary_loss_mlp": 0.01033443,
      "balance_loss_clip": 1.05759239,
      "balance_loss_mlp": 1.02511072,
      "epoch": 0.557806769674743,
      "flos": 17196040241280.0,
      "grad_norm": 3.5954764991962405,
      "language_loss": 0.84629977,
      "learning_rate": 1.7235738344253038e-06,
      "loss": 0.8685323,
      "num_input_tokens_seen": 100060610,
      "step": 4639,
      "time_per_iteration": 2.6011390686035156
    },
    {
      "auxiliary_loss_clip": 0.01189059,
      "auxiliary_loss_mlp": 0.01039738,
      "balance_loss_clip": 1.0576489,
      "balance_loss_mlp": 1.03043985,
      "epoch": 0.557927012565382,
      "flos": 24712717887360.0,
      "grad_norm": 2.0875965964959597,
      "language_loss": 0.82496452,
      "learning_rate": 1.72280235993222e-06,
      "loss": 0.84725243,
      "num_input_tokens_seen": 100078915,
      "step": 4640,
      "time_per_iteration": 2.6131863594055176
    },
    {
      "auxiliary_loss_clip": 0.01187753,
      "auxiliary_loss_mlp": 0.01033023,
      "balance_loss_clip": 1.05725312,
      "balance_loss_mlp": 1.02393973,
      "epoch": 0.5580472554560212,
      "flos": 16983090460800.0,
      "grad_norm": 3.5293623241708247,
      "language_loss": 0.69433951,
      "learning_rate": 1.722030927489798e-06,
      "loss": 0.71654725,
      "num_input_tokens_seen": 100096195,
      "step": 4641,
      "time_per_iteration": 3.496727466583252
    },
    {
      "auxiliary_loss_clip": 0.01165708,
      "auxiliary_loss_mlp": 0.01030934,
      "balance_loss_clip": 1.05605245,
      "balance_loss_mlp": 1.02207041,
      "epoch": 0.5581674983466602,
      "flos": 23509100027520.0,
      "grad_norm": 2.06094676021891,
      "language_loss": 0.74101704,
      "learning_rate": 1.7212595372150634e-06,
      "loss": 0.7629835,
      "num_input_tokens_seen": 100116175,
      "step": 4642,
      "time_per_iteration": 3.6121129989624023
    },
    {
      "auxiliary_loss_clip": 0.01173937,
      "auxiliary_loss_mlp": 0.01031954,
      "balance_loss_clip": 1.05742073,
      "balance_loss_mlp": 1.02381778,
      "epoch": 0.5582877412372993,
      "flos": 13480291969920.0,
      "grad_norm": 2.2841671650282143,
      "language_loss": 0.72902298,
      "learning_rate": 1.720488189225035e-06,
      "loss": 0.75108188,
      "num_input_tokens_seen": 100133875,
      "step": 4643,
      "time_per_iteration": 2.6097934246063232
    },
    {
      "auxiliary_loss_clip": 0.01188454,
      "auxiliary_loss_mlp": 0.01034109,
      "balance_loss_clip": 1.05697489,
      "balance_loss_mlp": 1.02473307,
      "epoch": 0.5584079841279385,
      "flos": 21903605827200.0,
      "grad_norm": 2.9925920701895166,
      "language_loss": 0.79612374,
      "learning_rate": 1.7197168836367265e-06,
      "loss": 0.81834936,
      "num_input_tokens_seen": 100150685,
      "step": 4644,
      "time_per_iteration": 3.6946702003479004
    },
    {
      "auxiliary_loss_clip": 0.01185653,
      "auxiliary_loss_mlp": 0.0103208,
      "balance_loss_clip": 1.05655849,
      "balance_loss_mlp": 1.02375293,
      "epoch": 0.5585282270185775,
      "flos": 18843550375680.0,
      "grad_norm": 6.2507852560636445,
      "language_loss": 0.82228649,
      "learning_rate": 1.7189456205671433e-06,
      "loss": 0.84446388,
      "num_input_tokens_seen": 100169530,
      "step": 4645,
      "time_per_iteration": 2.661844491958618
    },
    {
      "auxiliary_loss_clip": 0.0118297,
      "auxiliary_loss_mlp": 0.01033437,
      "balance_loss_clip": 1.06101096,
      "balance_loss_mlp": 1.02443695,
      "epoch": 0.5586484699092166,
      "flos": 21868449390720.0,
      "grad_norm": 1.9515020086324815,
      "language_loss": 0.82801962,
      "learning_rate": 1.7181744001332866e-06,
      "loss": 0.85018373,
      "num_input_tokens_seen": 100188140,
      "step": 4646,
      "time_per_iteration": 2.695798873901367
    },
    {
      "auxiliary_loss_clip": 0.01172777,
      "auxiliary_loss_mlp": 0.01029414,
      "balance_loss_clip": 1.06083798,
      "balance_loss_mlp": 1.02084851,
      "epoch": 0.5587687127998557,
      "flos": 22893232412160.0,
      "grad_norm": 3.2901224812608496,
      "language_loss": 0.63384116,
      "learning_rate": 1.7174032224521493e-06,
      "loss": 0.65586305,
      "num_input_tokens_seen": 100206850,
      "step": 4647,
      "time_per_iteration": 2.737607955932617
    },
    {
      "auxiliary_loss_clip": 0.01174961,
      "auxiliary_loss_mlp": 0.01026796,
      "balance_loss_clip": 1.05599999,
      "balance_loss_mlp": 1.01883268,
      "epoch": 0.5588889556904948,
      "flos": 20303067703680.0,
      "grad_norm": 2.310912556487726,
      "language_loss": 0.70474017,
      "learning_rate": 1.7166320876407184e-06,
      "loss": 0.72675765,
      "num_input_tokens_seen": 100226270,
      "step": 4648,
      "time_per_iteration": 2.6668472290039062
    },
    {
      "auxiliary_loss_clip": 0.01177186,
      "auxiliary_loss_mlp": 0.01028631,
      "balance_loss_clip": 1.06059718,
      "balance_loss_mlp": 1.02014899,
      "epoch": 0.5590091985811338,
      "flos": 16472153450880.0,
      "grad_norm": 3.79195341674743,
      "language_loss": 0.67994523,
      "learning_rate": 1.7158609958159742e-06,
      "loss": 0.70200342,
      "num_input_tokens_seen": 100243675,
      "step": 4649,
      "time_per_iteration": 2.6883530616760254
    },
    {
      "auxiliary_loss_clip": 0.01130657,
      "auxiliary_loss_mlp": 0.01032467,
      "balance_loss_clip": 1.04961276,
      "balance_loss_mlp": 1.02337146,
      "epoch": 0.559129441471773,
      "flos": 14532186781440.0,
      "grad_norm": 1.9372946299267269,
      "language_loss": 0.78470892,
      "learning_rate": 1.7150899470948911e-06,
      "loss": 0.80634016,
      "num_input_tokens_seen": 100258940,
      "step": 4650,
      "time_per_iteration": 2.6703009605407715
    },
    {
      "auxiliary_loss_clip": 0.01076119,
      "auxiliary_loss_mlp": 0.01001937,
      "balance_loss_clip": 1.01474249,
      "balance_loss_mlp": 1.00094724,
      "epoch": 0.5592496843624121,
      "flos": 60521009852160.0,
      "grad_norm": 0.8002402983593686,
      "language_loss": 0.56664163,
      "learning_rate": 1.7143189415944365e-06,
      "loss": 0.58742219,
      "num_input_tokens_seen": 100323400,
      "step": 4651,
      "time_per_iteration": 3.219689130783081
    },
    {
      "auxiliary_loss_clip": 0.01188138,
      "auxiliary_loss_mlp": 0.01029518,
      "balance_loss_clip": 1.05808103,
      "balance_loss_mlp": 1.02095306,
      "epoch": 0.5593699272530511,
      "flos": 20886256920960.0,
      "grad_norm": 1.7920074031877677,
      "language_loss": 0.76516962,
      "learning_rate": 1.7135479794315714e-06,
      "loss": 0.78734618,
      "num_input_tokens_seen": 100340355,
      "step": 4652,
      "time_per_iteration": 3.5475447177886963
    },
    {
      "auxiliary_loss_clip": 0.011884,
      "auxiliary_loss_mlp": 0.01031023,
      "balance_loss_clip": 1.05849946,
      "balance_loss_mlp": 1.023036,
      "epoch": 0.5594901701436903,
      "flos": 12896743616640.0,
      "grad_norm": 2.766592801673269,
      "language_loss": 0.7892493,
      "learning_rate": 1.7127770607232502e-06,
      "loss": 0.81144351,
      "num_input_tokens_seen": 100358900,
      "step": 4653,
      "time_per_iteration": 2.6663856506347656
    },
    {
      "auxiliary_loss_clip": 0.01170196,
      "auxiliary_loss_mlp": 0.01036897,
      "balance_loss_clip": 1.05453634,
      "balance_loss_mlp": 1.02792656,
      "epoch": 0.5596104130343293,
      "flos": 23112107936640.0,
      "grad_norm": 2.495039885637237,
      "language_loss": 0.80169332,
      "learning_rate": 1.7120061855864204e-06,
      "loss": 0.8237642,
      "num_input_tokens_seen": 100378910,
      "step": 4654,
      "time_per_iteration": 2.6861982345581055
    },
    {
      "auxiliary_loss_clip": 0.01172927,
      "auxiliary_loss_mlp": 0.01030583,
      "balance_loss_clip": 1.05820537,
      "balance_loss_mlp": 1.02174401,
      "epoch": 0.5597306559249684,
      "flos": 25957812977280.0,
      "grad_norm": 2.7706685555317723,
      "language_loss": 0.71200526,
      "learning_rate": 1.7112353541380233e-06,
      "loss": 0.73404032,
      "num_input_tokens_seen": 100398770,
      "step": 4655,
      "time_per_iteration": 2.6791889667510986
    },
    {
      "auxiliary_loss_clip": 0.01194352,
      "auxiliary_loss_mlp": 0.01035947,
      "balance_loss_clip": 1.06238651,
      "balance_loss_mlp": 1.02688694,
      "epoch": 0.5598508988156076,
      "flos": 22492289825280.0,
      "grad_norm": 2.730767629229948,
      "language_loss": 0.7221759,
      "learning_rate": 1.7104645664949931e-06,
      "loss": 0.74447888,
      "num_input_tokens_seen": 100421240,
      "step": 4656,
      "time_per_iteration": 2.7641937732696533
    },
    {
      "auxiliary_loss_clip": 0.01189114,
      "auxiliary_loss_mlp": 0.01032114,
      "balance_loss_clip": 1.05745506,
      "balance_loss_mlp": 1.02307165,
      "epoch": 0.5599711417062466,
      "flos": 23112538899840.0,
      "grad_norm": 1.9888298104390807,
      "language_loss": 0.71280485,
      "learning_rate": 1.7096938227742584e-06,
      "loss": 0.73501712,
      "num_input_tokens_seen": 100442370,
      "step": 4657,
      "time_per_iteration": 2.615947961807251
    },
    {
      "auxiliary_loss_clip": 0.01191699,
      "auxiliary_loss_mlp": 0.01035129,
      "balance_loss_clip": 1.06006813,
      "balance_loss_mlp": 1.02634907,
      "epoch": 0.5600913845968857,
      "flos": 22339345714560.0,
      "grad_norm": 1.7950791099648338,
      "language_loss": 0.84987164,
      "learning_rate": 1.70892312309274e-06,
      "loss": 0.87213999,
      "num_input_tokens_seen": 100460260,
      "step": 4658,
      "time_per_iteration": 2.6290500164031982
    },
    {
      "auxiliary_loss_clip": 0.01187817,
      "auxiliary_loss_mlp": 0.01029846,
      "balance_loss_clip": 1.05529141,
      "balance_loss_mlp": 1.02134013,
      "epoch": 0.5602116274875248,
      "flos": 17633791290240.0,
      "grad_norm": 2.192831593491079,
      "language_loss": 0.68742764,
      "learning_rate": 1.7081524675673523e-06,
      "loss": 0.70960426,
      "num_input_tokens_seen": 100475750,
      "step": 4659,
      "time_per_iteration": 2.5781571865081787
    },
    {
      "auxiliary_loss_clip": 0.01074135,
      "auxiliary_loss_mlp": 0.01000356,
      "balance_loss_clip": 1.01293635,
      "balance_loss_mlp": 0.99936068,
      "epoch": 0.5603318703781639,
      "flos": 70115945529600.0,
      "grad_norm": 0.7894430819551608,
      "language_loss": 0.59575891,
      "learning_rate": 1.7073818563150026e-06,
      "loss": 0.61650383,
      "num_input_tokens_seen": 100537830,
      "step": 4660,
      "time_per_iteration": 3.3187434673309326
    },
    {
      "auxiliary_loss_clip": 0.01152318,
      "auxiliary_loss_mlp": 0.01034401,
      "balance_loss_clip": 1.0548408,
      "balance_loss_mlp": 1.0256567,
      "epoch": 0.560452113268803,
      "flos": 18545850455040.0,
      "grad_norm": 2.044351235869328,
      "language_loss": 0.86756074,
      "learning_rate": 1.7066112894525935e-06,
      "loss": 0.88942796,
      "num_input_tokens_seen": 100555910,
      "step": 4661,
      "time_per_iteration": 2.7093915939331055
    },
    {
      "auxiliary_loss_clip": 0.01173349,
      "auxiliary_loss_mlp": 0.01033137,
      "balance_loss_clip": 1.05556417,
      "balance_loss_mlp": 1.02378535,
      "epoch": 0.5605723561594421,
      "flos": 25264665250560.0,
      "grad_norm": 1.8317353317378327,
      "language_loss": 0.72693998,
      "learning_rate": 1.7058407670970177e-06,
      "loss": 0.74900484,
      "num_input_tokens_seen": 100577385,
      "step": 4662,
      "time_per_iteration": 2.69270658493042
    },
    {
      "auxiliary_loss_clip": 0.01188496,
      "auxiliary_loss_mlp": 0.01030696,
      "balance_loss_clip": 1.05604899,
      "balance_loss_mlp": 1.02167809,
      "epoch": 0.5606925990500812,
      "flos": 20594949621120.0,
      "grad_norm": 2.1518857281895443,
      "language_loss": 0.61335111,
      "learning_rate": 1.7050702893651643e-06,
      "loss": 0.63554305,
      "num_input_tokens_seen": 100596965,
      "step": 4663,
      "time_per_iteration": 2.548037052154541
    },
    {
      "auxiliary_loss_clip": 0.01177899,
      "auxiliary_loss_mlp": 0.01032139,
      "balance_loss_clip": 1.0555768,
      "balance_loss_mlp": 1.02335882,
      "epoch": 0.5608128419407202,
      "flos": 35006044677120.0,
      "grad_norm": 2.3604701049369554,
      "language_loss": 0.75643122,
      "learning_rate": 1.7042998563739134e-06,
      "loss": 0.77853155,
      "num_input_tokens_seen": 100615315,
      "step": 4664,
      "time_per_iteration": 2.7857651710510254
    },
    {
      "auxiliary_loss_clip": 0.0119276,
      "auxiliary_loss_mlp": 0.01030017,
      "balance_loss_clip": 1.05936515,
      "balance_loss_mlp": 1.01984215,
      "epoch": 0.5609330848313594,
      "flos": 24639819235200.0,
      "grad_norm": 15.066891561045397,
      "language_loss": 0.71649206,
      "learning_rate": 1.703529468240139e-06,
      "loss": 0.73871988,
      "num_input_tokens_seen": 100634185,
      "step": 4665,
      "time_per_iteration": 2.624081611633301
    },
    {
      "auxiliary_loss_clip": 0.01156752,
      "auxiliary_loss_mlp": 0.01029901,
      "balance_loss_clip": 1.051651,
      "balance_loss_mlp": 1.02090645,
      "epoch": 0.5610533277219985,
      "flos": 18762894385920.0,
      "grad_norm": 2.9007550460387197,
      "language_loss": 0.73569179,
      "learning_rate": 1.7027591250807088e-06,
      "loss": 0.75755835,
      "num_input_tokens_seen": 100651360,
      "step": 4666,
      "time_per_iteration": 2.6556129455566406
    },
    {
      "auxiliary_loss_clip": 0.01176259,
      "auxiliary_loss_mlp": 0.01034164,
      "balance_loss_clip": 1.060287,
      "balance_loss_mlp": 1.02550364,
      "epoch": 0.5611735706126375,
      "flos": 15012384727680.0,
      "grad_norm": 2.875986225871481,
      "language_loss": 0.8468641,
      "learning_rate": 1.7019888270124825e-06,
      "loss": 0.86896837,
      "num_input_tokens_seen": 100668525,
      "step": 4667,
      "time_per_iteration": 2.650941848754883
    },
    {
      "auxiliary_loss_clip": 0.01168035,
      "auxiliary_loss_mlp": 0.01041958,
      "balance_loss_clip": 1.05704522,
      "balance_loss_mlp": 1.03351784,
      "epoch": 0.5612938135032767,
      "flos": 16468167041280.0,
      "grad_norm": 2.142139581360174,
      "language_loss": 0.82417625,
      "learning_rate": 1.7012185741523147e-06,
      "loss": 0.84627616,
      "num_input_tokens_seen": 100684850,
      "step": 4668,
      "time_per_iteration": 3.5734543800354004
    },
    {
      "auxiliary_loss_clip": 0.01189277,
      "auxiliary_loss_mlp": 0.01032162,
      "balance_loss_clip": 1.05905581,
      "balance_loss_mlp": 1.02320898,
      "epoch": 0.5614140563939157,
      "flos": 25666433850240.0,
      "grad_norm": 2.1655965006237743,
      "language_loss": 0.62740648,
      "learning_rate": 1.7004483666170514e-06,
      "loss": 0.64962089,
      "num_input_tokens_seen": 100705345,
      "step": 4669,
      "time_per_iteration": 2.6764891147613525
    },
    {
      "auxiliary_loss_clip": 0.01187929,
      "auxiliary_loss_mlp": 0.01031402,
      "balance_loss_clip": 1.0575645,
      "balance_loss_mlp": 1.02274704,
      "epoch": 0.5615342992845548,
      "flos": 24717566223360.0,
      "grad_norm": 2.0678983448736594,
      "language_loss": 0.8034929,
      "learning_rate": 1.699678204523533e-06,
      "loss": 0.82568622,
      "num_input_tokens_seen": 100725210,
      "step": 4670,
      "time_per_iteration": 3.61586856842041
    },
    {
      "auxiliary_loss_clip": 0.01195376,
      "auxiliary_loss_mlp": 0.01032998,
      "balance_loss_clip": 1.06160319,
      "balance_loss_mlp": 1.02410531,
      "epoch": 0.5616545421751938,
      "flos": 22015934634240.0,
      "grad_norm": 2.4679034951979055,
      "language_loss": 0.68347931,
      "learning_rate": 1.6989080879885918e-06,
      "loss": 0.7057631,
      "num_input_tokens_seen": 100743070,
      "step": 4671,
      "time_per_iteration": 2.6037144660949707
    },
    {
      "auxiliary_loss_clip": 0.01057569,
      "auxiliary_loss_mlp": 0.01007265,
      "balance_loss_clip": 1.02450967,
      "balance_loss_mlp": 1.00622785,
      "epoch": 0.561774785065833,
      "flos": 53760358690560.0,
      "grad_norm": 0.8997512450559172,
      "language_loss": 0.60942292,
      "learning_rate": 1.6981380171290544e-06,
      "loss": 0.63007128,
      "num_input_tokens_seen": 100804095,
      "step": 4672,
      "time_per_iteration": 3.202723741531372
    },
    {
      "auxiliary_loss_clip": 0.01176008,
      "auxiliary_loss_mlp": 0.01030778,
      "balance_loss_clip": 1.05566013,
      "balance_loss_mlp": 1.02167058,
      "epoch": 0.5618950279564721,
      "flos": 19750007018880.0,
      "grad_norm": 2.0566607849785843,
      "language_loss": 0.74169827,
      "learning_rate": 1.6973679920617396e-06,
      "loss": 0.76376611,
      "num_input_tokens_seen": 100821630,
      "step": 4673,
      "time_per_iteration": 2.632530450820923
    },
    {
      "auxiliary_loss_clip": 0.01191855,
      "auxiliary_loss_mlp": 0.01027813,
      "balance_loss_clip": 1.06154394,
      "balance_loss_mlp": 1.01891375,
      "epoch": 0.5620152708471111,
      "flos": 16800592435200.0,
      "grad_norm": 2.4686139289096114,
      "language_loss": 0.85149121,
      "learning_rate": 1.6965980129034603e-06,
      "loss": 0.87368786,
      "num_input_tokens_seen": 100839015,
      "step": 4674,
      "time_per_iteration": 2.645289182662964
    },
    {
      "auxiliary_loss_clip": 0.01178775,
      "auxiliary_loss_mlp": 0.01034148,
      "balance_loss_clip": 1.05746603,
      "balance_loss_mlp": 1.02498627,
      "epoch": 0.5621355137377503,
      "flos": 26797799502720.0,
      "grad_norm": 1.854086852499463,
      "language_loss": 0.77027881,
      "learning_rate": 1.6958280797710209e-06,
      "loss": 0.79240799,
      "num_input_tokens_seen": 100860940,
      "step": 4675,
      "time_per_iteration": 2.6622798442840576
    },
    {
      "auxiliary_loss_clip": 0.01075284,
      "auxiliary_loss_mlp": 0.01003702,
      "balance_loss_clip": 1.01414037,
      "balance_loss_mlp": 1.00268269,
      "epoch": 0.5622557566283893,
      "flos": 61207046686080.0,
      "grad_norm": 0.7131866799144972,
      "language_loss": 0.54686481,
      "learning_rate": 1.6950581927812198e-06,
      "loss": 0.56765467,
      "num_input_tokens_seen": 100920510,
      "step": 4676,
      "time_per_iteration": 3.092355728149414
    },
    {
      "auxiliary_loss_clip": 0.01171407,
      "auxiliary_loss_mlp": 0.01028049,
      "balance_loss_clip": 1.05503023,
      "balance_loss_mlp": 1.019889,
      "epoch": 0.5623759995190284,
      "flos": 26468534505600.0,
      "grad_norm": 2.074437197286809,
      "language_loss": 0.78905469,
      "learning_rate": 1.6942883520508486e-06,
      "loss": 0.81104928,
      "num_input_tokens_seen": 100939245,
      "step": 4677,
      "time_per_iteration": 2.6649725437164307
    },
    {
      "auxiliary_loss_clip": 0.01172977,
      "auxiliary_loss_mlp": 0.01032995,
      "balance_loss_clip": 1.05672884,
      "balance_loss_mlp": 1.02481711,
      "epoch": 0.5624962424096676,
      "flos": 19390900798080.0,
      "grad_norm": 2.4485865613104036,
      "language_loss": 0.77495754,
      "learning_rate": 1.693518557696691e-06,
      "loss": 0.79701722,
      "num_input_tokens_seen": 100958385,
      "step": 4678,
      "time_per_iteration": 3.5930943489074707
    },
    {
      "auxiliary_loss_clip": 0.01185155,
      "auxiliary_loss_mlp": 0.01031957,
      "balance_loss_clip": 1.0558176,
      "balance_loss_mlp": 1.02343321,
      "epoch": 0.5626164853003066,
      "flos": 20667345482880.0,
      "grad_norm": 2.3677696019479195,
      "language_loss": 0.88849628,
      "learning_rate": 1.6927488098355252e-06,
      "loss": 0.91066742,
      "num_input_tokens_seen": 100976015,
      "step": 4679,
      "time_per_iteration": 2.6324687004089355
    },
    {
      "auxiliary_loss_clip": 0.01074414,
      "auxiliary_loss_mlp": 0.01002364,
      "balance_loss_clip": 1.01351357,
      "balance_loss_mlp": 1.00133312,
      "epoch": 0.5627367281909457,
      "flos": 62766071665920.0,
      "grad_norm": 0.9085187283871573,
      "language_loss": 0.63104129,
      "learning_rate": 1.6919791085841201e-06,
      "loss": 0.6518091,
      "num_input_tokens_seen": 101033425,
      "step": 4680,
      "time_per_iteration": 3.121159315109253
    },
    {
      "auxiliary_loss_clip": 0.01135684,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.05392528,
      "balance_loss_mlp": 1.016675,
      "epoch": 0.5628569710815848,
      "flos": 12787144243200.0,
      "grad_norm": 3.58815323939524,
      "language_loss": 0.78883827,
      "learning_rate": 1.6912094540592396e-06,
      "loss": 0.81044734,
      "num_input_tokens_seen": 101048945,
      "step": 4681,
      "time_per_iteration": 2.731764316558838
    },
    {
      "auxiliary_loss_clip": 0.01187125,
      "auxiliary_loss_mlp": 0.01034976,
      "balance_loss_clip": 1.05677891,
      "balance_loss_mlp": 1.02631545,
      "epoch": 0.5629772139722239,
      "flos": 13762082165760.0,
      "grad_norm": 2.5060832115294565,
      "language_loss": 0.81434488,
      "learning_rate": 1.6904398463776393e-06,
      "loss": 0.83656585,
      "num_input_tokens_seen": 101062745,
      "step": 4682,
      "time_per_iteration": 2.521181344985962
    },
    {
      "auxiliary_loss_clip": 0.01186522,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 1.05793881,
      "balance_loss_mlp": 1.01866734,
      "epoch": 0.5630974568628629,
      "flos": 21467830026240.0,
      "grad_norm": 18.917247432830774,
      "language_loss": 0.72664279,
      "learning_rate": 1.6896702856560683e-06,
      "loss": 0.7487781,
      "num_input_tokens_seen": 101081840,
      "step": 4683,
      "time_per_iteration": 2.651432752609253
    },
    {
      "auxiliary_loss_clip": 0.0118847,
      "auxiliary_loss_mlp": 0.01028324,
      "balance_loss_clip": 1.05733657,
      "balance_loss_mlp": 1.01957393,
      "epoch": 0.5632176997535021,
      "flos": 14245907385600.0,
      "grad_norm": 2.4690765609510454,
      "language_loss": 0.69241929,
      "learning_rate": 1.6889007720112677e-06,
      "loss": 0.71458733,
      "num_input_tokens_seen": 101099585,
      "step": 4684,
      "time_per_iteration": 2.567504405975342
    },
    {
      "auxiliary_loss_clip": 0.01189842,
      "auxiliary_loss_mlp": 0.01026358,
      "balance_loss_clip": 1.0584693,
      "balance_loss_mlp": 1.01789427,
      "epoch": 0.5633379426441412,
      "flos": 20812244947200.0,
      "grad_norm": 1.6601153738147785,
      "language_loss": 0.77639323,
      "learning_rate": 1.6881313055599734e-06,
      "loss": 0.79855525,
      "num_input_tokens_seen": 101119515,
      "step": 4685,
      "time_per_iteration": 2.6359920501708984
    },
    {
      "auxiliary_loss_clip": 0.01191443,
      "auxiliary_loss_mlp": 0.01030098,
      "balance_loss_clip": 1.05859971,
      "balance_loss_mlp": 1.02125871,
      "epoch": 0.5634581855347802,
      "flos": 22600883617920.0,
      "grad_norm": 2.333603933135899,
      "language_loss": 0.825701,
      "learning_rate": 1.6873618864189117e-06,
      "loss": 0.84791648,
      "num_input_tokens_seen": 101135285,
      "step": 4686,
      "time_per_iteration": 2.5992393493652344
    },
    {
      "auxiliary_loss_clip": 0.01150107,
      "auxiliary_loss_mlp": 0.01036603,
      "balance_loss_clip": 1.05170107,
      "balance_loss_mlp": 1.02743602,
      "epoch": 0.5635784284254194,
      "flos": 21506972872320.0,
      "grad_norm": 4.340502042373835,
      "language_loss": 0.78280872,
      "learning_rate": 1.686592514704803e-06,
      "loss": 0.80467582,
      "num_input_tokens_seen": 101152680,
      "step": 4687,
      "time_per_iteration": 2.6856398582458496
    },
    {
      "auxiliary_loss_clip": 0.01171653,
      "auxiliary_loss_mlp": 0.01032333,
      "balance_loss_clip": 1.05725598,
      "balance_loss_mlp": 1.0240953,
      "epoch": 0.5636986713160584,
      "flos": 19827466698240.0,
      "grad_norm": 4.7367362640258825,
      "language_loss": 0.71247411,
      "learning_rate": 1.685823190534361e-06,
      "loss": 0.73451394,
      "num_input_tokens_seen": 101170920,
      "step": 4688,
      "time_per_iteration": 2.7007288932800293
    },
    {
      "auxiliary_loss_clip": 0.01155014,
      "auxiliary_loss_mlp": 0.01033628,
      "balance_loss_clip": 1.05439973,
      "balance_loss_mlp": 1.02350688,
      "epoch": 0.5638189142066975,
      "flos": 19792453916160.0,
      "grad_norm": 1.7809034573078564,
      "language_loss": 0.84344435,
      "learning_rate": 1.6850539140242907e-06,
      "loss": 0.8653307,
      "num_input_tokens_seen": 101190180,
      "step": 4689,
      "time_per_iteration": 2.7178595066070557
    },
    {
      "auxiliary_loss_clip": 0.01171398,
      "auxiliary_loss_mlp": 0.00762989,
      "balance_loss_clip": 1.05659628,
      "balance_loss_mlp": 1.00125182,
      "epoch": 0.5639391570973367,
      "flos": 22893771116160.0,
      "grad_norm": 3.6509692820862076,
      "language_loss": 0.82562256,
      "learning_rate": 1.684284685291292e-06,
      "loss": 0.84496641,
      "num_input_tokens_seen": 101211825,
      "step": 4690,
      "time_per_iteration": 2.736616373062134
    },
    {
      "auxiliary_loss_clip": 0.01161988,
      "auxiliary_loss_mlp": 0.01036503,
      "balance_loss_clip": 1.05640686,
      "balance_loss_mlp": 1.02756846,
      "epoch": 0.5640593999879757,
      "flos": 23727077712000.0,
      "grad_norm": 2.143639921811838,
      "language_loss": 0.81465572,
      "learning_rate": 1.683515504452055e-06,
      "loss": 0.8366406,
      "num_input_tokens_seen": 101229200,
      "step": 4691,
      "time_per_iteration": 2.7370970249176025
    },
    {
      "auxiliary_loss_clip": 0.01191623,
      "auxiliary_loss_mlp": 0.01034255,
      "balance_loss_clip": 1.05826497,
      "balance_loss_mlp": 1.02488565,
      "epoch": 0.5641796428786148,
      "flos": 22710123855360.0,
      "grad_norm": 1.6717715398278474,
      "language_loss": 0.66450453,
      "learning_rate": 1.6827463716232648e-06,
      "loss": 0.68676329,
      "num_input_tokens_seen": 101249860,
      "step": 4692,
      "time_per_iteration": 2.724325180053711
    },
    {
      "auxiliary_loss_clip": 0.01165211,
      "auxiliary_loss_mlp": 0.01030653,
      "balance_loss_clip": 1.05597508,
      "balance_loss_mlp": 1.02183104,
      "epoch": 0.5642998857692539,
      "flos": 19791987039360.0,
      "grad_norm": 1.7224397701745755,
      "language_loss": 0.76103699,
      "learning_rate": 1.6819772869215972e-06,
      "loss": 0.78299558,
      "num_input_tokens_seen": 101268940,
      "step": 4693,
      "time_per_iteration": 2.635532855987549
    },
    {
      "auxiliary_loss_clip": 0.01191215,
      "auxiliary_loss_mlp": 0.01030507,
      "balance_loss_clip": 1.05939269,
      "balance_loss_mlp": 1.02164364,
      "epoch": 0.564420128659893,
      "flos": 23185904428800.0,
      "grad_norm": 1.8806504800349393,
      "language_loss": 0.82625043,
      "learning_rate": 1.6812082504637228e-06,
      "loss": 0.84846765,
      "num_input_tokens_seen": 101290260,
      "step": 4694,
      "time_per_iteration": 4.564510345458984
    },
    {
      "auxiliary_loss_clip": 0.01162067,
      "auxiliary_loss_mlp": 0.01033328,
      "balance_loss_clip": 1.05554235,
      "balance_loss_mlp": 1.02521598,
      "epoch": 0.564540371550532,
      "flos": 23258264376960.0,
      "grad_norm": 1.5642045666854218,
      "language_loss": 0.7436595,
      "learning_rate": 1.6804392623663025e-06,
      "loss": 0.76561344,
      "num_input_tokens_seen": 101311465,
      "step": 4695,
      "time_per_iteration": 2.7332966327667236
    },
    {
      "auxiliary_loss_clip": 0.01186711,
      "auxiliary_loss_mlp": 0.01029818,
      "balance_loss_clip": 1.05917668,
      "balance_loss_mlp": 1.02167904,
      "epoch": 0.5646606144411712,
      "flos": 25010058672000.0,
      "grad_norm": 5.299342988956318,
      "language_loss": 0.78213459,
      "learning_rate": 1.6796703227459935e-06,
      "loss": 0.80429989,
      "num_input_tokens_seen": 101329420,
      "step": 4696,
      "time_per_iteration": 3.600594997406006
    },
    {
      "auxiliary_loss_clip": 0.01184865,
      "auxiliary_loss_mlp": 0.01031403,
      "balance_loss_clip": 1.05636477,
      "balance_loss_mlp": 1.02317739,
      "epoch": 0.5647808573318103,
      "flos": 36539645806080.0,
      "grad_norm": 1.86962168313731,
      "language_loss": 0.75945079,
      "learning_rate": 1.6789014317194407e-06,
      "loss": 0.78161347,
      "num_input_tokens_seen": 101350900,
      "step": 4697,
      "time_per_iteration": 2.7202813625335693
    },
    {
      "auxiliary_loss_clip": 0.01177645,
      "auxiliary_loss_mlp": 0.01035264,
      "balance_loss_clip": 1.05823541,
      "balance_loss_mlp": 1.02638257,
      "epoch": 0.5649011002224493,
      "flos": 22528451842560.0,
      "grad_norm": 3.4612026030780494,
      "language_loss": 0.72550559,
      "learning_rate": 1.6781325894032853e-06,
      "loss": 0.74763465,
      "num_input_tokens_seen": 101369860,
      "step": 4698,
      "time_per_iteration": 2.656632423400879
    },
    {
      "auxiliary_loss_clip": 0.0116971,
      "auxiliary_loss_mlp": 0.01031072,
      "balance_loss_clip": 1.055372,
      "balance_loss_mlp": 1.02245283,
      "epoch": 0.5650213431130885,
      "flos": 18515147304960.0,
      "grad_norm": 1.8745237312781489,
      "language_loss": 0.92047787,
      "learning_rate": 1.6773637959141608e-06,
      "loss": 0.94248575,
      "num_input_tokens_seen": 101386835,
      "step": 4699,
      "time_per_iteration": 2.6368651390075684
    },
    {
      "auxiliary_loss_clip": 0.01172694,
      "auxiliary_loss_mlp": 0.01027557,
      "balance_loss_clip": 1.05729616,
      "balance_loss_mlp": 1.01921868,
      "epoch": 0.5651415860037275,
      "flos": 17526310819200.0,
      "grad_norm": 2.5404375066107816,
      "language_loss": 0.66568148,
      "learning_rate": 1.6765950513686915e-06,
      "loss": 0.687684,
      "num_input_tokens_seen": 101404945,
      "step": 4700,
      "time_per_iteration": 2.651067018508911
    },
    {
      "auxiliary_loss_clip": 0.01153765,
      "auxiliary_loss_mlp": 0.01034545,
      "balance_loss_clip": 1.05130506,
      "balance_loss_mlp": 1.02587271,
      "epoch": 0.5652618288943666,
      "flos": 25520026014720.0,
      "grad_norm": 1.8182543946781684,
      "language_loss": 0.76410139,
      "learning_rate": 1.675826355883496e-06,
      "loss": 0.78598452,
      "num_input_tokens_seen": 101424160,
      "step": 4701,
      "time_per_iteration": 2.790888786315918
    },
    {
      "auxiliary_loss_clip": 0.01171662,
      "auxiliary_loss_mlp": 0.01032965,
      "balance_loss_clip": 1.05495524,
      "balance_loss_mlp": 1.02456081,
      "epoch": 0.5653820717850057,
      "flos": 19683105937920.0,
      "grad_norm": 2.3326930702307003,
      "language_loss": 0.79332995,
      "learning_rate": 1.6750577095751848e-06,
      "loss": 0.81537628,
      "num_input_tokens_seen": 101443270,
      "step": 4702,
      "time_per_iteration": 2.732083797454834
    },
    {
      "auxiliary_loss_clip": 0.01187396,
      "auxiliary_loss_mlp": 0.01029612,
      "balance_loss_clip": 1.05938661,
      "balance_loss_mlp": 1.02194977,
      "epoch": 0.5655023146756448,
      "flos": 26979722910720.0,
      "grad_norm": 2.2049858168606224,
      "language_loss": 0.72827065,
      "learning_rate": 1.6742891125603605e-06,
      "loss": 0.75044072,
      "num_input_tokens_seen": 101464175,
      "step": 4703,
      "time_per_iteration": 2.675788640975952
    },
    {
      "auxiliary_loss_clip": 0.0115916,
      "auxiliary_loss_mlp": 0.01035668,
      "balance_loss_clip": 1.0553813,
      "balance_loss_mlp": 1.02608347,
      "epoch": 0.5656225575662839,
      "flos": 27669351104640.0,
      "grad_norm": 2.4910884305841297,
      "language_loss": 0.72176272,
      "learning_rate": 1.6735205649556185e-06,
      "loss": 0.74371099,
      "num_input_tokens_seen": 101484045,
      "step": 4704,
      "time_per_iteration": 3.721771240234375
    },
    {
      "auxiliary_loss_clip": 0.01186574,
      "auxiliary_loss_mlp": 0.01038013,
      "balance_loss_clip": 1.0580852,
      "balance_loss_mlp": 1.0292747,
      "epoch": 0.5657428004569229,
      "flos": 24349732997760.0,
      "grad_norm": 1.7176707168981324,
      "language_loss": 0.84924316,
      "learning_rate": 1.6727520668775476e-06,
      "loss": 0.87148905,
      "num_input_tokens_seen": 101504330,
      "step": 4705,
      "time_per_iteration": 2.6734108924865723
    },
    {
      "auxiliary_loss_clip": 0.01171438,
      "auxiliary_loss_mlp": 0.01033315,
      "balance_loss_clip": 1.05360198,
      "balance_loss_mlp": 1.02465987,
      "epoch": 0.5658630433475621,
      "flos": 21944041562880.0,
      "grad_norm": 4.152213713518761,
      "language_loss": 0.75323588,
      "learning_rate": 1.6719836184427275e-06,
      "loss": 0.7752834,
      "num_input_tokens_seen": 101524635,
      "step": 4706,
      "time_per_iteration": 2.701679229736328
    },
    {
      "auxiliary_loss_clip": 0.01170076,
      "auxiliary_loss_mlp": 0.01033322,
      "balance_loss_clip": 1.0544591,
      "balance_loss_mlp": 1.02512646,
      "epoch": 0.5659832862382012,
      "flos": 30409012218240.0,
      "grad_norm": 1.7565484896397057,
      "language_loss": 0.64738613,
      "learning_rate": 1.671215219767733e-06,
      "loss": 0.66942012,
      "num_input_tokens_seen": 101544095,
      "step": 4707,
      "time_per_iteration": 2.7407376766204834
    },
    {
      "auxiliary_loss_clip": 0.01163809,
      "auxiliary_loss_mlp": 0.01027837,
      "balance_loss_clip": 1.05403614,
      "balance_loss_mlp": 1.01955223,
      "epoch": 0.5661035291288402,
      "flos": 13188194570880.0,
      "grad_norm": 2.437206838374055,
      "language_loss": 0.76097155,
      "learning_rate": 1.670446870969127e-06,
      "loss": 0.782888,
      "num_input_tokens_seen": 101561760,
      "step": 4708,
      "time_per_iteration": 2.6429240703582764
    },
    {
      "auxiliary_loss_clip": 0.01186945,
      "auxiliary_loss_mlp": 0.01031359,
      "balance_loss_clip": 1.05744863,
      "balance_loss_mlp": 1.02323532,
      "epoch": 0.5662237720194794,
      "flos": 16143032108160.0,
      "grad_norm": 2.123362337250392,
      "language_loss": 0.80417663,
      "learning_rate": 1.6696785721634685e-06,
      "loss": 0.82635975,
      "num_input_tokens_seen": 101576245,
      "step": 4709,
      "time_per_iteration": 2.6415822505950928
    },
    {
      "auxiliary_loss_clip": 0.01177841,
      "auxiliary_loss_mlp": 0.01036743,
      "balance_loss_clip": 1.06034648,
      "balance_loss_mlp": 1.02776694,
      "epoch": 0.5663440149101184,
      "flos": 17676848718720.0,
      "grad_norm": 1.9694425634116963,
      "language_loss": 0.73594856,
      "learning_rate": 1.6689103234673086e-06,
      "loss": 0.75809431,
      "num_input_tokens_seen": 101594565,
      "step": 4710,
      "time_per_iteration": 2.605853796005249
    },
    {
      "auxiliary_loss_clip": 0.01172796,
      "auxiliary_loss_mlp": 0.01028959,
      "balance_loss_clip": 1.05339336,
      "balance_loss_mlp": 1.02020264,
      "epoch": 0.5664642578007575,
      "flos": 23368330627200.0,
      "grad_norm": 3.0884175742190436,
      "language_loss": 0.77255464,
      "learning_rate": 1.668142124997189e-06,
      "loss": 0.79457217,
      "num_input_tokens_seen": 101614225,
      "step": 4711,
      "time_per_iteration": 2.6984171867370605
    },
    {
      "auxiliary_loss_clip": 0.01064708,
      "auxiliary_loss_mlp": 0.01002263,
      "balance_loss_clip": 1.01398766,
      "balance_loss_mlp": 1.00125015,
      "epoch": 0.5665845006913967,
      "flos": 65516470945920.0,
      "grad_norm": 0.7511689559707035,
      "language_loss": 0.59666443,
      "learning_rate": 1.6673739768696453e-06,
      "loss": 0.61733419,
      "num_input_tokens_seen": 101680795,
      "step": 4712,
      "time_per_iteration": 3.2970950603485107
    },
    {
      "auxiliary_loss_clip": 0.01185989,
      "auxiliary_loss_mlp": 0.0103613,
      "balance_loss_clip": 1.0569706,
      "balance_loss_mlp": 1.02739835,
      "epoch": 0.5667047435820357,
      "flos": 26140885620480.0,
      "grad_norm": 1.7833072796187985,
      "language_loss": 0.77776062,
      "learning_rate": 1.6666058792012052e-06,
      "loss": 0.79998183,
      "num_input_tokens_seen": 101701680,
      "step": 4713,
      "time_per_iteration": 2.6845226287841797
    },
    {
      "auxiliary_loss_clip": 0.01074696,
      "auxiliary_loss_mlp": 0.01002437,
      "balance_loss_clip": 1.01371205,
      "balance_loss_mlp": 1.00146568,
      "epoch": 0.5668249864726748,
      "flos": 71866949725440.0,
      "grad_norm": 0.8970178857324137,
      "language_loss": 0.68767846,
      "learning_rate": 1.6658378321083878e-06,
      "loss": 0.70844978,
      "num_input_tokens_seen": 101766010,
      "step": 4714,
      "time_per_iteration": 3.245415210723877
    },
    {
      "auxiliary_loss_clip": 0.0117147,
      "auxiliary_loss_mlp": 0.01028648,
      "balance_loss_clip": 1.05710614,
      "balance_loss_mlp": 1.02070284,
      "epoch": 0.5669452293633139,
      "flos": 22195667312640.0,
      "grad_norm": 1.79625369356692,
      "language_loss": 0.82729799,
      "learning_rate": 1.6650698357077055e-06,
      "loss": 0.84929913,
      "num_input_tokens_seen": 101783055,
      "step": 4715,
      "time_per_iteration": 2.6164662837982178
    },
    {
      "auxiliary_loss_clip": 0.01168859,
      "auxiliary_loss_mlp": 0.01030866,
      "balance_loss_clip": 1.05470669,
      "balance_loss_mlp": 1.02242613,
      "epoch": 0.567065472253953,
      "flos": 18223193560320.0,
      "grad_norm": 5.085347397188681,
      "language_loss": 0.81297684,
      "learning_rate": 1.6643018901156632e-06,
      "loss": 0.83497405,
      "num_input_tokens_seen": 101802150,
      "step": 4716,
      "time_per_iteration": 2.707951784133911
    },
    {
      "auxiliary_loss_clip": 0.01150058,
      "auxiliary_loss_mlp": 0.01034367,
      "balance_loss_clip": 1.05394423,
      "balance_loss_mlp": 1.02643371,
      "epoch": 0.567185715144592,
      "flos": 20371548983040.0,
      "grad_norm": 2.6294203406627124,
      "language_loss": 0.79829621,
      "learning_rate": 1.6635339954487566e-06,
      "loss": 0.82014048,
      "num_input_tokens_seen": 101818025,
      "step": 4717,
      "time_per_iteration": 2.6622214317321777
    },
    {
      "auxiliary_loss_clip": 0.01162233,
      "auxiliary_loss_mlp": 0.01034063,
      "balance_loss_clip": 1.05712807,
      "balance_loss_mlp": 1.02553344,
      "epoch": 0.5673059580352312,
      "flos": 23221348174080.0,
      "grad_norm": 1.7363914776912872,
      "language_loss": 0.82260096,
      "learning_rate": 1.6627661518234765e-06,
      "loss": 0.84456396,
      "num_input_tokens_seen": 101837280,
      "step": 4718,
      "time_per_iteration": 2.7461154460906982
    },
    {
      "auxiliary_loss_clip": 0.0117446,
      "auxiliary_loss_mlp": 0.01029198,
      "balance_loss_clip": 1.05535924,
      "balance_loss_mlp": 1.02034676,
      "epoch": 0.5674262009258703,
      "flos": 21719599430400.0,
      "grad_norm": 1.6512688366347144,
      "language_loss": 0.85542685,
      "learning_rate": 1.661998359356302e-06,
      "loss": 0.8774634,
      "num_input_tokens_seen": 101856310,
      "step": 4719,
      "time_per_iteration": 2.622361183166504
    },
    {
      "auxiliary_loss_clip": 0.01055309,
      "auxiliary_loss_mlp": 0.01005892,
      "balance_loss_clip": 1.01313674,
      "balance_loss_mlp": 1.00467575,
      "epoch": 0.5675464438165093,
      "flos": 67470369114240.0,
      "grad_norm": 0.7486673926356426,
      "language_loss": 0.5568645,
      "learning_rate": 1.6612306181637077e-06,
      "loss": 0.5774765,
      "num_input_tokens_seen": 101915635,
      "step": 4720,
      "time_per_iteration": 5.028491497039795
    },
    {
      "auxiliary_loss_clip": 0.01173036,
      "auxiliary_loss_mlp": 0.01035153,
      "balance_loss_clip": 1.05622768,
      "balance_loss_mlp": 1.02589083,
      "epoch": 0.5676666867071485,
      "flos": 18879173688960.0,
      "grad_norm": 2.1647890299053256,
      "language_loss": 0.65563452,
      "learning_rate": 1.6604629283621598e-06,
      "loss": 0.67771643,
      "num_input_tokens_seen": 101933565,
      "step": 4721,
      "time_per_iteration": 2.594994306564331
    },
    {
      "auxiliary_loss_clip": 0.01172531,
      "auxiliary_loss_mlp": 0.01028951,
      "balance_loss_clip": 1.05702281,
      "balance_loss_mlp": 1.02092791,
      "epoch": 0.5677869295977875,
      "flos": 33546778744320.0,
      "grad_norm": 2.6900761679180087,
      "language_loss": 0.74446225,
      "learning_rate": 1.6596952900681152e-06,
      "loss": 0.76647699,
      "num_input_tokens_seen": 101954325,
      "step": 4722,
      "time_per_iteration": 3.750709056854248
    },
    {
      "auxiliary_loss_clip": 0.01150512,
      "auxiliary_loss_mlp": 0.01033787,
      "balance_loss_clip": 1.05346119,
      "balance_loss_mlp": 1.02520955,
      "epoch": 0.5679071724884266,
      "flos": 28037256157440.0,
      "grad_norm": 2.517280932704674,
      "language_loss": 0.82435679,
      "learning_rate": 1.658927703398025e-06,
      "loss": 0.84619969,
      "num_input_tokens_seen": 101974390,
      "step": 4723,
      "time_per_iteration": 2.7243621349334717
    },
    {
      "auxiliary_loss_clip": 0.01128304,
      "auxiliary_loss_mlp": 0.01028155,
      "balance_loss_clip": 1.04943013,
      "balance_loss_mlp": 1.01979828,
      "epoch": 0.5680274153790658,
      "flos": 23550110380800.0,
      "grad_norm": 2.5201837386715296,
      "language_loss": 0.78298187,
      "learning_rate": 1.6581601684683309e-06,
      "loss": 0.80454648,
      "num_input_tokens_seen": 101994815,
      "step": 4724,
      "time_per_iteration": 2.776250123977661
    },
    {
      "auxiliary_loss_clip": 0.01164318,
      "auxiliary_loss_mlp": 0.01034607,
      "balance_loss_clip": 1.05456328,
      "balance_loss_mlp": 1.02567232,
      "epoch": 0.5681476582697048,
      "flos": 22455158140800.0,
      "grad_norm": 2.718944493265181,
      "language_loss": 0.68487895,
      "learning_rate": 1.6573926853954674e-06,
      "loss": 0.70686817,
      "num_input_tokens_seen": 102012400,
      "step": 4725,
      "time_per_iteration": 2.6655211448669434
    },
    {
      "auxiliary_loss_clip": 0.0115723,
      "auxiliary_loss_mlp": 0.01032458,
      "balance_loss_clip": 1.05602407,
      "balance_loss_mlp": 1.02456594,
      "epoch": 0.5682679011603439,
      "flos": 19536913584000.0,
      "grad_norm": 1.8970248824945253,
      "language_loss": 0.83037561,
      "learning_rate": 1.6566252542958608e-06,
      "loss": 0.85227251,
      "num_input_tokens_seen": 102031900,
      "step": 4726,
      "time_per_iteration": 2.703519105911255
    },
    {
      "auxiliary_loss_clip": 0.0118784,
      "auxiliary_loss_mlp": 0.01034784,
      "balance_loss_clip": 1.05844331,
      "balance_loss_mlp": 1.02610016,
      "epoch": 0.568388144050983,
      "flos": 28765488493440.0,
      "grad_norm": 2.306419429389782,
      "language_loss": 0.78653967,
      "learning_rate": 1.6558578752859305e-06,
      "loss": 0.80876589,
      "num_input_tokens_seen": 102050860,
      "step": 4727,
      "time_per_iteration": 2.656843662261963
    },
    {
      "auxiliary_loss_clip": 0.01189029,
      "auxiliary_loss_mlp": 0.0104074,
      "balance_loss_clip": 1.06003928,
      "balance_loss_mlp": 1.0322938,
      "epoch": 0.5685083869416221,
      "flos": 21209452519680.0,
      "grad_norm": 2.7564105828705725,
      "language_loss": 0.78981841,
      "learning_rate": 1.6550905484820865e-06,
      "loss": 0.81211603,
      "num_input_tokens_seen": 102069320,
      "step": 4728,
      "time_per_iteration": 2.637528896331787
    },
    {
      "auxiliary_loss_clip": 0.01175656,
      "auxiliary_loss_mlp": 0.01035769,
      "balance_loss_clip": 1.05712366,
      "balance_loss_mlp": 1.02681613,
      "epoch": 0.5686286298322611,
      "flos": 24827021942400.0,
      "grad_norm": 3.338084813424666,
      "language_loss": 0.78800631,
      "learning_rate": 1.6543232740007328e-06,
      "loss": 0.81012064,
      "num_input_tokens_seen": 102086435,
      "step": 4729,
      "time_per_iteration": 2.686702251434326
    },
    {
      "auxiliary_loss_clip": 0.01189952,
      "auxiliary_loss_mlp": 0.01035102,
      "balance_loss_clip": 1.05974054,
      "balance_loss_mlp": 1.02675152,
      "epoch": 0.5687488727229003,
      "flos": 26615121909120.0,
      "grad_norm": 2.9146324290208994,
      "language_loss": 0.67455971,
      "learning_rate": 1.653556051958263e-06,
      "loss": 0.69681025,
      "num_input_tokens_seen": 102106115,
      "step": 4730,
      "time_per_iteration": 2.706664562225342
    },
    {
      "auxiliary_loss_clip": 0.01174502,
      "auxiliary_loss_mlp": 0.01032044,
      "balance_loss_clip": 1.05527079,
      "balance_loss_mlp": 1.02337122,
      "epoch": 0.5688691156135394,
      "flos": 20808725414400.0,
      "grad_norm": 2.043766104717023,
      "language_loss": 0.73926735,
      "learning_rate": 1.6527888824710642e-06,
      "loss": 0.76133281,
      "num_input_tokens_seen": 102125715,
      "step": 4731,
      "time_per_iteration": 3.571708917617798
    },
    {
      "auxiliary_loss_clip": 0.01167494,
      "auxiliary_loss_mlp": 0.01030002,
      "balance_loss_clip": 1.05509925,
      "balance_loss_mlp": 1.02091217,
      "epoch": 0.5689893585041784,
      "flos": 25880963829120.0,
      "grad_norm": 2.439036545416874,
      "language_loss": 0.76864231,
      "learning_rate": 1.6520217656555166e-06,
      "loss": 0.79061729,
      "num_input_tokens_seen": 102145005,
      "step": 4732,
      "time_per_iteration": 2.714082717895508
    },
    {
      "auxiliary_loss_clip": 0.01159193,
      "auxiliary_loss_mlp": 0.01030786,
      "balance_loss_clip": 1.05729365,
      "balance_loss_mlp": 1.02247167,
      "epoch": 0.5691096013948175,
      "flos": 23477463123840.0,
      "grad_norm": 1.6021432630830201,
      "language_loss": 0.71028394,
      "learning_rate": 1.65125470162799e-06,
      "loss": 0.73218369,
      "num_input_tokens_seen": 102165360,
      "step": 4733,
      "time_per_iteration": 2.7634198665618896
    },
    {
      "auxiliary_loss_clip": 0.0117569,
      "auxiliary_loss_mlp": 0.01035796,
      "balance_loss_clip": 1.05448198,
      "balance_loss_mlp": 1.02687287,
      "epoch": 0.5692298442854566,
      "flos": 18075600576000.0,
      "grad_norm": 4.417928156694791,
      "language_loss": 0.70048404,
      "learning_rate": 1.6504876905048485e-06,
      "loss": 0.72259885,
      "num_input_tokens_seen": 102182320,
      "step": 4734,
      "time_per_iteration": 2.586452007293701
    },
    {
      "auxiliary_loss_clip": 0.01157855,
      "auxiliary_loss_mlp": 0.01034453,
      "balance_loss_clip": 1.05235684,
      "balance_loss_mlp": 1.02715397,
      "epoch": 0.5693500871760957,
      "flos": 23039317025280.0,
      "grad_norm": 1.9835382594294282,
      "language_loss": 0.72477758,
      "learning_rate": 1.6497207324024464e-06,
      "loss": 0.74670064,
      "num_input_tokens_seen": 102201220,
      "step": 4735,
      "time_per_iteration": 2.7220458984375
    },
    {
      "auxiliary_loss_clip": 0.011875,
      "auxiliary_loss_mlp": 0.01027147,
      "balance_loss_clip": 1.05801368,
      "balance_loss_mlp": 1.01824784,
      "epoch": 0.5694703300667348,
      "flos": 18989670902400.0,
      "grad_norm": 2.113876370602585,
      "language_loss": 0.82618856,
      "learning_rate": 1.6489538274371305e-06,
      "loss": 0.84833503,
      "num_input_tokens_seen": 102219825,
      "step": 4736,
      "time_per_iteration": 2.575838804244995
    },
    {
      "auxiliary_loss_clip": 0.01169161,
      "auxiliary_loss_mlp": 0.01030041,
      "balance_loss_clip": 1.05372906,
      "balance_loss_mlp": 1.02194691,
      "epoch": 0.5695905729573739,
      "flos": 21908705558400.0,
      "grad_norm": 2.2714601116811957,
      "language_loss": 0.83017194,
      "learning_rate": 1.6481869757252396e-06,
      "loss": 0.85216391,
      "num_input_tokens_seen": 102238160,
      "step": 4737,
      "time_per_iteration": 2.661827325820923
    },
    {
      "auxiliary_loss_clip": 0.0118612,
      "auxiliary_loss_mlp": 0.01036272,
      "balance_loss_clip": 1.05852079,
      "balance_loss_mlp": 1.02817822,
      "epoch": 0.569710815848013,
      "flos": 28476659232000.0,
      "grad_norm": 1.7836154848205388,
      "language_loss": 0.71643507,
      "learning_rate": 1.647420177383105e-06,
      "loss": 0.73865896,
      "num_input_tokens_seen": 102261030,
      "step": 4738,
      "time_per_iteration": 2.6519124507904053
    },
    {
      "auxiliary_loss_clip": 0.01161399,
      "auxiliary_loss_mlp": 0.01037372,
      "balance_loss_clip": 1.05498719,
      "balance_loss_mlp": 1.02933168,
      "epoch": 0.569831058738652,
      "flos": 28366162018560.0,
      "grad_norm": 1.791771683937763,
      "language_loss": 0.72942126,
      "learning_rate": 1.646653432527049e-06,
      "loss": 0.75140893,
      "num_input_tokens_seen": 102281670,
      "step": 4739,
      "time_per_iteration": 2.7183682918548584
    },
    {
      "auxiliary_loss_clip": 0.01166883,
      "auxiliary_loss_mlp": 0.00763344,
      "balance_loss_clip": 1.05638313,
      "balance_loss_mlp": 1.00143123,
      "epoch": 0.5699513016292912,
      "flos": 25849973370240.0,
      "grad_norm": 2.775414306818618,
      "language_loss": 0.74302769,
      "learning_rate": 1.645886741273387e-06,
      "loss": 0.76232994,
      "num_input_tokens_seen": 102303485,
      "step": 4740,
      "time_per_iteration": 2.676187038421631
    },
    {
      "auxiliary_loss_clip": 0.01170622,
      "auxiliary_loss_mlp": 0.01026527,
      "balance_loss_clip": 1.05577576,
      "balance_loss_mlp": 1.01880217,
      "epoch": 0.5700715445199303,
      "flos": 18037858360320.0,
      "grad_norm": 2.3384085354857556,
      "language_loss": 0.74341047,
      "learning_rate": 1.645120103738424e-06,
      "loss": 0.76538193,
      "num_input_tokens_seen": 102320995,
      "step": 4741,
      "time_per_iteration": 2.6455883979797363
    },
    {
      "auxiliary_loss_clip": 0.01181364,
      "auxiliary_loss_mlp": 0.01025082,
      "balance_loss_clip": 1.05630136,
      "balance_loss_mlp": 1.01750374,
      "epoch": 0.5701917874105693,
      "flos": 11473352392320.0,
      "grad_norm": 3.1882951490661537,
      "language_loss": 0.8388148,
      "learning_rate": 1.6443535200384591e-06,
      "loss": 0.86087918,
      "num_input_tokens_seen": 102339170,
      "step": 4742,
      "time_per_iteration": 2.5940356254577637
    },
    {
      "auxiliary_loss_clip": 0.01169941,
      "auxiliary_loss_mlp": 0.01029875,
      "balance_loss_clip": 1.05683732,
      "balance_loss_mlp": 1.02114856,
      "epoch": 0.5703120303012085,
      "flos": 21761759018880.0,
      "grad_norm": 2.449297178390394,
      "language_loss": 0.70582634,
      "learning_rate": 1.6435869902897827e-06,
      "loss": 0.72782445,
      "num_input_tokens_seen": 102357750,
      "step": 4743,
      "time_per_iteration": 2.704873561859131
    },
    {
      "auxiliary_loss_clip": 0.01071881,
      "auxiliary_loss_mlp": 0.00999936,
      "balance_loss_clip": 1.01165175,
      "balance_loss_mlp": 0.99899977,
      "epoch": 0.5704322731918475,
      "flos": 56746258513920.0,
      "grad_norm": 0.7962136801078376,
      "language_loss": 0.62052596,
      "learning_rate": 1.6428205146086764e-06,
      "loss": 0.64124417,
      "num_input_tokens_seen": 102419730,
      "step": 4744,
      "time_per_iteration": 3.2359325885772705
    },
    {
      "auxiliary_loss_clip": 0.01187815,
      "auxiliary_loss_mlp": 0.01029508,
      "balance_loss_clip": 1.0574919,
      "balance_loss_mlp": 1.02110946,
      "epoch": 0.5705525160824866,
      "flos": 20741141975040.0,
      "grad_norm": 1.9201612502025753,
      "language_loss": 0.71004945,
      "learning_rate": 1.6420540931114142e-06,
      "loss": 0.73222268,
      "num_input_tokens_seen": 102440320,
      "step": 4745,
      "time_per_iteration": 2.7450530529022217
    },
    {
      "auxiliary_loss_clip": 0.01156167,
      "auxiliary_loss_mlp": 0.01034295,
      "balance_loss_clip": 1.05237007,
      "balance_loss_mlp": 1.02557492,
      "epoch": 0.5706727589731257,
      "flos": 18771262254720.0,
      "grad_norm": 1.956745019447067,
      "language_loss": 0.78970575,
      "learning_rate": 1.6412877259142616e-06,
      "loss": 0.81161034,
      "num_input_tokens_seen": 102460240,
      "step": 4746,
      "time_per_iteration": 3.709456205368042
    },
    {
      "auxiliary_loss_clip": 0.01175219,
      "auxiliary_loss_mlp": 0.01033121,
      "balance_loss_clip": 1.0551151,
      "balance_loss_mlp": 1.02542603,
      "epoch": 0.5707930018637648,
      "flos": 27634733372160.0,
      "grad_norm": 2.0864243010610606,
      "language_loss": 0.73928607,
      "learning_rate": 1.6405214131334757e-06,
      "loss": 0.76136947,
      "num_input_tokens_seen": 102478765,
      "step": 4747,
      "time_per_iteration": 3.588549852371216
    },
    {
      "auxiliary_loss_clip": 0.01187871,
      "auxiliary_loss_mlp": 0.01031067,
      "balance_loss_clip": 1.05946064,
      "balance_loss_mlp": 1.02301991,
      "epoch": 0.5709132447544039,
      "flos": 27597673514880.0,
      "grad_norm": 2.0686077058798724,
      "language_loss": 0.80425668,
      "learning_rate": 1.6397551548853052e-06,
      "loss": 0.82644606,
      "num_input_tokens_seen": 102496930,
      "step": 4748,
      "time_per_iteration": 3.5633034706115723
    },
    {
      "auxiliary_loss_clip": 0.0118681,
      "auxiliary_loss_mlp": 0.01031034,
      "balance_loss_clip": 1.05534053,
      "balance_loss_mlp": 1.02181888,
      "epoch": 0.571033487645043,
      "flos": 21686095019520.0,
      "grad_norm": 1.7760327239811897,
      "language_loss": 0.71008945,
      "learning_rate": 1.6389889512859917e-06,
      "loss": 0.73226798,
      "num_input_tokens_seen": 102516590,
      "step": 4749,
      "time_per_iteration": 2.672755002975464
    },
    {
      "auxiliary_loss_clip": 0.01043828,
      "auxiliary_loss_mlp": 0.01001901,
      "balance_loss_clip": 1.01032567,
      "balance_loss_mlp": 1.0010246,
      "epoch": 0.5711537305356821,
      "flos": 70181445980160.0,
      "grad_norm": 0.8118663258960831,
      "language_loss": 0.60336328,
      "learning_rate": 1.638222802451767e-06,
      "loss": 0.62382054,
      "num_input_tokens_seen": 102578070,
      "step": 4750,
      "time_per_iteration": 3.2475104331970215
    },
    {
      "auxiliary_loss_clip": 0.0115194,
      "auxiliary_loss_mlp": 0.0102934,
      "balance_loss_clip": 1.0554688,
      "balance_loss_mlp": 1.02116263,
      "epoch": 0.5712739734263211,
      "flos": 24717494396160.0,
      "grad_norm": 1.8521992841017096,
      "language_loss": 0.75240242,
      "learning_rate": 1.6374567084988561e-06,
      "loss": 0.77421522,
      "num_input_tokens_seen": 102599255,
      "step": 4751,
      "time_per_iteration": 2.789700746536255
    },
    {
      "auxiliary_loss_clip": 0.01146203,
      "auxiliary_loss_mlp": 0.0076296,
      "balance_loss_clip": 1.05374026,
      "balance_loss_mlp": 1.00118971,
      "epoch": 0.5713942163169603,
      "flos": 26578169792640.0,
      "grad_norm": 1.9161461279094056,
      "language_loss": 0.76906312,
      "learning_rate": 1.6366906695434738e-06,
      "loss": 0.78815472,
      "num_input_tokens_seen": 102621775,
      "step": 4752,
      "time_per_iteration": 2.755354166030884
    },
    {
      "auxiliary_loss_clip": 0.01164476,
      "auxiliary_loss_mlp": 0.01033714,
      "balance_loss_clip": 1.05643296,
      "balance_loss_mlp": 1.02563202,
      "epoch": 0.5715144592075994,
      "flos": 21142443697920.0,
      "grad_norm": 2.3631458872824207,
      "language_loss": 0.86152327,
      "learning_rate": 1.6359246857018275e-06,
      "loss": 0.88350517,
      "num_input_tokens_seen": 102639305,
      "step": 4753,
      "time_per_iteration": 2.6725428104400635
    },
    {
      "auxiliary_loss_clip": 0.01187701,
      "auxiliary_loss_mlp": 0.01029356,
      "balance_loss_clip": 1.05823445,
      "balance_loss_mlp": 1.02099919,
      "epoch": 0.5716347020982384,
      "flos": 23330265189120.0,
      "grad_norm": 5.342092968905558,
      "language_loss": 0.78548443,
      "learning_rate": 1.6351587570901178e-06,
      "loss": 0.80765498,
      "num_input_tokens_seen": 102659430,
      "step": 4754,
      "time_per_iteration": 2.6536035537719727
    },
    {
      "auxiliary_loss_clip": 0.01174848,
      "auxiliary_loss_mlp": 0.01029784,
      "balance_loss_clip": 1.05503893,
      "balance_loss_mlp": 1.02144575,
      "epoch": 0.5717549449888776,
      "flos": 17009555806080.0,
      "grad_norm": 2.4179441943134745,
      "language_loss": 0.7562201,
      "learning_rate": 1.634392883824534e-06,
      "loss": 0.77826643,
      "num_input_tokens_seen": 102671430,
      "step": 4755,
      "time_per_iteration": 2.624983072280884
    },
    {
      "auxiliary_loss_clip": 0.01174467,
      "auxiliary_loss_mlp": 0.01042379,
      "balance_loss_clip": 1.05849862,
      "balance_loss_mlp": 1.0333488,
      "epoch": 0.5718751878795166,
      "flos": 35518130922240.0,
      "grad_norm": 1.702200155524711,
      "language_loss": 0.68168962,
      "learning_rate": 1.6336270660212595e-06,
      "loss": 0.70385814,
      "num_input_tokens_seen": 102693025,
      "step": 4756,
      "time_per_iteration": 2.787978172302246
    },
    {
      "auxiliary_loss_clip": 0.01179355,
      "auxiliary_loss_mlp": 0.01029222,
      "balance_loss_clip": 1.05813301,
      "balance_loss_mlp": 1.01972079,
      "epoch": 0.5719954307701557,
      "flos": 38613989255040.0,
      "grad_norm": 2.7708249088886343,
      "language_loss": 0.66212201,
      "learning_rate": 1.6328613037964676e-06,
      "loss": 0.6842078,
      "num_input_tokens_seen": 102716090,
      "step": 4757,
      "time_per_iteration": 3.7272958755493164
    },
    {
      "auxiliary_loss_clip": 0.01171961,
      "auxiliary_loss_mlp": 0.01035601,
      "balance_loss_clip": 1.0524776,
      "balance_loss_mlp": 1.02675605,
      "epoch": 0.5721156736607949,
      "flos": 20631111638400.0,
      "grad_norm": 1.7334793767473313,
      "language_loss": 0.68106747,
      "learning_rate": 1.6320955972663241e-06,
      "loss": 0.70314312,
      "num_input_tokens_seen": 102735685,
      "step": 4758,
      "time_per_iteration": 2.6423051357269287
    },
    {
      "auxiliary_loss_clip": 0.01184965,
      "auxiliary_loss_mlp": 0.01031357,
      "balance_loss_clip": 1.05485368,
      "balance_loss_mlp": 1.02297699,
      "epoch": 0.5722359165514339,
      "flos": 37415076076800.0,
      "grad_norm": 18.546122108985468,
      "language_loss": 0.65579402,
      "learning_rate": 1.6313299465469857e-06,
      "loss": 0.67795724,
      "num_input_tokens_seen": 102758415,
      "step": 4759,
      "time_per_iteration": 2.8194103240966797
    },
    {
      "auxiliary_loss_clip": 0.01188302,
      "auxiliary_loss_mlp": 0.01032943,
      "balance_loss_clip": 1.05781376,
      "balance_loss_mlp": 1.02401423,
      "epoch": 0.572356159442073,
      "flos": 21972877205760.0,
      "grad_norm": 2.8537361469678184,
      "language_loss": 0.79696405,
      "learning_rate": 1.6305643517546014e-06,
      "loss": 0.81917655,
      "num_input_tokens_seen": 102773795,
      "step": 4760,
      "time_per_iteration": 2.5561790466308594
    },
    {
      "auxiliary_loss_clip": 0.01119897,
      "auxiliary_loss_mlp": 0.01028669,
      "balance_loss_clip": 1.04288101,
      "balance_loss_mlp": 1.02025878,
      "epoch": 0.5724764023327121,
      "flos": 19135540033920.0,
      "grad_norm": 1.8557155082470127,
      "language_loss": 0.84828579,
      "learning_rate": 1.629798813005311e-06,
      "loss": 0.86977148,
      "num_input_tokens_seen": 102793515,
      "step": 4761,
      "time_per_iteration": 2.722791910171509
    },
    {
      "auxiliary_loss_clip": 0.01188788,
      "auxiliary_loss_mlp": 0.0102955,
      "balance_loss_clip": 1.0595175,
      "balance_loss_mlp": 1.02152443,
      "epoch": 0.5725966452233512,
      "flos": 22819759142400.0,
      "grad_norm": 8.080717449123986,
      "language_loss": 0.71179128,
      "learning_rate": 1.6290333304152473e-06,
      "loss": 0.73397458,
      "num_input_tokens_seen": 102813390,
      "step": 4762,
      "time_per_iteration": 2.5924441814422607
    },
    {
      "auxiliary_loss_clip": 0.01173581,
      "auxiliary_loss_mlp": 0.01029172,
      "balance_loss_clip": 1.05484343,
      "balance_loss_mlp": 1.02064312,
      "epoch": 0.5727168881139902,
      "flos": 41496610498560.0,
      "grad_norm": 1.9956329267580823,
      "language_loss": 0.57203686,
      "learning_rate": 1.6282679041005314e-06,
      "loss": 0.59406441,
      "num_input_tokens_seen": 102838980,
      "step": 4763,
      "time_per_iteration": 2.8464887142181396
    },
    {
      "auxiliary_loss_clip": 0.011685,
      "auxiliary_loss_mlp": 0.01029303,
      "balance_loss_clip": 1.0545783,
      "balance_loss_mlp": 1.02063084,
      "epoch": 0.5728371310046293,
      "flos": 14647675985280.0,
      "grad_norm": 3.3566613055318624,
      "language_loss": 0.87236756,
      "learning_rate": 1.6275025341772789e-06,
      "loss": 0.89434558,
      "num_input_tokens_seen": 102855285,
      "step": 4764,
      "time_per_iteration": 2.649409294128418
    },
    {
      "auxiliary_loss_clip": 0.01144153,
      "auxiliary_loss_mlp": 0.01031754,
      "balance_loss_clip": 1.05248833,
      "balance_loss_mlp": 1.02257538,
      "epoch": 0.5729573738952685,
      "flos": 21506613736320.0,
      "grad_norm": 2.9482027149001957,
      "language_loss": 0.81813633,
      "learning_rate": 1.626737220761596e-06,
      "loss": 0.83989543,
      "num_input_tokens_seen": 102872750,
      "step": 4765,
      "time_per_iteration": 2.722421407699585
    },
    {
      "auxiliary_loss_clip": 0.01174953,
      "auxiliary_loss_mlp": 0.01031916,
      "balance_loss_clip": 1.05995989,
      "balance_loss_mlp": 1.02333927,
      "epoch": 0.5730776167859075,
      "flos": 23621680229760.0,
      "grad_norm": 1.8769168126631464,
      "language_loss": 0.79241824,
      "learning_rate": 1.62597196396958e-06,
      "loss": 0.81448698,
      "num_input_tokens_seen": 102890920,
      "step": 4766,
      "time_per_iteration": 2.6176912784576416
    },
    {
      "auxiliary_loss_clip": 0.01160527,
      "auxiliary_loss_mlp": 0.01026589,
      "balance_loss_clip": 1.05187702,
      "balance_loss_mlp": 1.01805353,
      "epoch": 0.5731978596765466,
      "flos": 25739224761600.0,
      "grad_norm": 1.8954253391317295,
      "language_loss": 0.85756433,
      "learning_rate": 1.6252067639173197e-06,
      "loss": 0.87943554,
      "num_input_tokens_seen": 102912830,
      "step": 4767,
      "time_per_iteration": 2.7807915210723877
    },
    {
      "auxiliary_loss_clip": 0.01176117,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 1.05648088,
      "balance_loss_mlp": 1.01930881,
      "epoch": 0.5733181025671857,
      "flos": 26359509749760.0,
      "grad_norm": 1.9593520252499441,
      "language_loss": 0.69963014,
      "learning_rate": 1.6244416207208956e-06,
      "loss": 0.72167253,
      "num_input_tokens_seen": 102933765,
      "step": 4768,
      "time_per_iteration": 2.684128999710083
    },
    {
      "auxiliary_loss_clip": 0.01162821,
      "auxiliary_loss_mlp": 0.01035743,
      "balance_loss_clip": 1.0570637,
      "balance_loss_mlp": 1.02696896,
      "epoch": 0.5734383454578248,
      "flos": 29423874833280.0,
      "grad_norm": 1.6626524893085182,
      "language_loss": 0.73839927,
      "learning_rate": 1.6236765344963787e-06,
      "loss": 0.76038492,
      "num_input_tokens_seen": 102955025,
      "step": 4769,
      "time_per_iteration": 2.745209217071533
    },
    {
      "auxiliary_loss_clip": 0.0116172,
      "auxiliary_loss_mlp": 0.01030523,
      "balance_loss_clip": 1.05514336,
      "balance_loss_mlp": 1.02146363,
      "epoch": 0.5735585883484638,
      "flos": 34969954487040.0,
      "grad_norm": 2.2845526632752677,
      "language_loss": 0.69324756,
      "learning_rate": 1.6229115053598322e-06,
      "loss": 0.71516997,
      "num_input_tokens_seen": 102976780,
      "step": 4770,
      "time_per_iteration": 2.87479567527771
    },
    {
      "auxiliary_loss_clip": 0.01192805,
      "auxiliary_loss_mlp": 0.0103106,
      "balance_loss_clip": 1.06166089,
      "balance_loss_mlp": 1.02213717,
      "epoch": 0.573678831239103,
      "flos": 18770759464320.0,
      "grad_norm": 8.294998846569303,
      "language_loss": 0.72222942,
      "learning_rate": 1.6221465334273108e-06,
      "loss": 0.74446809,
      "num_input_tokens_seen": 102995990,
      "step": 4771,
      "time_per_iteration": 2.608417510986328
    },
    {
      "auxiliary_loss_clip": 0.0116199,
      "auxiliary_loss_mlp": 0.0103187,
      "balance_loss_clip": 1.05397105,
      "balance_loss_mlp": 1.02340603,
      "epoch": 0.5737990741297421,
      "flos": 25702883176320.0,
      "grad_norm": 2.0214644295523314,
      "language_loss": 0.62125242,
      "learning_rate": 1.6213816188148593e-06,
      "loss": 0.64319098,
      "num_input_tokens_seen": 103014695,
      "step": 4772,
      "time_per_iteration": 4.635494947433472
    },
    {
      "auxiliary_loss_clip": 0.01151206,
      "auxiliary_loss_mlp": 0.01030197,
      "balance_loss_clip": 1.05648828,
      "balance_loss_mlp": 1.02163792,
      "epoch": 0.5739193170203811,
      "flos": 27269234530560.0,
      "grad_norm": 4.158577277874725,
      "language_loss": 0.77249277,
      "learning_rate": 1.6206167616385162e-06,
      "loss": 0.79430687,
      "num_input_tokens_seen": 103035760,
      "step": 4773,
      "time_per_iteration": 2.775782823562622
    },
    {
      "auxiliary_loss_clip": 0.01161609,
      "auxiliary_loss_mlp": 0.01034189,
      "balance_loss_clip": 1.059412,
      "balance_loss_mlp": 1.02495074,
      "epoch": 0.5740395599110203,
      "flos": 12239721993600.0,
      "grad_norm": 2.2982764066063277,
      "language_loss": 0.73660946,
      "learning_rate": 1.6198519620143078e-06,
      "loss": 0.75856745,
      "num_input_tokens_seen": 103052915,
      "step": 4774,
      "time_per_iteration": 3.689227819442749
    },
    {
      "auxiliary_loss_clip": 0.01144038,
      "auxiliary_loss_mlp": 0.00762687,
      "balance_loss_clip": 1.05377507,
      "balance_loss_mlp": 1.00138116,
      "epoch": 0.5741598028016593,
      "flos": 25921399564800.0,
      "grad_norm": 1.9716251248975118,
      "language_loss": 0.78262937,
      "learning_rate": 1.6190872200582546e-06,
      "loss": 0.80169666,
      "num_input_tokens_seen": 103074655,
      "step": 4775,
      "time_per_iteration": 2.812562942504883
    },
    {
      "auxiliary_loss_clip": 0.01176054,
      "auxiliary_loss_mlp": 0.01034061,
      "balance_loss_clip": 1.05790949,
      "balance_loss_mlp": 1.02560318,
      "epoch": 0.5742800456922984,
      "flos": 19244133826560.0,
      "grad_norm": 2.176799823908845,
      "language_loss": 0.7828728,
      "learning_rate": 1.6183225358863676e-06,
      "loss": 0.80497396,
      "num_input_tokens_seen": 103091550,
      "step": 4776,
      "time_per_iteration": 2.5887296199798584
    },
    {
      "auxiliary_loss_clip": 0.01137288,
      "auxiliary_loss_mlp": 0.00763001,
      "balance_loss_clip": 1.04993951,
      "balance_loss_mlp": 1.00135159,
      "epoch": 0.5744002885829376,
      "flos": 30920487932160.0,
      "grad_norm": 2.7937660973242835,
      "language_loss": 0.7167474,
      "learning_rate": 1.617557909614648e-06,
      "loss": 0.73575032,
      "num_input_tokens_seen": 103110985,
      "step": 4777,
      "time_per_iteration": 2.8212320804595947
    },
    {
      "auxiliary_loss_clip": 0.01175931,
      "auxiliary_loss_mlp": 0.01031614,
      "balance_loss_clip": 1.05533779,
      "balance_loss_mlp": 1.02337623,
      "epoch": 0.5745205314735766,
      "flos": 23840017050240.0,
      "grad_norm": 1.9319160120930736,
      "language_loss": 0.86043,
      "learning_rate": 1.6167933413590899e-06,
      "loss": 0.88250542,
      "num_input_tokens_seen": 103129890,
      "step": 4778,
      "time_per_iteration": 2.7148191928863525
    },
    {
      "auxiliary_loss_clip": 0.01178085,
      "auxiliary_loss_mlp": 0.01026119,
      "balance_loss_clip": 1.0590384,
      "balance_loss_mlp": 1.01804256,
      "epoch": 0.5746407743642157,
      "flos": 12311902373760.0,
      "grad_norm": 3.028758883929261,
      "language_loss": 0.91075003,
      "learning_rate": 1.6160288312356773e-06,
      "loss": 0.93279207,
      "num_input_tokens_seen": 103147020,
      "step": 4779,
      "time_per_iteration": 2.6356289386749268
    },
    {
      "auxiliary_loss_clip": 0.01187902,
      "auxiliary_loss_mlp": 0.01029538,
      "balance_loss_clip": 1.05735159,
      "balance_loss_mlp": 1.02124095,
      "epoch": 0.5747610172548548,
      "flos": 24133658734080.0,
      "grad_norm": 2.556729221696292,
      "language_loss": 0.81734598,
      "learning_rate": 1.6152643793603857e-06,
      "loss": 0.83952045,
      "num_input_tokens_seen": 103167370,
      "step": 4780,
      "time_per_iteration": 2.668020486831665
    },
    {
      "auxiliary_loss_clip": 0.01187591,
      "auxiliary_loss_mlp": 0.01030409,
      "balance_loss_clip": 1.05889678,
      "balance_loss_mlp": 1.02171278,
      "epoch": 0.5748812601454939,
      "flos": 25408451393280.0,
      "grad_norm": 1.7127337566204648,
      "language_loss": 0.87592822,
      "learning_rate": 1.6144999858491815e-06,
      "loss": 0.89810824,
      "num_input_tokens_seen": 103186000,
      "step": 4781,
      "time_per_iteration": 2.6603314876556396
    },
    {
      "auxiliary_loss_clip": 0.01191006,
      "auxiliary_loss_mlp": 0.01029094,
      "balance_loss_clip": 1.05774355,
      "balance_loss_mlp": 1.0195272,
      "epoch": 0.575001503036133,
      "flos": 30624942827520.0,
      "grad_norm": 1.6384618696890727,
      "language_loss": 0.8617394,
      "learning_rate": 1.6137356508180232e-06,
      "loss": 0.88394046,
      "num_input_tokens_seen": 103207710,
      "step": 4782,
      "time_per_iteration": 2.6879687309265137
    },
    {
      "auxiliary_loss_clip": 0.01184296,
      "auxiliary_loss_mlp": 0.01034638,
      "balance_loss_clip": 1.05472493,
      "balance_loss_mlp": 1.02557838,
      "epoch": 0.5751217459267721,
      "flos": 21726566668800.0,
      "grad_norm": 2.9477531830377903,
      "language_loss": 0.81412691,
      "learning_rate": 1.6129713743828593e-06,
      "loss": 0.83631623,
      "num_input_tokens_seen": 103226720,
      "step": 4783,
      "time_per_iteration": 3.5063889026641846
    },
    {
      "auxiliary_loss_clip": 0.01172499,
      "auxiliary_loss_mlp": 0.00762895,
      "balance_loss_clip": 1.05487156,
      "balance_loss_mlp": 1.00114131,
      "epoch": 0.5752419888174112,
      "flos": 21651620941440.0,
      "grad_norm": 1.5349542148825221,
      "language_loss": 0.75789654,
      "learning_rate": 1.6122071566596306e-06,
      "loss": 0.77725053,
      "num_input_tokens_seen": 103246995,
      "step": 4784,
      "time_per_iteration": 2.6666228771209717
    },
    {
      "auxiliary_loss_clip": 0.01175098,
      "auxiliary_loss_mlp": 0.01033872,
      "balance_loss_clip": 1.05659962,
      "balance_loss_mlp": 1.02502704,
      "epoch": 0.5753622317080502,
      "flos": 17775997234560.0,
      "grad_norm": 2.342651679723281,
      "language_loss": 0.82936001,
      "learning_rate": 1.6114429977642674e-06,
      "loss": 0.85144973,
      "num_input_tokens_seen": 103261500,
      "step": 4785,
      "time_per_iteration": 2.5499186515808105
    },
    {
      "auxiliary_loss_clip": 0.01180241,
      "auxiliary_loss_mlp": 0.01026344,
      "balance_loss_clip": 1.05859613,
      "balance_loss_mlp": 1.01708794,
      "epoch": 0.5754824745986894,
      "flos": 19789616741760.0,
      "grad_norm": 2.0635537178544285,
      "language_loss": 0.73943281,
      "learning_rate": 1.6106788978126926e-06,
      "loss": 0.76149869,
      "num_input_tokens_seen": 103280475,
      "step": 4786,
      "time_per_iteration": 2.7213761806488037
    },
    {
      "auxiliary_loss_clip": 0.01153469,
      "auxiliary_loss_mlp": 0.00763016,
      "balance_loss_clip": 1.0540123,
      "balance_loss_mlp": 1.00116634,
      "epoch": 0.5756027174893285,
      "flos": 30985665160320.0,
      "grad_norm": 2.4195184754490597,
      "language_loss": 0.79077333,
      "learning_rate": 1.6099148569208196e-06,
      "loss": 0.80993819,
      "num_input_tokens_seen": 103297695,
      "step": 4787,
      "time_per_iteration": 2.747776985168457
    },
    {
      "auxiliary_loss_clip": 0.01189886,
      "auxiliary_loss_mlp": 0.01031491,
      "balance_loss_clip": 1.05831671,
      "balance_loss_mlp": 1.02256179,
      "epoch": 0.5757229603799675,
      "flos": 28546864364160.0,
      "grad_norm": 2.2962988588041027,
      "language_loss": 0.63227677,
      "learning_rate": 1.6091508752045523e-06,
      "loss": 0.65449059,
      "num_input_tokens_seen": 103318575,
      "step": 4788,
      "time_per_iteration": 2.6838977336883545
    },
    {
      "auxiliary_loss_clip": 0.01185798,
      "auxiliary_loss_mlp": 0.01029837,
      "balance_loss_clip": 1.05736971,
      "balance_loss_mlp": 1.02171922,
      "epoch": 0.5758432032706067,
      "flos": 22999024944000.0,
      "grad_norm": 1.7197750598158015,
      "language_loss": 0.86846995,
      "learning_rate": 1.608386952779787e-06,
      "loss": 0.89062631,
      "num_input_tokens_seen": 103337945,
      "step": 4789,
      "time_per_iteration": 2.6581411361694336
    },
    {
      "auxiliary_loss_clip": 0.01187007,
      "auxiliary_loss_mlp": 0.01038624,
      "balance_loss_clip": 1.05836844,
      "balance_loss_mlp": 1.03038049,
      "epoch": 0.5759634461612457,
      "flos": 25739727552000.0,
      "grad_norm": 2.356651412556104,
      "language_loss": 0.74596322,
      "learning_rate": 1.6076230897624098e-06,
      "loss": 0.76821953,
      "num_input_tokens_seen": 103360150,
      "step": 4790,
      "time_per_iteration": 2.598196268081665
    },
    {
      "auxiliary_loss_clip": 0.01161145,
      "auxiliary_loss_mlp": 0.01026258,
      "balance_loss_clip": 1.05310726,
      "balance_loss_mlp": 1.01748383,
      "epoch": 0.5760836890518848,
      "flos": 30591761639040.0,
      "grad_norm": 5.572763467239789,
      "language_loss": 0.77141988,
      "learning_rate": 1.6068592862682974e-06,
      "loss": 0.79329395,
      "num_input_tokens_seen": 103378305,
      "step": 4791,
      "time_per_iteration": 2.887887954711914
    },
    {
      "auxiliary_loss_clip": 0.0113369,
      "auxiliary_loss_mlp": 0.01032086,
      "balance_loss_clip": 1.05199945,
      "balance_loss_mlp": 1.02356887,
      "epoch": 0.576203931942524,
      "flos": 36538963447680.0,
      "grad_norm": 1.9433783970602005,
      "language_loss": 0.73775434,
      "learning_rate": 1.6060955424133187e-06,
      "loss": 0.75941217,
      "num_input_tokens_seen": 103399230,
      "step": 4792,
      "time_per_iteration": 2.887948751449585
    },
    {
      "auxiliary_loss_clip": 0.01190163,
      "auxiliary_loss_mlp": 0.0103241,
      "balance_loss_clip": 1.05948389,
      "balance_loss_mlp": 1.02379107,
      "epoch": 0.576324174833163,
      "flos": 25516937445120.0,
      "grad_norm": 1.8819313413787517,
      "language_loss": 0.89928567,
      "learning_rate": 1.6053318583133332e-06,
      "loss": 0.92151129,
      "num_input_tokens_seen": 103420100,
      "step": 4793,
      "time_per_iteration": 2.684718132019043
    },
    {
      "auxiliary_loss_clip": 0.01144748,
      "auxiliary_loss_mlp": 0.01028626,
      "balance_loss_clip": 1.04837751,
      "balance_loss_mlp": 1.01959038,
      "epoch": 0.5764444177238021,
      "flos": 25119262995840.0,
      "grad_norm": 2.117486291911164,
      "language_loss": 0.75014961,
      "learning_rate": 1.6045682340841907e-06,
      "loss": 0.77188337,
      "num_input_tokens_seen": 103439025,
      "step": 4794,
      "time_per_iteration": 2.7044179439544678
    },
    {
      "auxiliary_loss_clip": 0.01062522,
      "auxiliary_loss_mlp": 0.0100129,
      "balance_loss_clip": 1.01295495,
      "balance_loss_mlp": 1.00024068,
      "epoch": 0.5765646606144411,
      "flos": 62212687758720.0,
      "grad_norm": 0.7540972499721436,
      "language_loss": 0.57935792,
      "learning_rate": 1.6038046698417336e-06,
      "loss": 0.59999609,
      "num_input_tokens_seen": 103499920,
      "step": 4795,
      "time_per_iteration": 3.1806440353393555
    },
    {
      "auxiliary_loss_clip": 0.01164266,
      "auxiliary_loss_mlp": 0.01037022,
      "balance_loss_clip": 1.05333614,
      "balance_loss_mlp": 1.02802157,
      "epoch": 0.5766849035050803,
      "flos": 25118760205440.0,
      "grad_norm": 2.430795721892345,
      "language_loss": 0.69002342,
      "learning_rate": 1.6030411657017919e-06,
      "loss": 0.71203625,
      "num_input_tokens_seen": 103519575,
      "step": 4796,
      "time_per_iteration": 2.645263671875
    },
    {
      "auxiliary_loss_clip": 0.01163107,
      "auxiliary_loss_mlp": 0.01032336,
      "balance_loss_clip": 1.05355084,
      "balance_loss_mlp": 1.02409828,
      "epoch": 0.5768051463957193,
      "flos": 15991093578240.0,
      "grad_norm": 4.0541521076110465,
      "language_loss": 0.84498203,
      "learning_rate": 1.6022777217801903e-06,
      "loss": 0.86693645,
      "num_input_tokens_seen": 103536530,
      "step": 4797,
      "time_per_iteration": 2.6304357051849365
    },
    {
      "auxiliary_loss_clip": 0.01188301,
      "auxiliary_loss_mlp": 0.01028789,
      "balance_loss_clip": 1.05869985,
      "balance_loss_mlp": 1.01990247,
      "epoch": 0.5769253892863584,
      "flos": 22163635359360.0,
      "grad_norm": 2.72107921358793,
      "language_loss": 0.74049163,
      "learning_rate": 1.601514338192742e-06,
      "loss": 0.76266247,
      "num_input_tokens_seen": 103556460,
      "step": 4798,
      "time_per_iteration": 3.605069160461426
    },
    {
      "auxiliary_loss_clip": 0.01183305,
      "auxiliary_loss_mlp": 0.01029591,
      "balance_loss_clip": 1.05595231,
      "balance_loss_mlp": 1.02172959,
      "epoch": 0.5770456321769976,
      "flos": 22856388036480.0,
      "grad_norm": 2.297045599138009,
      "language_loss": 0.71574849,
      "learning_rate": 1.6007510150552514e-06,
      "loss": 0.73787743,
      "num_input_tokens_seen": 103574520,
      "step": 4799,
      "time_per_iteration": 3.751023530960083
    },
    {
      "auxiliary_loss_clip": 0.01165339,
      "auxiliary_loss_mlp": 0.01035649,
      "balance_loss_clip": 1.05458438,
      "balance_loss_mlp": 1.02658939,
      "epoch": 0.5771658750676366,
      "flos": 46353672489600.0,
      "grad_norm": 1.5733133215827968,
      "language_loss": 0.62479657,
      "learning_rate": 1.599987752483515e-06,
      "loss": 0.64680648,
      "num_input_tokens_seen": 103598965,
      "step": 4800,
      "time_per_iteration": 3.9431397914886475
    },
    {
      "auxiliary_loss_clip": 0.01177405,
      "auxiliary_loss_mlp": 0.01029607,
      "balance_loss_clip": 1.05913985,
      "balance_loss_mlp": 1.02110767,
      "epoch": 0.5772861179582757,
      "flos": 22159972172160.0,
      "grad_norm": 1.658534855915253,
      "language_loss": 0.68035173,
      "learning_rate": 1.5992245505933184e-06,
      "loss": 0.70242184,
      "num_input_tokens_seen": 103618665,
      "step": 4801,
      "time_per_iteration": 2.6651530265808105
    },
    {
      "auxiliary_loss_clip": 0.01188266,
      "auxiliary_loss_mlp": 0.01032721,
      "balance_loss_clip": 1.05904293,
      "balance_loss_mlp": 1.02432883,
      "epoch": 0.5774063608489148,
      "flos": 31248926916480.0,
      "grad_norm": 2.345691133033551,
      "language_loss": 0.7150737,
      "learning_rate": 1.5984614095004388e-06,
      "loss": 0.73728359,
      "num_input_tokens_seen": 103639800,
      "step": 4802,
      "time_per_iteration": 2.7956159114837646
    },
    {
      "auxiliary_loss_clip": 0.01160568,
      "auxiliary_loss_mlp": 0.01037187,
      "balance_loss_clip": 1.05349064,
      "balance_loss_mlp": 1.02873468,
      "epoch": 0.5775266037395539,
      "flos": 22527123039360.0,
      "grad_norm": 2.563395273703491,
      "language_loss": 0.81365412,
      "learning_rate": 1.5976983293206438e-06,
      "loss": 0.83563167,
      "num_input_tokens_seen": 103655605,
      "step": 4803,
      "time_per_iteration": 2.645519256591797
    },
    {
      "auxiliary_loss_clip": 0.01176316,
      "auxiliary_loss_mlp": 0.01037958,
      "balance_loss_clip": 1.05536962,
      "balance_loss_mlp": 1.0298512,
      "epoch": 0.577646846630193,
      "flos": 21068790860160.0,
      "grad_norm": 2.5555295479713376,
      "language_loss": 0.71406591,
      "learning_rate": 1.5969353101696928e-06,
      "loss": 0.73620856,
      "num_input_tokens_seen": 103674045,
      "step": 4804,
      "time_per_iteration": 2.5700459480285645
    },
    {
      "auxiliary_loss_clip": 0.01186918,
      "auxiliary_loss_mlp": 0.00763057,
      "balance_loss_clip": 1.05924535,
      "balance_loss_mlp": 1.00110376,
      "epoch": 0.5777670895208321,
      "flos": 29714284293120.0,
      "grad_norm": 1.6846790657964879,
      "language_loss": 0.80051893,
      "learning_rate": 1.5961723521633341e-06,
      "loss": 0.82001865,
      "num_input_tokens_seen": 103695285,
      "step": 4805,
      "time_per_iteration": 2.700695276260376
    },
    {
      "auxiliary_loss_clip": 0.01175124,
      "auxiliary_loss_mlp": 0.01041337,
      "balance_loss_clip": 1.05464053,
      "balance_loss_mlp": 1.03270638,
      "epoch": 0.5778873324114712,
      "flos": 19500428344320.0,
      "grad_norm": 2.0687379509655113,
      "language_loss": 0.90734458,
      "learning_rate": 1.5954094554173097e-06,
      "loss": 0.92950922,
      "num_input_tokens_seen": 103713275,
      "step": 4806,
      "time_per_iteration": 2.604494333267212
    },
    {
      "auxiliary_loss_clip": 0.01155438,
      "auxiliary_loss_mlp": 0.01028695,
      "balance_loss_clip": 1.05475092,
      "balance_loss_mlp": 1.02034485,
      "epoch": 0.5780075753021102,
      "flos": 14136846716160.0,
      "grad_norm": 3.1861980250450213,
      "language_loss": 0.79092312,
      "learning_rate": 1.5946466200473482e-06,
      "loss": 0.81276441,
      "num_input_tokens_seen": 103731185,
      "step": 4807,
      "time_per_iteration": 2.6408236026763916
    },
    {
      "auxiliary_loss_clip": 0.01162868,
      "auxiliary_loss_mlp": 0.01034174,
      "balance_loss_clip": 1.05629325,
      "balance_loss_mlp": 1.02531099,
      "epoch": 0.5781278181927494,
      "flos": 15262178883840.0,
      "grad_norm": 1.8015051314603205,
      "language_loss": 0.83446103,
      "learning_rate": 1.5938838461691723e-06,
      "loss": 0.85643148,
      "num_input_tokens_seen": 103748095,
      "step": 4808,
      "time_per_iteration": 2.73909068107605
    },
    {
      "auxiliary_loss_clip": 0.01111658,
      "auxiliary_loss_mlp": 0.01035314,
      "balance_loss_clip": 1.04627299,
      "balance_loss_mlp": 1.02681446,
      "epoch": 0.5782480610833884,
      "flos": 16726831856640.0,
      "grad_norm": 2.547013276012754,
      "language_loss": 0.83288121,
      "learning_rate": 1.593121133898494e-06,
      "loss": 0.85435092,
      "num_input_tokens_seen": 103765300,
      "step": 4809,
      "time_per_iteration": 3.654456615447998
    },
    {
      "auxiliary_loss_clip": 0.01190246,
      "auxiliary_loss_mlp": 0.0102536,
      "balance_loss_clip": 1.05798578,
      "balance_loss_mlp": 1.01687193,
      "epoch": 0.5783683039740275,
      "flos": 25482140144640.0,
      "grad_norm": 2.255853728357511,
      "language_loss": 0.7931056,
      "learning_rate": 1.592358483351016e-06,
      "loss": 0.8152616,
      "num_input_tokens_seen": 103785475,
      "step": 4810,
      "time_per_iteration": 2.7075881958007812
    },
    {
      "auxiliary_loss_clip": 0.01170389,
      "auxiliary_loss_mlp": 0.01033125,
      "balance_loss_clip": 1.05748153,
      "balance_loss_mlp": 1.02446413,
      "epoch": 0.5784885468646667,
      "flos": 18405835240320.0,
      "grad_norm": 1.8887753279603599,
      "language_loss": 0.72751462,
      "learning_rate": 1.5915958946424326e-06,
      "loss": 0.74954981,
      "num_input_tokens_seen": 103804160,
      "step": 4811,
      "time_per_iteration": 2.5456628799438477
    },
    {
      "auxiliary_loss_clip": 0.01171061,
      "auxiliary_loss_mlp": 0.01033781,
      "balance_loss_clip": 1.05609679,
      "balance_loss_mlp": 1.0250963,
      "epoch": 0.5786087897553057,
      "flos": 46100717936640.0,
      "grad_norm": 1.6057466556120552,
      "language_loss": 0.74523878,
      "learning_rate": 1.5908333678884271e-06,
      "loss": 0.76728714,
      "num_input_tokens_seen": 103830580,
      "step": 4812,
      "time_per_iteration": 2.8597006797790527
    },
    {
      "auxiliary_loss_clip": 0.01188164,
      "auxiliary_loss_mlp": 0.01029014,
      "balance_loss_clip": 1.05748296,
      "balance_loss_mlp": 1.02035344,
      "epoch": 0.5787290326459448,
      "flos": 12385950261120.0,
      "grad_norm": 2.919175454372506,
      "language_loss": 0.73985618,
      "learning_rate": 1.5900709032046743e-06,
      "loss": 0.76202798,
      "num_input_tokens_seen": 103848655,
      "step": 4813,
      "time_per_iteration": 2.639059543609619
    },
    {
      "auxiliary_loss_clip": 0.01161905,
      "auxiliary_loss_mlp": 0.01030773,
      "balance_loss_clip": 1.05666566,
      "balance_loss_mlp": 1.02256584,
      "epoch": 0.5788492755365839,
      "flos": 23290332243840.0,
      "grad_norm": 2.4593718311217154,
      "language_loss": 0.78294814,
      "learning_rate": 1.5893085007068391e-06,
      "loss": 0.8048749,
      "num_input_tokens_seen": 103866215,
      "step": 4814,
      "time_per_iteration": 2.7344534397125244
    },
    {
      "auxiliary_loss_clip": 0.01187245,
      "auxiliary_loss_mlp": 0.01029236,
      "balance_loss_clip": 1.05829096,
      "balance_loss_mlp": 1.02119553,
      "epoch": 0.578969518427223,
      "flos": 24061047390720.0,
      "grad_norm": 2.1788308339878784,
      "language_loss": 0.71084607,
      "learning_rate": 1.5885461605105786e-06,
      "loss": 0.73301089,
      "num_input_tokens_seen": 103887815,
      "step": 4815,
      "time_per_iteration": 2.6875994205474854
    },
    {
      "auxiliary_loss_clip": 0.01190173,
      "auxiliary_loss_mlp": 0.01028977,
      "balance_loss_clip": 1.05804586,
      "balance_loss_mlp": 1.02005982,
      "epoch": 0.579089761317862,
      "flos": 21871825269120.0,
      "grad_norm": 1.8998658811283637,
      "language_loss": 0.76928568,
      "learning_rate": 1.5877838827315375e-06,
      "loss": 0.7914772,
      "num_input_tokens_seen": 103906360,
      "step": 4816,
      "time_per_iteration": 2.606285810470581
    },
    {
      "auxiliary_loss_clip": 0.01185974,
      "auxiliary_loss_mlp": 0.01028435,
      "balance_loss_clip": 1.05727315,
      "balance_loss_mlp": 1.0203557,
      "epoch": 0.5792100042085012,
      "flos": 22929681738240.0,
      "grad_norm": 2.1042616040921276,
      "language_loss": 0.70877808,
      "learning_rate": 1.587021667485355e-06,
      "loss": 0.73092222,
      "num_input_tokens_seen": 103925730,
      "step": 4817,
      "time_per_iteration": 2.637103319168091
    },
    {
      "auxiliary_loss_clip": 0.01186567,
      "auxiliary_loss_mlp": 0.01028024,
      "balance_loss_clip": 1.05653405,
      "balance_loss_mlp": 1.02068067,
      "epoch": 0.5793302470991403,
      "flos": 21470056669440.0,
      "grad_norm": 1.8071186042028222,
      "language_loss": 0.7860918,
      "learning_rate": 1.5862595148876559e-06,
      "loss": 0.80823767,
      "num_input_tokens_seen": 103945835,
      "step": 4818,
      "time_per_iteration": 2.554431200027466
    },
    {
      "auxiliary_loss_clip": 0.01159655,
      "auxiliary_loss_mlp": 0.01036412,
      "balance_loss_clip": 1.05671656,
      "balance_loss_mlp": 1.02735186,
      "epoch": 0.5794504899897793,
      "flos": 12711013367040.0,
      "grad_norm": 2.4691413859632307,
      "language_loss": 0.77048135,
      "learning_rate": 1.58549742505406e-06,
      "loss": 0.79244196,
      "num_input_tokens_seen": 103960580,
      "step": 4819,
      "time_per_iteration": 2.648088216781616
    },
    {
      "auxiliary_loss_clip": 0.01172839,
      "auxiliary_loss_mlp": 0.01032217,
      "balance_loss_clip": 1.0562799,
      "balance_loss_mlp": 1.02339554,
      "epoch": 0.5795707328804185,
      "flos": 14867054300160.0,
      "grad_norm": 2.605310448023083,
      "language_loss": 0.75579536,
      "learning_rate": 1.5847353981001747e-06,
      "loss": 0.77784586,
      "num_input_tokens_seen": 103977760,
      "step": 4820,
      "time_per_iteration": 2.5774986743927
    },
    {
      "auxiliary_loss_clip": 0.01176207,
      "auxiliary_loss_mlp": 0.01030079,
      "balance_loss_clip": 1.05567288,
      "balance_loss_mlp": 1.02156997,
      "epoch": 0.5796909757710575,
      "flos": 36430046432640.0,
      "grad_norm": 1.7142180295286384,
      "language_loss": 0.70133185,
      "learning_rate": 1.5839734341415993e-06,
      "loss": 0.72339475,
      "num_input_tokens_seen": 103999960,
      "step": 4821,
      "time_per_iteration": 2.751129388809204
    },
    {
      "auxiliary_loss_clip": 0.01185736,
      "auxiliary_loss_mlp": 0.01031627,
      "balance_loss_clip": 1.05983722,
      "balance_loss_mlp": 1.02327609,
      "epoch": 0.5798112186616966,
      "flos": 23039891642880.0,
      "grad_norm": 12.42088517039944,
      "language_loss": 0.76427007,
      "learning_rate": 1.5832115332939238e-06,
      "loss": 0.78644371,
      "num_input_tokens_seen": 104018400,
      "step": 4822,
      "time_per_iteration": 2.679206371307373
    },
    {
      "auxiliary_loss_clip": 0.01113524,
      "auxiliary_loss_mlp": 0.01029274,
      "balance_loss_clip": 1.04759359,
      "balance_loss_mlp": 1.02123356,
      "epoch": 0.5799314615523358,
      "flos": 16652604401280.0,
      "grad_norm": 3.6176422747007773,
      "language_loss": 0.74529028,
      "learning_rate": 1.5824496956727272e-06,
      "loss": 0.76671821,
      "num_input_tokens_seen": 104035605,
      "step": 4823,
      "time_per_iteration": 2.8282504081726074
    },
    {
      "auxiliary_loss_clip": 0.01154997,
      "auxiliary_loss_mlp": 0.01026422,
      "balance_loss_clip": 1.05462313,
      "balance_loss_mlp": 1.01869774,
      "epoch": 0.5800517044429748,
      "flos": 20485673470080.0,
      "grad_norm": 3.615206042222389,
      "language_loss": 0.73549259,
      "learning_rate": 1.5816879213935797e-06,
      "loss": 0.75730681,
      "num_input_tokens_seen": 104054415,
      "step": 4824,
      "time_per_iteration": 3.691692352294922
    },
    {
      "auxiliary_loss_clip": 0.01171201,
      "auxiliary_loss_mlp": 0.01027238,
      "balance_loss_clip": 1.05706847,
      "balance_loss_mlp": 1.01937008,
      "epoch": 0.5801719473336139,
      "flos": 31538258968320.0,
      "grad_norm": 1.5561312996801586,
      "language_loss": 0.79968798,
      "learning_rate": 1.5809262105720416e-06,
      "loss": 0.82167238,
      "num_input_tokens_seen": 104075455,
      "step": 4825,
      "time_per_iteration": 3.676110029220581
    },
    {
      "auxiliary_loss_clip": 0.01172188,
      "auxiliary_loss_mlp": 0.01033179,
      "balance_loss_clip": 1.05400157,
      "balance_loss_mlp": 1.02458358,
      "epoch": 0.580292190224253,
      "flos": 20375966355840.0,
      "grad_norm": 2.2400180100369234,
      "language_loss": 0.79886985,
      "learning_rate": 1.5801645633236644e-06,
      "loss": 0.82092351,
      "num_input_tokens_seen": 104096440,
      "step": 4826,
      "time_per_iteration": 3.610269784927368
    },
    {
      "auxiliary_loss_clip": 0.01155235,
      "auxiliary_loss_mlp": 0.0104006,
      "balance_loss_clip": 1.05563486,
      "balance_loss_mlp": 1.03162849,
      "epoch": 0.5804124331148921,
      "flos": 26615373304320.0,
      "grad_norm": 2.107164941339185,
      "language_loss": 0.77704608,
      "learning_rate": 1.579402979763989e-06,
      "loss": 0.79899901,
      "num_input_tokens_seen": 104116775,
      "step": 4827,
      "time_per_iteration": 2.712637424468994
    },
    {
      "auxiliary_loss_clip": 0.01153335,
      "auxiliary_loss_mlp": 0.01033301,
      "balance_loss_clip": 1.05032825,
      "balance_loss_mlp": 1.02527821,
      "epoch": 0.5805326760055312,
      "flos": 13478496289920.0,
      "grad_norm": 3.0847271258031173,
      "language_loss": 0.81502974,
      "learning_rate": 1.578641460008548e-06,
      "loss": 0.83689606,
      "num_input_tokens_seen": 104134510,
      "step": 4828,
      "time_per_iteration": 2.6694412231445312
    },
    {
      "auxiliary_loss_clip": 0.01185109,
      "auxiliary_loss_mlp": 0.01026035,
      "balance_loss_clip": 1.05722165,
      "balance_loss_mlp": 1.01871872,
      "epoch": 0.5806529188961702,
      "flos": 12091374823680.0,
      "grad_norm": 2.352804000264984,
      "language_loss": 0.68023622,
      "learning_rate": 1.5778800041728613e-06,
      "loss": 0.70234764,
      "num_input_tokens_seen": 104150800,
      "step": 4829,
      "time_per_iteration": 2.6971993446350098
    },
    {
      "auxiliary_loss_clip": 0.01157454,
      "auxiliary_loss_mlp": 0.0103305,
      "balance_loss_clip": 1.05414081,
      "balance_loss_mlp": 1.02510428,
      "epoch": 0.5807731617868094,
      "flos": 26214107495040.0,
      "grad_norm": 1.5337607034599436,
      "language_loss": 0.66362864,
      "learning_rate": 1.577118612372443e-06,
      "loss": 0.6855337,
      "num_input_tokens_seen": 104172640,
      "step": 4830,
      "time_per_iteration": 2.6893327236175537
    },
    {
      "auxiliary_loss_clip": 0.011868,
      "auxiliary_loss_mlp": 0.0103593,
      "balance_loss_clip": 1.05566752,
      "balance_loss_mlp": 1.02698302,
      "epoch": 0.5808934046774484,
      "flos": 37962139190400.0,
      "grad_norm": 1.8133064486439097,
      "language_loss": 0.7068119,
      "learning_rate": 1.5763572847227943e-06,
      "loss": 0.72903919,
      "num_input_tokens_seen": 104193525,
      "step": 4831,
      "time_per_iteration": 2.7868402004241943
    },
    {
      "auxiliary_loss_clip": 0.01167213,
      "auxiliary_loss_mlp": 0.01028328,
      "balance_loss_clip": 1.05014777,
      "balance_loss_mlp": 1.02073693,
      "epoch": 0.5810136475680875,
      "flos": 20485853038080.0,
      "grad_norm": 1.8321601352413117,
      "language_loss": 0.81412852,
      "learning_rate": 1.5755960213394091e-06,
      "loss": 0.83608395,
      "num_input_tokens_seen": 104210625,
      "step": 4832,
      "time_per_iteration": 2.6045467853546143
    },
    {
      "auxiliary_loss_clip": 0.01169767,
      "auxiliary_loss_mlp": 0.01028812,
      "balance_loss_clip": 1.05725288,
      "balance_loss_mlp": 1.02080154,
      "epoch": 0.5811338904587267,
      "flos": 17530153574400.0,
      "grad_norm": 2.5774855116193174,
      "language_loss": 0.78715813,
      "learning_rate": 1.5748348223377703e-06,
      "loss": 0.8091439,
      "num_input_tokens_seen": 104228180,
      "step": 4833,
      "time_per_iteration": 2.6518168449401855
    },
    {
      "auxiliary_loss_clip": 0.01160369,
      "auxiliary_loss_mlp": 0.01025149,
      "balance_loss_clip": 1.05055404,
      "balance_loss_mlp": 1.01648271,
      "epoch": 0.5812541333493657,
      "flos": 19458017360640.0,
      "grad_norm": 4.540021909080885,
      "language_loss": 0.78110313,
      "learning_rate": 1.5740736878333507e-06,
      "loss": 0.80295831,
      "num_input_tokens_seen": 104246020,
      "step": 4834,
      "time_per_iteration": 2.6396868228912354
    },
    {
      "auxiliary_loss_clip": 0.01188743,
      "auxiliary_loss_mlp": 0.01028606,
      "balance_loss_clip": 1.05696154,
      "balance_loss_mlp": 1.02029729,
      "epoch": 0.5813743762400048,
      "flos": 20594949621120.0,
      "grad_norm": 2.535472974353137,
      "language_loss": 0.78026545,
      "learning_rate": 1.5733126179416143e-06,
      "loss": 0.80243891,
      "num_input_tokens_seen": 104260505,
      "step": 4835,
      "time_per_iteration": 3.5200865268707275
    },
    {
      "auxiliary_loss_clip": 0.01186008,
      "auxiliary_loss_mlp": 0.01028589,
      "balance_loss_clip": 1.05646062,
      "balance_loss_mlp": 1.02024424,
      "epoch": 0.5814946191306439,
      "flos": 33178227246720.0,
      "grad_norm": 2.102112619556292,
      "language_loss": 0.72544485,
      "learning_rate": 1.5725516127780137e-06,
      "loss": 0.74759078,
      "num_input_tokens_seen": 104282640,
      "step": 4836,
      "time_per_iteration": 2.7694671154022217
    },
    {
      "auxiliary_loss_clip": 0.01189562,
      "auxiliary_loss_mlp": 0.01031033,
      "balance_loss_clip": 1.05676341,
      "balance_loss_mlp": 1.02166891,
      "epoch": 0.581614862021283,
      "flos": 16143283503360.0,
      "grad_norm": 4.348762169203179,
      "language_loss": 0.88470811,
      "learning_rate": 1.5717906724579943e-06,
      "loss": 0.90691406,
      "num_input_tokens_seen": 104299700,
      "step": 4837,
      "time_per_iteration": 2.595287799835205
    },
    {
      "auxiliary_loss_clip": 0.01175974,
      "auxiliary_loss_mlp": 0.01033496,
      "balance_loss_clip": 1.05875564,
      "balance_loss_mlp": 1.02509809,
      "epoch": 0.581735104911922,
      "flos": 33802642298880.0,
      "grad_norm": 2.1275782097684823,
      "language_loss": 0.68456912,
      "learning_rate": 1.571029797096989e-06,
      "loss": 0.70666385,
      "num_input_tokens_seen": 104320805,
      "step": 4838,
      "time_per_iteration": 2.7294161319732666
    },
    {
      "auxiliary_loss_clip": 0.01183806,
      "auxiliary_loss_mlp": 0.00762578,
      "balance_loss_clip": 1.05551183,
      "balance_loss_mlp": 1.00134277,
      "epoch": 0.5818553478025612,
      "flos": 23331163029120.0,
      "grad_norm": 2.8677190142822995,
      "language_loss": 0.7927537,
      "learning_rate": 1.570268986810423e-06,
      "loss": 0.81221747,
      "num_input_tokens_seen": 104340700,
      "step": 4839,
      "time_per_iteration": 2.595191717147827
    },
    {
      "auxiliary_loss_clip": 0.01173612,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.05511355,
      "balance_loss_mlp": 1.01965272,
      "epoch": 0.5819755906932003,
      "flos": 20996143603200.0,
      "grad_norm": 2.0560517489438808,
      "language_loss": 0.75427765,
      "learning_rate": 1.5695082417137096e-06,
      "loss": 0.77629143,
      "num_input_tokens_seen": 104358575,
      "step": 4840,
      "time_per_iteration": 2.636615037918091
    },
    {
      "auxiliary_loss_clip": 0.01164421,
      "auxiliary_loss_mlp": 0.01027226,
      "balance_loss_clip": 1.05186033,
      "balance_loss_mlp": 1.01948369,
      "epoch": 0.5820958335838393,
      "flos": 21431668008960.0,
      "grad_norm": 4.889167495288096,
      "language_loss": 0.75621045,
      "learning_rate": 1.5687475619222539e-06,
      "loss": 0.77812684,
      "num_input_tokens_seen": 104378530,
      "step": 4841,
      "time_per_iteration": 2.6465485095977783
    },
    {
      "auxiliary_loss_clip": 0.01188826,
      "auxiliary_loss_mlp": 0.01035484,
      "balance_loss_clip": 1.05762434,
      "balance_loss_mlp": 1.02667761,
      "epoch": 0.5822160764744785,
      "flos": 17967473660160.0,
      "grad_norm": 3.412907266247043,
      "language_loss": 0.73454636,
      "learning_rate": 1.5679869475514496e-06,
      "loss": 0.75678945,
      "num_input_tokens_seen": 104395465,
      "step": 4842,
      "time_per_iteration": 2.6405301094055176
    },
    {
      "auxiliary_loss_clip": 0.01128322,
      "auxiliary_loss_mlp": 0.01029947,
      "balance_loss_clip": 1.04834914,
      "balance_loss_mlp": 1.02082705,
      "epoch": 0.5823363193651175,
      "flos": 23033858158080.0,
      "grad_norm": 2.524917707379673,
      "language_loss": 0.81096113,
      "learning_rate": 1.567226398716682e-06,
      "loss": 0.83254385,
      "num_input_tokens_seen": 104415380,
      "step": 4843,
      "time_per_iteration": 2.7195334434509277
    },
    {
      "auxiliary_loss_clip": 0.01188718,
      "auxiliary_loss_mlp": 0.01033604,
      "balance_loss_clip": 1.05712783,
      "balance_loss_mlp": 1.02404356,
      "epoch": 0.5824565622557566,
      "flos": 32891840110080.0,
      "grad_norm": 2.1298290675580205,
      "language_loss": 0.6213901,
      "learning_rate": 1.566465915533326e-06,
      "loss": 0.64361334,
      "num_input_tokens_seen": 104437410,
      "step": 4844,
      "time_per_iteration": 2.761836290359497
    },
    {
      "auxiliary_loss_clip": 0.01170951,
      "auxiliary_loss_mlp": 0.01032171,
      "balance_loss_clip": 1.05432796,
      "balance_loss_mlp": 1.02374899,
      "epoch": 0.5825768051463958,
      "flos": 22229674513920.0,
      "grad_norm": 1.970959989200819,
      "language_loss": 0.880826,
      "learning_rate": 1.5657054981167458e-06,
      "loss": 0.90285724,
      "num_input_tokens_seen": 104456305,
      "step": 4845,
      "time_per_iteration": 2.6906440258026123
    },
    {
      "auxiliary_loss_clip": 0.0118675,
      "auxiliary_loss_mlp": 0.01024362,
      "balance_loss_clip": 1.05983567,
      "balance_loss_mlp": 1.01660764,
      "epoch": 0.5826970480370348,
      "flos": 28001561016960.0,
      "grad_norm": 2.020032561932444,
      "language_loss": 0.68254602,
      "learning_rate": 1.5649451465822965e-06,
      "loss": 0.70465708,
      "num_input_tokens_seen": 104477695,
      "step": 4846,
      "time_per_iteration": 2.677086114883423
    },
    {
      "auxiliary_loss_clip": 0.01187871,
      "auxiliary_loss_mlp": 0.01032117,
      "balance_loss_clip": 1.05950427,
      "balance_loss_mlp": 1.02402556,
      "epoch": 0.5828172909276739,
      "flos": 17858053854720.0,
      "grad_norm": 1.9540889691384031,
      "language_loss": 0.83889568,
      "learning_rate": 1.5641848610453218e-06,
      "loss": 0.86109555,
      "num_input_tokens_seen": 104496355,
      "step": 4847,
      "time_per_iteration": 2.543454170227051
    },
    {
      "auxiliary_loss_clip": 0.0115165,
      "auxiliary_loss_mlp": 0.01026391,
      "balance_loss_clip": 1.05272377,
      "balance_loss_mlp": 1.01781368,
      "epoch": 0.582937533818313,
      "flos": 19865244827520.0,
      "grad_norm": 2.0338114539977097,
      "language_loss": 0.85860342,
      "learning_rate": 1.563424641621158e-06,
      "loss": 0.88038385,
      "num_input_tokens_seen": 104515535,
      "step": 4848,
      "time_per_iteration": 2.7189297676086426
    },
    {
      "auxiliary_loss_clip": 0.01188087,
      "auxiliary_loss_mlp": 0.01031301,
      "balance_loss_clip": 1.05754769,
      "balance_loss_mlp": 1.02290583,
      "epoch": 0.5830577767089521,
      "flos": 26870734068480.0,
      "grad_norm": 2.0558992024949316,
      "language_loss": 0.69902438,
      "learning_rate": 1.5626644884251282e-06,
      "loss": 0.72121829,
      "num_input_tokens_seen": 104535055,
      "step": 4849,
      "time_per_iteration": 2.6393890380859375
    },
    {
      "auxiliary_loss_clip": 0.01148991,
      "auxiliary_loss_mlp": 0.01030381,
      "balance_loss_clip": 1.04929543,
      "balance_loss_mlp": 1.02223277,
      "epoch": 0.5831780195995911,
      "flos": 25298205575040.0,
      "grad_norm": 1.6628683088450262,
      "language_loss": 0.88027382,
      "learning_rate": 1.5619044015725488e-06,
      "loss": 0.90206754,
      "num_input_tokens_seen": 104554745,
      "step": 4850,
      "time_per_iteration": 3.644956350326538
    },
    {
      "auxiliary_loss_clip": 0.01174142,
      "auxiliary_loss_mlp": 0.01030301,
      "balance_loss_clip": 1.05924797,
      "balance_loss_mlp": 1.0206157,
      "epoch": 0.5832982624902303,
      "flos": 14756988049920.0,
      "grad_norm": 2.728742839111194,
      "language_loss": 0.86837316,
      "learning_rate": 1.5611443811787224e-06,
      "loss": 0.89041758,
      "num_input_tokens_seen": 104568870,
      "step": 4851,
      "time_per_iteration": 3.4985337257385254
    },
    {
      "auxiliary_loss_clip": 0.01169413,
      "auxiliary_loss_mlp": 0.01031843,
      "balance_loss_clip": 1.05487561,
      "balance_loss_mlp": 1.02343917,
      "epoch": 0.5834185053808694,
      "flos": 20444555376000.0,
      "grad_norm": 2.3719268812317957,
      "language_loss": 0.69301867,
      "learning_rate": 1.560384427358945e-06,
      "loss": 0.71503127,
      "num_input_tokens_seen": 104588415,
      "step": 4852,
      "time_per_iteration": 3.5695064067840576
    },
    {
      "auxiliary_loss_clip": 0.01186231,
      "auxiliary_loss_mlp": 0.01026699,
      "balance_loss_clip": 1.05535865,
      "balance_loss_mlp": 1.01778221,
      "epoch": 0.5835387482715084,
      "flos": 27200394115200.0,
      "grad_norm": 1.5632298607218758,
      "language_loss": 0.73213446,
      "learning_rate": 1.5596245402284998e-06,
      "loss": 0.75426376,
      "num_input_tokens_seen": 104611940,
      "step": 4853,
      "time_per_iteration": 2.6203250885009766
    },
    {
      "auxiliary_loss_clip": 0.01140461,
      "auxiliary_loss_mlp": 0.01036904,
      "balance_loss_clip": 1.05012512,
      "balance_loss_mlp": 1.02877998,
      "epoch": 0.5836589911621476,
      "flos": 16654615562880.0,
      "grad_norm": 1.6622609661513017,
      "language_loss": 0.82209742,
      "learning_rate": 1.5588647199026619e-06,
      "loss": 0.84387106,
      "num_input_tokens_seen": 104629675,
      "step": 4854,
      "time_per_iteration": 2.6931326389312744
    },
    {
      "auxiliary_loss_clip": 0.01138001,
      "auxiliary_loss_mlp": 0.01032562,
      "balance_loss_clip": 1.05088615,
      "balance_loss_mlp": 1.02433038,
      "epoch": 0.5837792340527866,
      "flos": 20446817932800.0,
      "grad_norm": 4.422462744181491,
      "language_loss": 0.87627554,
      "learning_rate": 1.5581049664966956e-06,
      "loss": 0.89798117,
      "num_input_tokens_seen": 104647435,
      "step": 4855,
      "time_per_iteration": 2.6961617469787598
    },
    {
      "auxiliary_loss_clip": 0.01073453,
      "auxiliary_loss_mlp": 0.0099909,
      "balance_loss_clip": 1.01330066,
      "balance_loss_mlp": 0.9980709,
      "epoch": 0.5838994769434257,
      "flos": 65995480765440.0,
      "grad_norm": 0.9848074474318036,
      "language_loss": 0.65043831,
      "learning_rate": 1.5573452801258545e-06,
      "loss": 0.67116374,
      "num_input_tokens_seen": 104694605,
      "step": 4856,
      "time_per_iteration": 3.023379325866699
    },
    {
      "auxiliary_loss_clip": 0.01156988,
      "auxiliary_loss_mlp": 0.01029609,
      "balance_loss_clip": 1.05592752,
      "balance_loss_mlp": 1.02131176,
      "epoch": 0.5840197198340649,
      "flos": 21470523546240.0,
      "grad_norm": 2.081929659463262,
      "language_loss": 0.63429224,
      "learning_rate": 1.5565856609053824e-06,
      "loss": 0.65615821,
      "num_input_tokens_seen": 104713400,
      "step": 4857,
      "time_per_iteration": 2.7194290161132812
    },
    {
      "auxiliary_loss_clip": 0.01188023,
      "auxiliary_loss_mlp": 0.01039178,
      "balance_loss_clip": 1.05897212,
      "balance_loss_mlp": 1.02963495,
      "epoch": 0.5841399627247039,
      "flos": 19135144984320.0,
      "grad_norm": 2.359775057469064,
      "language_loss": 0.79941297,
      "learning_rate": 1.5558261089505127e-06,
      "loss": 0.82168496,
      "num_input_tokens_seen": 104732130,
      "step": 4858,
      "time_per_iteration": 2.703918695449829
    },
    {
      "auxiliary_loss_clip": 0.01174898,
      "auxiliary_loss_mlp": 0.01024985,
      "balance_loss_clip": 1.05653882,
      "balance_loss_mlp": 1.0163424,
      "epoch": 0.584260205615343,
      "flos": 26425692558720.0,
      "grad_norm": 2.728929723760384,
      "language_loss": 0.80102944,
      "learning_rate": 1.5550666243764697e-06,
      "loss": 0.82302833,
      "num_input_tokens_seen": 104750290,
      "step": 4859,
      "time_per_iteration": 2.63867449760437
    },
    {
      "auxiliary_loss_clip": 0.01185344,
      "auxiliary_loss_mlp": 0.0103354,
      "balance_loss_clip": 1.0560627,
      "balance_loss_mlp": 1.02517772,
      "epoch": 0.584380448505982,
      "flos": 13881809174400.0,
      "grad_norm": 3.1903925795514323,
      "language_loss": 0.77348822,
      "learning_rate": 1.554307207298465e-06,
      "loss": 0.79567701,
      "num_input_tokens_seen": 104768550,
      "step": 4860,
      "time_per_iteration": 2.6086325645446777
    },
    {
      "auxiliary_loss_clip": 0.01139599,
      "auxiliary_loss_mlp": 0.01036053,
      "balance_loss_clip": 1.05028939,
      "balance_loss_mlp": 1.02716589,
      "epoch": 0.5845006913966212,
      "flos": 21543709507200.0,
      "grad_norm": 2.32796090300952,
      "language_loss": 0.78816628,
      "learning_rate": 1.553547857831704e-06,
      "loss": 0.80992281,
      "num_input_tokens_seen": 104785060,
      "step": 4861,
      "time_per_iteration": 3.5141756534576416
    },
    {
      "auxiliary_loss_clip": 0.01073707,
      "auxiliary_loss_mlp": 0.01000826,
      "balance_loss_clip": 1.01374209,
      "balance_loss_mlp": 0.99980116,
      "epoch": 0.5846209342872603,
      "flos": 58375452712320.0,
      "grad_norm": 0.8766944144916349,
      "language_loss": 0.64183384,
      "learning_rate": 1.5527885760913771e-06,
      "loss": 0.66257918,
      "num_input_tokens_seen": 104834950,
      "step": 4862,
      "time_per_iteration": 3.036273717880249
    },
    {
      "auxiliary_loss_clip": 0.01175272,
      "auxiliary_loss_mlp": 0.01033844,
      "balance_loss_clip": 1.05876637,
      "balance_loss_mlp": 1.02569032,
      "epoch": 0.5847411771778993,
      "flos": 18588045957120.0,
      "grad_norm": 1.72929991325767,
      "language_loss": 0.76469404,
      "learning_rate": 1.552029362192668e-06,
      "loss": 0.78678519,
      "num_input_tokens_seen": 104854210,
      "step": 4863,
      "time_per_iteration": 2.632976531982422
    },
    {
      "auxiliary_loss_clip": 0.01187077,
      "auxiliary_loss_mlp": 0.0103394,
      "balance_loss_clip": 1.05775738,
      "balance_loss_mlp": 1.02554142,
      "epoch": 0.5848614200685385,
      "flos": 24240780069120.0,
      "grad_norm": 1.9169994389359977,
      "language_loss": 0.72458822,
      "learning_rate": 1.5512702162507478e-06,
      "loss": 0.7467984,
      "num_input_tokens_seen": 104874525,
      "step": 4864,
      "time_per_iteration": 2.5881214141845703
    },
    {
      "auxiliary_loss_clip": 0.01058839,
      "auxiliary_loss_mlp": 0.00999886,
      "balance_loss_clip": 1.01222825,
      "balance_loss_mlp": 0.99884284,
      "epoch": 0.5849816629591775,
      "flos": 71660245933440.0,
      "grad_norm": 1.1169535881608952,
      "language_loss": 0.55727816,
      "learning_rate": 1.5505111383807792e-06,
      "loss": 0.57786536,
      "num_input_tokens_seen": 104937195,
      "step": 4865,
      "time_per_iteration": 3.233776807785034
    },
    {
      "auxiliary_loss_clip": 0.01183316,
      "auxiliary_loss_mlp": 0.01032021,
      "balance_loss_clip": 1.05358613,
      "balance_loss_mlp": 1.02365243,
      "epoch": 0.5851019058498166,
      "flos": 23802095266560.0,
      "grad_norm": 1.9937558137350737,
      "language_loss": 0.80617088,
      "learning_rate": 1.5497521286979138e-06,
      "loss": 0.82832426,
      "num_input_tokens_seen": 104957435,
      "step": 4866,
      "time_per_iteration": 2.6285324096679688
    },
    {
      "auxiliary_loss_clip": 0.01188532,
      "auxiliary_loss_mlp": 0.00763578,
      "balance_loss_clip": 1.05551577,
      "balance_loss_mlp": 1.00108945,
      "epoch": 0.5852221487404557,
      "flos": 24388516707840.0,
      "grad_norm": 2.038203896817191,
      "language_loss": 0.73992264,
      "learning_rate": 1.5489931873172927e-06,
      "loss": 0.75944376,
      "num_input_tokens_seen": 104978755,
      "step": 4867,
      "time_per_iteration": 2.664177894592285
    },
    {
      "auxiliary_loss_clip": 0.01152209,
      "auxiliary_loss_mlp": 0.01037806,
      "balance_loss_clip": 1.05225801,
      "balance_loss_mlp": 1.0289067,
      "epoch": 0.5853423916310948,
      "flos": 27271425260160.0,
      "grad_norm": 1.7151890756320698,
      "language_loss": 0.79244173,
      "learning_rate": 1.5482343143540467e-06,
      "loss": 0.8143419,
      "num_input_tokens_seen": 105000020,
      "step": 4868,
      "time_per_iteration": 2.7059988975524902
    },
    {
      "auxiliary_loss_clip": 0.0118499,
      "auxiliary_loss_mlp": 0.01030421,
      "balance_loss_clip": 1.05589294,
      "balance_loss_mlp": 1.02175999,
      "epoch": 0.5854626345217339,
      "flos": 11983786611840.0,
      "grad_norm": 1.8988620971901269,
      "language_loss": 0.82582057,
      "learning_rate": 1.547475509923295e-06,
      "loss": 0.84797478,
      "num_input_tokens_seen": 105017060,
      "step": 4869,
      "time_per_iteration": 2.5190341472625732
    },
    {
      "auxiliary_loss_clip": 0.01063217,
      "auxiliary_loss_mlp": 0.01006374,
      "balance_loss_clip": 1.01233423,
      "balance_loss_mlp": 1.00530136,
      "epoch": 0.585582877412373,
      "flos": 64342335173760.0,
      "grad_norm": 0.7210705255249636,
      "language_loss": 0.55960858,
      "learning_rate": 1.5467167741401495e-06,
      "loss": 0.5803045,
      "num_input_tokens_seen": 105078540,
      "step": 4870,
      "time_per_iteration": 3.2325499057769775
    },
    {
      "auxiliary_loss_clip": 0.01169919,
      "auxiliary_loss_mlp": 0.01036351,
      "balance_loss_clip": 1.05702472,
      "balance_loss_mlp": 1.02828074,
      "epoch": 0.5857031203030121,
      "flos": 17011926103680.0,
      "grad_norm": 2.7514755979293395,
      "language_loss": 0.71469277,
      "learning_rate": 1.5459581071197083e-06,
      "loss": 0.73675549,
      "num_input_tokens_seen": 105094200,
      "step": 4871,
      "time_per_iteration": 2.6181130409240723
    },
    {
      "auxiliary_loss_clip": 0.01192234,
      "auxiliary_loss_mlp": 0.01033701,
      "balance_loss_clip": 1.06004584,
      "balance_loss_mlp": 1.02538598,
      "epoch": 0.5858233631936511,
      "flos": 20885682303360.0,
      "grad_norm": 2.5836382627847825,
      "language_loss": 0.83721137,
      "learning_rate": 1.5451995089770624e-06,
      "loss": 0.85947073,
      "num_input_tokens_seen": 105113985,
      "step": 4872,
      "time_per_iteration": 2.6669692993164062
    },
    {
      "auxiliary_loss_clip": 0.01182157,
      "auxiliary_loss_mlp": 0.01022706,
      "balance_loss_clip": 1.05549145,
      "balance_loss_mlp": 1.01499915,
      "epoch": 0.5859436060842903,
      "flos": 23191902000000.0,
      "grad_norm": 1.758311797085883,
      "language_loss": 0.71999526,
      "learning_rate": 1.5444409798272885e-06,
      "loss": 0.74204385,
      "num_input_tokens_seen": 105138075,
      "step": 4873,
      "time_per_iteration": 2.709961414337158
    },
    {
      "auxiliary_loss_clip": 0.01175692,
      "auxiliary_loss_mlp": 0.01032911,
      "balance_loss_clip": 1.05443478,
      "balance_loss_mlp": 1.02408338,
      "epoch": 0.5860638489749294,
      "flos": 22492648961280.0,
      "grad_norm": 15.079183750098355,
      "language_loss": 0.80865872,
      "learning_rate": 1.543682519785456e-06,
      "loss": 0.83074474,
      "num_input_tokens_seen": 105156555,
      "step": 4874,
      "time_per_iteration": 2.667330026626587
    },
    {
      "auxiliary_loss_clip": 0.01159821,
      "auxiliary_loss_mlp": 0.01034341,
      "balance_loss_clip": 1.0542227,
      "balance_loss_mlp": 1.02672935,
      "epoch": 0.5861840918655684,
      "flos": 17566243764480.0,
      "grad_norm": 2.9537553514154915,
      "language_loss": 0.81306028,
      "learning_rate": 1.5429241289666219e-06,
      "loss": 0.83500189,
      "num_input_tokens_seen": 105174055,
      "step": 4875,
      "time_per_iteration": 2.6917288303375244
    },
    {
      "auxiliary_loss_clip": 0.01175294,
      "auxiliary_loss_mlp": 0.0102983,
      "balance_loss_clip": 1.05630744,
      "balance_loss_mlp": 1.02146792,
      "epoch": 0.5863043347562076,
      "flos": 25556152118400.0,
      "grad_norm": 2.0368710054800756,
      "language_loss": 0.69796216,
      "learning_rate": 1.5421658074858342e-06,
      "loss": 0.72001344,
      "num_input_tokens_seen": 105192160,
      "step": 4876,
      "time_per_iteration": 3.668820858001709
    },
    {
      "auxiliary_loss_clip": 0.01166679,
      "auxiliary_loss_mlp": 0.01033787,
      "balance_loss_clip": 1.05389929,
      "balance_loss_mlp": 1.02451837,
      "epoch": 0.5864245776468466,
      "flos": 20667525050880.0,
      "grad_norm": 3.238619269604009,
      "language_loss": 0.67247301,
      "learning_rate": 1.5414075554581298e-06,
      "loss": 0.69447768,
      "num_input_tokens_seen": 105210205,
      "step": 4877,
      "time_per_iteration": 3.527804136276245
    },
    {
      "auxiliary_loss_clip": 0.01189539,
      "auxiliary_loss_mlp": 0.01030182,
      "balance_loss_clip": 1.05825913,
      "balance_loss_mlp": 1.02143192,
      "epoch": 0.5865448205374857,
      "flos": 28913907490560.0,
      "grad_norm": 2.332195246380138,
      "language_loss": 0.79193211,
      "learning_rate": 1.5406493729985348e-06,
      "loss": 0.81412935,
      "num_input_tokens_seen": 105229400,
      "step": 4878,
      "time_per_iteration": 3.663923740386963
    },
    {
      "auxiliary_loss_clip": 0.01155952,
      "auxiliary_loss_mlp": 0.01034652,
      "balance_loss_clip": 1.05428314,
      "balance_loss_mlp": 1.0256691,
      "epoch": 0.5866650634281249,
      "flos": 25842575168640.0,
      "grad_norm": 3.7654584015751174,
      "language_loss": 0.71958363,
      "learning_rate": 1.5398912602220644e-06,
      "loss": 0.74148965,
      "num_input_tokens_seen": 105248675,
      "step": 4879,
      "time_per_iteration": 2.7156982421875
    },
    {
      "auxiliary_loss_clip": 0.01185767,
      "auxiliary_loss_mlp": 0.01031967,
      "balance_loss_clip": 1.05659294,
      "balance_loss_mlp": 1.02290154,
      "epoch": 0.5867853063187639,
      "flos": 17052325925760.0,
      "grad_norm": 11.68487566305747,
      "language_loss": 0.78632027,
      "learning_rate": 1.539133217243724e-06,
      "loss": 0.80849761,
      "num_input_tokens_seen": 105265695,
      "step": 4880,
      "time_per_iteration": 2.5980968475341797
    },
    {
      "auxiliary_loss_clip": 0.01148503,
      "auxiliary_loss_mlp": 0.0103325,
      "balance_loss_clip": 1.05344534,
      "balance_loss_mlp": 1.02370715,
      "epoch": 0.586905549209403,
      "flos": 24645026707200.0,
      "grad_norm": 3.02404854601179,
      "language_loss": 0.76608551,
      "learning_rate": 1.5383752441785081e-06,
      "loss": 0.78790307,
      "num_input_tokens_seen": 105284920,
      "step": 4881,
      "time_per_iteration": 2.680840015411377
    },
    {
      "auxiliary_loss_clip": 0.01171864,
      "auxiliary_loss_mlp": 0.01040006,
      "balance_loss_clip": 1.05754912,
      "balance_loss_mlp": 1.03114927,
      "epoch": 0.5870257921000421,
      "flos": 14720538723840.0,
      "grad_norm": 2.7560177921913653,
      "language_loss": 0.85567164,
      "learning_rate": 1.5376173411414003e-06,
      "loss": 0.87779039,
      "num_input_tokens_seen": 105302960,
      "step": 4882,
      "time_per_iteration": 2.689366340637207
    },
    {
      "auxiliary_loss_clip": 0.0118812,
      "auxiliary_loss_mlp": 0.01033685,
      "balance_loss_clip": 1.05818772,
      "balance_loss_mlp": 1.02472663,
      "epoch": 0.5871460349906812,
      "flos": 23914998691200.0,
      "grad_norm": 1.9281871384574025,
      "language_loss": 0.78837544,
      "learning_rate": 1.5368595082473753e-06,
      "loss": 0.81059349,
      "num_input_tokens_seen": 105321260,
      "step": 4883,
      "time_per_iteration": 2.6728904247283936
    },
    {
      "auxiliary_loss_clip": 0.01184822,
      "auxiliary_loss_mlp": 0.01035852,
      "balance_loss_clip": 1.05658782,
      "balance_loss_mlp": 1.02709055,
      "epoch": 0.5872662778813202,
      "flos": 22164174063360.0,
      "grad_norm": 1.921913686623922,
      "language_loss": 0.77989674,
      "learning_rate": 1.5361017456113935e-06,
      "loss": 0.80210346,
      "num_input_tokens_seen": 105341610,
      "step": 4884,
      "time_per_iteration": 2.679058313369751
    },
    {
      "auxiliary_loss_clip": 0.01187976,
      "auxiliary_loss_mlp": 0.01034722,
      "balance_loss_clip": 1.05699205,
      "balance_loss_mlp": 1.02545357,
      "epoch": 0.5873865207719594,
      "flos": 18441925430400.0,
      "grad_norm": 2.172072034223494,
      "language_loss": 0.86028302,
      "learning_rate": 1.5353440533484085e-06,
      "loss": 0.88250995,
      "num_input_tokens_seen": 105360465,
      "step": 4885,
      "time_per_iteration": 2.5674009323120117
    },
    {
      "auxiliary_loss_clip": 0.01170654,
      "auxiliary_loss_mlp": 0.01039185,
      "balance_loss_clip": 1.05448151,
      "balance_loss_mlp": 1.03044105,
      "epoch": 0.5875067636625985,
      "flos": 54015321427200.0,
      "grad_norm": 1.9869696457316874,
      "language_loss": 0.65944701,
      "learning_rate": 1.534586431573361e-06,
      "loss": 0.68154538,
      "num_input_tokens_seen": 105385405,
      "step": 4886,
      "time_per_iteration": 2.934495210647583
    },
    {
      "auxiliary_loss_clip": 0.01143522,
      "auxiliary_loss_mlp": 0.01030841,
      "balance_loss_clip": 1.05263019,
      "balance_loss_mlp": 1.0221684,
      "epoch": 0.5876270065532375,
      "flos": 27995707100160.0,
      "grad_norm": 2.40654599031684,
      "language_loss": 0.79156458,
      "learning_rate": 1.5338288804011817e-06,
      "loss": 0.81330818,
      "num_input_tokens_seen": 105404905,
      "step": 4887,
      "time_per_iteration": 2.7406044006347656
    },
    {
      "auxiliary_loss_clip": 0.01138255,
      "auxiliary_loss_mlp": 0.01033032,
      "balance_loss_clip": 1.04899287,
      "balance_loss_mlp": 1.02434158,
      "epoch": 0.5877472494438767,
      "flos": 21361462876800.0,
      "grad_norm": 2.2184355562427363,
      "language_loss": 0.71178114,
      "learning_rate": 1.533071399946791e-06,
      "loss": 0.73349398,
      "num_input_tokens_seen": 105423650,
      "step": 4888,
      "time_per_iteration": 3.6190195083618164
    },
    {
      "auxiliary_loss_clip": 0.01184888,
      "auxiliary_loss_mlp": 0.01031054,
      "balance_loss_clip": 1.05525947,
      "balance_loss_mlp": 1.02305484,
      "epoch": 0.5878674923345157,
      "flos": 22383013674240.0,
      "grad_norm": 1.8439461754103348,
      "language_loss": 0.57969815,
      "learning_rate": 1.5323139903250977e-06,
      "loss": 0.60185754,
      "num_input_tokens_seen": 105444255,
      "step": 4889,
      "time_per_iteration": 2.660200357437134
    },
    {
      "auxiliary_loss_clip": 0.01177484,
      "auxiliary_loss_mlp": 0.0103027,
      "balance_loss_clip": 1.05663764,
      "balance_loss_mlp": 1.02158582,
      "epoch": 0.5879877352251548,
      "flos": 21868664872320.0,
      "grad_norm": 2.65742282142852,
      "language_loss": 0.77136487,
      "learning_rate": 1.5315566516510002e-06,
      "loss": 0.79344237,
      "num_input_tokens_seen": 105462425,
      "step": 4890,
      "time_per_iteration": 2.596900224685669
    },
    {
      "auxiliary_loss_clip": 0.01190171,
      "auxiliary_loss_mlp": 0.01034593,
      "balance_loss_clip": 1.06029272,
      "balance_loss_mlp": 1.02592659,
      "epoch": 0.5881079781157939,
      "flos": 17493811989120.0,
      "grad_norm": 2.7440371770665193,
      "language_loss": 0.67931765,
      "learning_rate": 1.5307993840393857e-06,
      "loss": 0.70156527,
      "num_input_tokens_seen": 105480505,
      "step": 4891,
      "time_per_iteration": 2.6373565196990967
    },
    {
      "auxiliary_loss_clip": 0.01166089,
      "auxiliary_loss_mlp": 0.01030077,
      "balance_loss_clip": 1.05271173,
      "balance_loss_mlp": 1.02193475,
      "epoch": 0.588228221006433,
      "flos": 22601853285120.0,
      "grad_norm": 2.276604527856801,
      "language_loss": 0.8050257,
      "learning_rate": 1.530042187605132e-06,
      "loss": 0.82698727,
      "num_input_tokens_seen": 105499760,
      "step": 4892,
      "time_per_iteration": 2.6198954582214355
    },
    {
      "auxiliary_loss_clip": 0.01174857,
      "auxiliary_loss_mlp": 0.01034984,
      "balance_loss_clip": 1.05885816,
      "balance_loss_mlp": 1.02665758,
      "epoch": 0.5883484638970721,
      "flos": 26176939896960.0,
      "grad_norm": 1.713117379025358,
      "language_loss": 0.84295845,
      "learning_rate": 1.5292850624631044e-06,
      "loss": 0.86505687,
      "num_input_tokens_seen": 105521955,
      "step": 4893,
      "time_per_iteration": 2.692291498184204
    },
    {
      "auxiliary_loss_clip": 0.01175125,
      "auxiliary_loss_mlp": 0.00763001,
      "balance_loss_clip": 1.05530405,
      "balance_loss_mlp": 1.00114036,
      "epoch": 0.5884687067877111,
      "flos": 30443737691520.0,
      "grad_norm": 2.0067276113275945,
      "language_loss": 0.799667,
      "learning_rate": 1.5285280087281593e-06,
      "loss": 0.81904829,
      "num_input_tokens_seen": 105542685,
      "step": 4894,
      "time_per_iteration": 2.68879771232605
    },
    {
      "auxiliary_loss_clip": 0.01072611,
      "auxiliary_loss_mlp": 0.01000757,
      "balance_loss_clip": 1.01315117,
      "balance_loss_mlp": 0.99971408,
      "epoch": 0.5885889496783503,
      "flos": 70507550580480.0,
      "grad_norm": 0.6421611644464562,
      "language_loss": 0.56501925,
      "learning_rate": 1.5277710265151398e-06,
      "loss": 0.5857529,
      "num_input_tokens_seen": 105612165,
      "step": 4895,
      "time_per_iteration": 3.3165698051452637
    },
    {
      "auxiliary_loss_clip": 0.01191243,
      "auxiliary_loss_mlp": 0.01033601,
      "balance_loss_clip": 1.060004,
      "balance_loss_mlp": 1.02507758,
      "epoch": 0.5887091925689893,
      "flos": 19098767485440.0,
      "grad_norm": 4.28569483534611,
      "language_loss": 0.77379346,
      "learning_rate": 1.5270141159388803e-06,
      "loss": 0.79604197,
      "num_input_tokens_seen": 105629185,
      "step": 4896,
      "time_per_iteration": 2.568991184234619
    },
    {
      "auxiliary_loss_clip": 0.01185035,
      "auxiliary_loss_mlp": 0.01035069,
      "balance_loss_clip": 1.05499816,
      "balance_loss_mlp": 1.02670646,
      "epoch": 0.5888294354596284,
      "flos": 23294282739840.0,
      "grad_norm": 1.835097107402161,
      "language_loss": 0.80602235,
      "learning_rate": 1.526257277114203e-06,
      "loss": 0.82822335,
      "num_input_tokens_seen": 105650260,
      "step": 4897,
      "time_per_iteration": 2.633322238922119
    },
    {
      "auxiliary_loss_clip": 0.01156176,
      "auxiliary_loss_mlp": 0.01032347,
      "balance_loss_clip": 1.05292463,
      "balance_loss_mlp": 1.02394283,
      "epoch": 0.5889496783502676,
      "flos": 21981532383360.0,
      "grad_norm": 2.855054640736091,
      "language_loss": 0.80350691,
      "learning_rate": 1.5255005101559201e-06,
      "loss": 0.82539213,
      "num_input_tokens_seen": 105667870,
      "step": 4898,
      "time_per_iteration": 2.7101941108703613
    },
    {
      "auxiliary_loss_clip": 0.01185596,
      "auxiliary_loss_mlp": 0.01027193,
      "balance_loss_clip": 1.05705655,
      "balance_loss_mlp": 1.01916409,
      "epoch": 0.5890699212409066,
      "flos": 21685233093120.0,
      "grad_norm": 3.8423084924221778,
      "language_loss": 0.77200162,
      "learning_rate": 1.524743815178833e-06,
      "loss": 0.79412949,
      "num_input_tokens_seen": 105685830,
      "step": 4899,
      "time_per_iteration": 2.651569128036499
    },
    {
      "auxiliary_loss_clip": 0.01168415,
      "auxiliary_loss_mlp": 0.01028636,
      "balance_loss_clip": 1.05497146,
      "balance_loss_mlp": 1.020697,
      "epoch": 0.5891901641315457,
      "flos": 19464553635840.0,
      "grad_norm": 1.7466307989128178,
      "language_loss": 0.80931079,
      "learning_rate": 1.5239871922977315e-06,
      "loss": 0.8312813,
      "num_input_tokens_seen": 105705745,
      "step": 4900,
      "time_per_iteration": 2.6592857837677
    },
    {
      "auxiliary_loss_clip": 0.01164202,
      "auxiliary_loss_mlp": 0.01034131,
      "balance_loss_clip": 1.05447209,
      "balance_loss_mlp": 1.02603996,
      "epoch": 0.5893104070221848,
      "flos": 19609884063360.0,
      "grad_norm": 1.9743605643450974,
      "language_loss": 0.89986867,
      "learning_rate": 1.523230641627394e-06,
      "loss": 0.92185199,
      "num_input_tokens_seen": 105724730,
      "step": 4901,
      "time_per_iteration": 2.6900532245635986
    },
    {
      "auxiliary_loss_clip": 0.01167401,
      "auxiliary_loss_mlp": 0.01029863,
      "balance_loss_clip": 1.05579591,
      "balance_loss_mlp": 1.02209616,
      "epoch": 0.5894306499128239,
      "flos": 29060063930880.0,
      "grad_norm": 2.132407822444261,
      "language_loss": 0.73024505,
      "learning_rate": 1.5224741632825888e-06,
      "loss": 0.75221771,
      "num_input_tokens_seen": 105744920,
      "step": 4902,
      "time_per_iteration": 3.6938090324401855
    },
    {
      "auxiliary_loss_clip": 0.01172285,
      "auxiliary_loss_mlp": 0.01043521,
      "balance_loss_clip": 1.05395842,
      "balance_loss_mlp": 1.03427672,
      "epoch": 0.589550892803463,
      "flos": 42298890721920.0,
      "grad_norm": 1.8124616968540523,
      "language_loss": 0.6950323,
      "learning_rate": 1.521717757378074e-06,
      "loss": 0.71719038,
      "num_input_tokens_seen": 105765465,
      "step": 4903,
      "time_per_iteration": 3.7603697776794434
    },
    {
      "auxiliary_loss_clip": 0.01173834,
      "auxiliary_loss_mlp": 0.01028457,
      "balance_loss_clip": 1.05569255,
      "balance_loss_mlp": 1.01958823,
      "epoch": 0.5896711356941021,
      "flos": 14137062197760.0,
      "grad_norm": 3.4161515197899974,
      "language_loss": 0.69564128,
      "learning_rate": 1.5209614240285943e-06,
      "loss": 0.71766424,
      "num_input_tokens_seen": 105783120,
      "step": 4904,
      "time_per_iteration": 2.5901055335998535
    },
    {
      "auxiliary_loss_clip": 0.01184336,
      "auxiliary_loss_mlp": 0.01029496,
      "balance_loss_clip": 1.05600739,
      "balance_loss_mlp": 1.02044857,
      "epoch": 0.5897913785847412,
      "flos": 17201355454080.0,
      "grad_norm": 2.132126883550458,
      "language_loss": 0.85333037,
      "learning_rate": 1.520205163348887e-06,
      "loss": 0.87546873,
      "num_input_tokens_seen": 105801055,
      "step": 4905,
      "time_per_iteration": 3.695014238357544
    },
    {
      "auxiliary_loss_clip": 0.01057557,
      "auxiliary_loss_mlp": 0.0100042,
      "balance_loss_clip": 1.01133466,
      "balance_loss_mlp": 0.99946004,
      "epoch": 0.5899116214753802,
      "flos": 48794164202880.0,
      "grad_norm": 0.7223627590741024,
      "language_loss": 0.5687204,
      "learning_rate": 1.519448975453674e-06,
      "loss": 0.58930022,
      "num_input_tokens_seen": 105856155,
      "step": 4906,
      "time_per_iteration": 3.103104591369629
    },
    {
      "auxiliary_loss_clip": 0.01188789,
      "auxiliary_loss_mlp": 0.01035701,
      "balance_loss_clip": 1.05876565,
      "balance_loss_mlp": 1.02676058,
      "epoch": 0.5900318643660194,
      "flos": 21103659987840.0,
      "grad_norm": 2.109883434238171,
      "language_loss": 0.76235574,
      "learning_rate": 1.5186928604576696e-06,
      "loss": 0.78460068,
      "num_input_tokens_seen": 105873350,
      "step": 4907,
      "time_per_iteration": 2.634201765060425
    },
    {
      "auxiliary_loss_clip": 0.01172692,
      "auxiliary_loss_mlp": 0.01033876,
      "balance_loss_clip": 1.05605793,
      "balance_loss_mlp": 1.02568054,
      "epoch": 0.5901521072566585,
      "flos": 21178390233600.0,
      "grad_norm": 2.655858096419142,
      "language_loss": 0.77457035,
      "learning_rate": 1.5179368184755752e-06,
      "loss": 0.79663604,
      "num_input_tokens_seen": 105891435,
      "step": 4908,
      "time_per_iteration": 2.6824071407318115
    },
    {
      "auxiliary_loss_clip": 0.01181758,
      "auxiliary_loss_mlp": 0.01027241,
      "balance_loss_clip": 1.05575323,
      "balance_loss_mlp": 1.01912904,
      "epoch": 0.5902723501472975,
      "flos": 20225967160320.0,
      "grad_norm": 1.7726613452260875,
      "language_loss": 0.82470137,
      "learning_rate": 1.5171808496220821e-06,
      "loss": 0.84679139,
      "num_input_tokens_seen": 105910190,
      "step": 4909,
      "time_per_iteration": 2.589775323867798
    },
    {
      "auxiliary_loss_clip": 0.01185313,
      "auxiliary_loss_mlp": 0.01033551,
      "balance_loss_clip": 1.05653596,
      "balance_loss_mlp": 1.02493203,
      "epoch": 0.5903925930379367,
      "flos": 22964407211520.0,
      "grad_norm": 27.485630897673712,
      "language_loss": 0.8150394,
      "learning_rate": 1.5164249540118708e-06,
      "loss": 0.83722806,
      "num_input_tokens_seen": 105929315,
      "step": 4910,
      "time_per_iteration": 2.6326417922973633
    },
    {
      "auxiliary_loss_clip": 0.01176043,
      "auxiliary_loss_mlp": 0.01028861,
      "balance_loss_clip": 1.05716908,
      "balance_loss_mlp": 1.02062356,
      "epoch": 0.5905128359285757,
      "flos": 23367720096000.0,
      "grad_norm": 1.8807676223071343,
      "language_loss": 0.83328366,
      "learning_rate": 1.5156691317596093e-06,
      "loss": 0.85533273,
      "num_input_tokens_seen": 105950740,
      "step": 4911,
      "time_per_iteration": 2.7061240673065186
    },
    {
      "auxiliary_loss_clip": 0.01187585,
      "auxiliary_loss_mlp": 0.01035024,
      "balance_loss_clip": 1.05878699,
      "balance_loss_mlp": 1.02626252,
      "epoch": 0.5906330788192148,
      "flos": 28032335994240.0,
      "grad_norm": 2.6055566229696785,
      "language_loss": 0.66928577,
      "learning_rate": 1.5149133829799556e-06,
      "loss": 0.69151187,
      "num_input_tokens_seen": 105968735,
      "step": 4912,
      "time_per_iteration": 2.6341404914855957
    },
    {
      "auxiliary_loss_clip": 0.01169367,
      "auxiliary_loss_mlp": 0.01028013,
      "balance_loss_clip": 1.0567466,
      "balance_loss_mlp": 1.01992142,
      "epoch": 0.590753321709854,
      "flos": 18477943793280.0,
      "grad_norm": 2.068346912487531,
      "language_loss": 0.81087196,
      "learning_rate": 1.5141577077875556e-06,
      "loss": 0.83284575,
      "num_input_tokens_seen": 105986060,
      "step": 4913,
      "time_per_iteration": 2.5811703205108643
    },
    {
      "auxiliary_loss_clip": 0.01171183,
      "auxiliary_loss_mlp": 0.01030781,
      "balance_loss_clip": 1.0569427,
      "balance_loss_mlp": 1.02263868,
      "epoch": 0.590873564600493,
      "flos": 16873706568960.0,
      "grad_norm": 1.838197928941313,
      "language_loss": 0.72417986,
      "learning_rate": 1.5134021062970451e-06,
      "loss": 0.74619949,
      "num_input_tokens_seen": 106004440,
      "step": 4914,
      "time_per_iteration": 3.626840591430664
    },
    {
      "auxiliary_loss_clip": 0.01169827,
      "auxiliary_loss_mlp": 0.01034275,
      "balance_loss_clip": 1.055619,
      "balance_loss_mlp": 1.02656233,
      "epoch": 0.5909938074911321,
      "flos": 13516166678400.0,
      "grad_norm": 2.968859124848016,
      "language_loss": 0.81346416,
      "learning_rate": 1.5126465786230483e-06,
      "loss": 0.83550525,
      "num_input_tokens_seen": 106021215,
      "step": 4915,
      "time_per_iteration": 2.624659538269043
    },
    {
      "auxiliary_loss_clip": 0.01152524,
      "auxiliary_loss_mlp": 0.01032266,
      "balance_loss_clip": 1.0529021,
      "balance_loss_mlp": 1.02393293,
      "epoch": 0.5911140503817712,
      "flos": 26024067613440.0,
      "grad_norm": 1.7679187177870896,
      "language_loss": 0.82128567,
      "learning_rate": 1.5118911248801787e-06,
      "loss": 0.84313357,
      "num_input_tokens_seen": 106039225,
      "step": 4916,
      "time_per_iteration": 2.688887596130371
    },
    {
      "auxiliary_loss_clip": 0.01143864,
      "auxiliary_loss_mlp": 0.0102549,
      "balance_loss_clip": 1.05079317,
      "balance_loss_mlp": 1.01814115,
      "epoch": 0.5912342932724103,
      "flos": 23258731253760.0,
      "grad_norm": 3.711882000100448,
      "language_loss": 0.80080515,
      "learning_rate": 1.5111357451830364e-06,
      "loss": 0.82249868,
      "num_input_tokens_seen": 106057920,
      "step": 4917,
      "time_per_iteration": 2.74518084526062
    },
    {
      "auxiliary_loss_clip": 0.01182771,
      "auxiliary_loss_mlp": 0.01032095,
      "balance_loss_clip": 1.05642366,
      "balance_loss_mlp": 1.02345252,
      "epoch": 0.5913545361630493,
      "flos": 19573039687680.0,
      "grad_norm": 2.3708605487172294,
      "language_loss": 0.71212339,
      "learning_rate": 1.5103804396462131e-06,
      "loss": 0.73427212,
      "num_input_tokens_seen": 106077855,
      "step": 4918,
      "time_per_iteration": 2.5854721069335938
    },
    {
      "auxiliary_loss_clip": 0.0117581,
      "auxiliary_loss_mlp": 0.01030926,
      "balance_loss_clip": 1.05466235,
      "balance_loss_mlp": 1.02170515,
      "epoch": 0.5914747790536885,
      "flos": 26213532877440.0,
      "grad_norm": 1.9515225019637699,
      "language_loss": 0.80422318,
      "learning_rate": 1.5096252083842877e-06,
      "loss": 0.82629049,
      "num_input_tokens_seen": 106097065,
      "step": 4919,
      "time_per_iteration": 2.744478225708008
    },
    {
      "auxiliary_loss_clip": 0.01159896,
      "auxiliary_loss_mlp": 0.01030432,
      "balance_loss_clip": 1.05311108,
      "balance_loss_mlp": 1.02181911,
      "epoch": 0.5915950219443276,
      "flos": 27417545786880.0,
      "grad_norm": 2.5179635214688476,
      "language_loss": 0.85828429,
      "learning_rate": 1.5088700515118285e-06,
      "loss": 0.88018763,
      "num_input_tokens_seen": 106116385,
      "step": 4920,
      "time_per_iteration": 2.684994697570801
    },
    {
      "auxiliary_loss_clip": 0.01150896,
      "auxiliary_loss_mlp": 0.0103162,
      "balance_loss_clip": 1.05353808,
      "balance_loss_mlp": 1.02331698,
      "epoch": 0.5917152648349666,
      "flos": 21907879545600.0,
      "grad_norm": 2.0072567520566196,
      "language_loss": 0.66551614,
      "learning_rate": 1.508114969143392e-06,
      "loss": 0.68734133,
      "num_input_tokens_seen": 106136370,
      "step": 4921,
      "time_per_iteration": 2.690340042114258
    },
    {
      "auxiliary_loss_clip": 0.01159707,
      "auxiliary_loss_mlp": 0.01029745,
      "balance_loss_clip": 1.05136871,
      "balance_loss_mlp": 1.0219487,
      "epoch": 0.5918355077256057,
      "flos": 28109185142400.0,
      "grad_norm": 2.106178556339575,
      "language_loss": 0.7738415,
      "learning_rate": 1.5073599613935238e-06,
      "loss": 0.79573601,
      "num_input_tokens_seen": 106158490,
      "step": 4922,
      "time_per_iteration": 2.653378963470459
    },
    {
      "auxiliary_loss_clip": 0.0116821,
      "auxiliary_loss_mlp": 0.01033441,
      "balance_loss_clip": 1.05461264,
      "balance_loss_mlp": 1.02504277,
      "epoch": 0.5919557506162448,
      "flos": 28183807647360.0,
      "grad_norm": 2.0735467610776492,
      "language_loss": 0.57904154,
      "learning_rate": 1.5066050283767574e-06,
      "loss": 0.60105807,
      "num_input_tokens_seen": 106179170,
      "step": 4923,
      "time_per_iteration": 2.8018951416015625
    },
    {
      "auxiliary_loss_clip": 0.01183636,
      "auxiliary_loss_mlp": 0.01031933,
      "balance_loss_clip": 1.05758834,
      "balance_loss_mlp": 1.02407146,
      "epoch": 0.5920759935068839,
      "flos": 12094355652480.0,
      "grad_norm": 2.138771270580399,
      "language_loss": 0.83343101,
      "learning_rate": 1.505850170207616e-06,
      "loss": 0.85558671,
      "num_input_tokens_seen": 106196035,
      "step": 4924,
      "time_per_iteration": 2.5527560710906982
    },
    {
      "auxiliary_loss_clip": 0.01184167,
      "auxiliary_loss_mlp": 0.01035549,
      "balance_loss_clip": 1.0559622,
      "balance_loss_mlp": 1.02694845,
      "epoch": 0.592196236397523,
      "flos": 29424772673280.0,
      "grad_norm": 2.516295606398879,
      "language_loss": 0.7792207,
      "learning_rate": 1.505095387000611e-06,
      "loss": 0.80141783,
      "num_input_tokens_seen": 106218335,
      "step": 4925,
      "time_per_iteration": 2.686041831970215
    },
    {
      "auxiliary_loss_clip": 0.01155255,
      "auxiliary_loss_mlp": 0.01037965,
      "balance_loss_clip": 1.05377996,
      "balance_loss_mlp": 1.03031194,
      "epoch": 0.5923164792881621,
      "flos": 24384709866240.0,
      "grad_norm": 2.0000212345916286,
      "language_loss": 0.74712902,
      "learning_rate": 1.504340678870242e-06,
      "loss": 0.76906121,
      "num_input_tokens_seen": 106236550,
      "step": 4926,
      "time_per_iteration": 2.685742139816284
    },
    {
      "auxiliary_loss_clip": 0.01168304,
      "auxiliary_loss_mlp": 0.01024543,
      "balance_loss_clip": 1.05217397,
      "balance_loss_mlp": 1.01668072,
      "epoch": 0.5924367221788012,
      "flos": 24024238928640.0,
      "grad_norm": 1.9934425637567963,
      "language_loss": 0.89857918,
      "learning_rate": 1.5035860459309989e-06,
      "loss": 0.92050767,
      "num_input_tokens_seen": 106254265,
      "step": 4927,
      "time_per_iteration": 2.6820528507232666
    },
    {
      "auxiliary_loss_clip": 0.01170598,
      "auxiliary_loss_mlp": 0.01032714,
      "balance_loss_clip": 1.05596542,
      "balance_loss_mlp": 1.02374315,
      "epoch": 0.5925569650694402,
      "flos": 26870590414080.0,
      "grad_norm": 2.0196204721621855,
      "language_loss": 0.6377939,
      "learning_rate": 1.5028314882973568e-06,
      "loss": 0.65982705,
      "num_input_tokens_seen": 106274670,
      "step": 4928,
      "time_per_iteration": 3.63431978225708
    },
    {
      "auxiliary_loss_clip": 0.01167562,
      "auxiliary_loss_mlp": 0.01041007,
      "balance_loss_clip": 1.0543468,
      "balance_loss_mlp": 1.03300762,
      "epoch": 0.5926772079600794,
      "flos": 22302788647680.0,
      "grad_norm": 2.075281494379288,
      "language_loss": 0.85275036,
      "learning_rate": 1.502077006083783e-06,
      "loss": 0.87483609,
      "num_input_tokens_seen": 106293330,
      "step": 4929,
      "time_per_iteration": 3.619333028793335
    },
    {
      "auxiliary_loss_clip": 0.01172455,
      "auxiliary_loss_mlp": 0.01033732,
      "balance_loss_clip": 1.05592704,
      "balance_loss_mlp": 1.02576303,
      "epoch": 0.5927974508507184,
      "flos": 19865244827520.0,
      "grad_norm": 1.709158217340794,
      "language_loss": 0.76763546,
      "learning_rate": 1.5013225994047315e-06,
      "loss": 0.78969735,
      "num_input_tokens_seen": 106310960,
      "step": 4930,
      "time_per_iteration": 3.6144487857818604
    },
    {
      "auxiliary_loss_clip": 0.01168628,
      "auxiliary_loss_mlp": 0.01032069,
      "balance_loss_clip": 1.05728006,
      "balance_loss_mlp": 1.0237478,
      "epoch": 0.5929176937413575,
      "flos": 15776743167360.0,
      "grad_norm": 1.606373134674039,
      "language_loss": 0.81160969,
      "learning_rate": 1.5005682683746452e-06,
      "loss": 0.83361661,
      "num_input_tokens_seen": 106329475,
      "step": 4931,
      "time_per_iteration": 2.7102630138397217
    },
    {
      "auxiliary_loss_clip": 0.01165273,
      "auxiliary_loss_mlp": 0.00762469,
      "balance_loss_clip": 1.05438209,
      "balance_loss_mlp": 1.0012424,
      "epoch": 0.5930379366319967,
      "flos": 17601472028160.0,
      "grad_norm": 2.709685516766345,
      "language_loss": 0.7314471,
      "learning_rate": 1.4998140131079553e-06,
      "loss": 0.75072455,
      "num_input_tokens_seen": 106345565,
      "step": 4932,
      "time_per_iteration": 2.584968090057373
    },
    {
      "auxiliary_loss_clip": 0.01150027,
      "auxiliary_loss_mlp": 0.00762198,
      "balance_loss_clip": 1.05482852,
      "balance_loss_mlp": 1.00102794,
      "epoch": 0.5931581795226357,
      "flos": 17704283731200.0,
      "grad_norm": 1.7988583355525392,
      "language_loss": 0.73642933,
      "learning_rate": 1.4990598337190821e-06,
      "loss": 0.75555158,
      "num_input_tokens_seen": 106361920,
      "step": 4933,
      "time_per_iteration": 2.627645254135132
    },
    {
      "auxiliary_loss_clip": 0.01168074,
      "auxiliary_loss_mlp": 0.0103079,
      "balance_loss_clip": 1.05297995,
      "balance_loss_mlp": 1.02257597,
      "epoch": 0.5932784224132748,
      "flos": 24280102483200.0,
      "grad_norm": 1.7955928010598878,
      "language_loss": 0.68253863,
      "learning_rate": 1.4983057303224338e-06,
      "loss": 0.70452732,
      "num_input_tokens_seen": 106381735,
      "step": 4934,
      "time_per_iteration": 2.735931634902954
    },
    {
      "auxiliary_loss_clip": 0.0116921,
      "auxiliary_loss_mlp": 0.01036759,
      "balance_loss_clip": 1.05809534,
      "balance_loss_mlp": 1.02813387,
      "epoch": 0.5933986653039139,
      "flos": 22926700909440.0,
      "grad_norm": 2.329106202736107,
      "language_loss": 0.87873828,
      "learning_rate": 1.4975517030324072e-06,
      "loss": 0.90079796,
      "num_input_tokens_seen": 106399745,
      "step": 4935,
      "time_per_iteration": 2.6648542881011963
    },
    {
      "auxiliary_loss_clip": 0.01032555,
      "auxiliary_loss_mlp": 0.01009951,
      "balance_loss_clip": 1.0118382,
      "balance_loss_mlp": 1.00885427,
      "epoch": 0.593518908194553,
      "flos": 71121730256640.0,
      "grad_norm": 0.7911935557577141,
      "language_loss": 0.61765933,
      "learning_rate": 1.4967977519633882e-06,
      "loss": 0.63808441,
      "num_input_tokens_seen": 106457205,
      "step": 4936,
      "time_per_iteration": 3.391345262527466
    },
    {
      "auxiliary_loss_clip": 0.01187591,
      "auxiliary_loss_mlp": 0.01031414,
      "balance_loss_clip": 1.05764341,
      "balance_loss_mlp": 1.02292597,
      "epoch": 0.593639151085192,
      "flos": 20448649526400.0,
      "grad_norm": 2.1074335099942076,
      "language_loss": 0.78221917,
      "learning_rate": 1.4960438772297494e-06,
      "loss": 0.80440927,
      "num_input_tokens_seen": 106474250,
      "step": 4937,
      "time_per_iteration": 2.6089205741882324
    },
    {
      "auxiliary_loss_clip": 0.01187357,
      "auxiliary_loss_mlp": 0.0103677,
      "balance_loss_clip": 1.05661511,
      "balance_loss_mlp": 1.02784157,
      "epoch": 0.5937593939758312,
      "flos": 30883428074880.0,
      "grad_norm": 2.0545190648481615,
      "language_loss": 0.73519939,
      "learning_rate": 1.495290078945855e-06,
      "loss": 0.75744063,
      "num_input_tokens_seen": 106494015,
      "step": 4938,
      "time_per_iteration": 2.670753002166748
    },
    {
      "auxiliary_loss_clip": 0.01168018,
      "auxiliary_loss_mlp": 0.01030049,
      "balance_loss_clip": 1.05567241,
      "balance_loss_mlp": 1.02179956,
      "epoch": 0.5938796368664703,
      "flos": 36898069668480.0,
      "grad_norm": 1.7526208920134259,
      "language_loss": 0.74216741,
      "learning_rate": 1.4945363572260529e-06,
      "loss": 0.76414806,
      "num_input_tokens_seen": 106515010,
      "step": 4939,
      "time_per_iteration": 3.668515205383301
    },
    {
      "auxiliary_loss_clip": 0.01185431,
      "auxiliary_loss_mlp": 0.01035031,
      "balance_loss_clip": 1.05583978,
      "balance_loss_mlp": 1.02632868,
      "epoch": 0.5939998797571093,
      "flos": 23842926051840.0,
      "grad_norm": 4.174388150966821,
      "language_loss": 0.67966473,
      "learning_rate": 1.4937827121846845e-06,
      "loss": 0.70186937,
      "num_input_tokens_seen": 106535265,
      "step": 4940,
      "time_per_iteration": 2.627500295639038
    },
    {
      "auxiliary_loss_clip": 0.01189469,
      "auxiliary_loss_mlp": 0.01032666,
      "balance_loss_clip": 1.06061196,
      "balance_loss_mlp": 1.0246079,
      "epoch": 0.5941201226477485,
      "flos": 25191407462400.0,
      "grad_norm": 1.5548020918888115,
      "language_loss": 0.73443711,
      "learning_rate": 1.4930291439360755e-06,
      "loss": 0.75665843,
      "num_input_tokens_seen": 106557830,
      "step": 4941,
      "time_per_iteration": 2.6581902503967285
    },
    {
      "auxiliary_loss_clip": 0.01189601,
      "auxiliary_loss_mlp": 0.01035339,
      "balance_loss_clip": 1.0581398,
      "balance_loss_mlp": 1.02719736,
      "epoch": 0.5942403655383875,
      "flos": 22418996123520.0,
      "grad_norm": 2.2052503279488107,
      "language_loss": 0.79451978,
      "learning_rate": 1.4922756525945427e-06,
      "loss": 0.81676924,
      "num_input_tokens_seen": 106577140,
      "step": 4942,
      "time_per_iteration": 2.6720662117004395
    },
    {
      "auxiliary_loss_clip": 0.01061083,
      "auxiliary_loss_mlp": 0.010047,
      "balance_loss_clip": 1.01236594,
      "balance_loss_mlp": 1.00366867,
      "epoch": 0.5943606084290266,
      "flos": 67629310796160.0,
      "grad_norm": 0.7759605881690064,
      "language_loss": 0.59511852,
      "learning_rate": 1.4915222382743894e-06,
      "loss": 0.61577642,
      "num_input_tokens_seen": 106635975,
      "step": 4943,
      "time_per_iteration": 3.276686906814575
    },
    {
      "auxiliary_loss_clip": 0.01190186,
      "auxiliary_loss_mlp": 0.01029032,
      "balance_loss_clip": 1.06003082,
      "balance_loss_mlp": 1.02032948,
      "epoch": 0.5944808513196658,
      "flos": 18223157646720.0,
      "grad_norm": 1.8787206933909462,
      "language_loss": 0.71980047,
      "learning_rate": 1.4907689010899085e-06,
      "loss": 0.74199271,
      "num_input_tokens_seen": 106653555,
      "step": 4944,
      "time_per_iteration": 2.5927765369415283
    },
    {
      "auxiliary_loss_clip": 0.01183136,
      "auxiliary_loss_mlp": 0.01028598,
      "balance_loss_clip": 1.05453801,
      "balance_loss_mlp": 1.02027178,
      "epoch": 0.5946010942103048,
      "flos": 24790824011520.0,
      "grad_norm": 2.398444865049158,
      "language_loss": 0.62861019,
      "learning_rate": 1.4900156411553804e-06,
      "loss": 0.65072751,
      "num_input_tokens_seen": 106673385,
      "step": 4945,
      "time_per_iteration": 2.6176998615264893
    },
    {
      "auxiliary_loss_clip": 0.01178766,
      "auxiliary_loss_mlp": 0.01036885,
      "balance_loss_clip": 1.05802596,
      "balance_loss_mlp": 1.02848697,
      "epoch": 0.5947213371009439,
      "flos": 15231619388160.0,
      "grad_norm": 1.9278702067400313,
      "language_loss": 0.85687768,
      "learning_rate": 1.4892624585850739e-06,
      "loss": 0.87903416,
      "num_input_tokens_seen": 106691740,
      "step": 4946,
      "time_per_iteration": 2.619886875152588
    },
    {
      "auxiliary_loss_clip": 0.01175208,
      "auxiliary_loss_mlp": 0.01029399,
      "balance_loss_clip": 1.05562401,
      "balance_loss_mlp": 1.02126288,
      "epoch": 0.594841579991583,
      "flos": 25848069949440.0,
      "grad_norm": 2.2140074358289885,
      "language_loss": 0.80020905,
      "learning_rate": 1.4885093534932465e-06,
      "loss": 0.82225513,
      "num_input_tokens_seen": 106709705,
      "step": 4947,
      "time_per_iteration": 2.741140127182007
    },
    {
      "auxiliary_loss_clip": 0.01175808,
      "auxiliary_loss_mlp": 0.01035896,
      "balance_loss_clip": 1.05745471,
      "balance_loss_mlp": 1.02744365,
      "epoch": 0.5949618228822221,
      "flos": 23981109672960.0,
      "grad_norm": 2.1192598663809794,
      "language_loss": 0.71782893,
      "learning_rate": 1.4877563259941433e-06,
      "loss": 0.73994589,
      "num_input_tokens_seen": 106727560,
      "step": 4948,
      "time_per_iteration": 2.631538152694702
    },
    {
      "auxiliary_loss_clip": 0.01174465,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 1.05396056,
      "balance_loss_mlp": 1.02290857,
      "epoch": 0.5950820657728612,
      "flos": 40547491476480.0,
      "grad_norm": 2.1899408022716194,
      "language_loss": 0.67762983,
      "learning_rate": 1.4870033762019988e-06,
      "loss": 0.69968677,
      "num_input_tokens_seen": 106747725,
      "step": 4949,
      "time_per_iteration": 2.8105673789978027
    },
    {
      "auxiliary_loss_clip": 0.01173613,
      "auxiliary_loss_mlp": 0.01035523,
      "balance_loss_clip": 1.05718875,
      "balance_loss_mlp": 1.02718997,
      "epoch": 0.5952023086635003,
      "flos": 23184467884800.0,
      "grad_norm": 1.7437755296121789,
      "language_loss": 0.73080605,
      "learning_rate": 1.4862505042310334e-06,
      "loss": 0.75289744,
      "num_input_tokens_seen": 106767010,
      "step": 4950,
      "time_per_iteration": 2.5949790477752686
    },
    {
      "auxiliary_loss_clip": 0.01134331,
      "auxiliary_loss_mlp": 0.01030257,
      "balance_loss_clip": 1.05101967,
      "balance_loss_mlp": 1.0219959,
      "epoch": 0.5953225515541394,
      "flos": 33653289548160.0,
      "grad_norm": 1.8275418469804383,
      "language_loss": 0.69888788,
      "learning_rate": 1.4854977101954587e-06,
      "loss": 0.72053379,
      "num_input_tokens_seen": 106789230,
      "step": 4951,
      "time_per_iteration": 2.846792459487915
    },
    {
      "auxiliary_loss_clip": 0.01185041,
      "auxiliary_loss_mlp": 0.01032859,
      "balance_loss_clip": 1.05561781,
      "balance_loss_mlp": 1.02434754,
      "epoch": 0.5954427944447784,
      "flos": 24459619680000.0,
      "grad_norm": 1.9403571539684232,
      "language_loss": 0.86605561,
      "learning_rate": 1.4847449942094716e-06,
      "loss": 0.88823462,
      "num_input_tokens_seen": 106808110,
      "step": 4952,
      "time_per_iteration": 2.595863103866577
    },
    {
      "auxiliary_loss_clip": 0.01180864,
      "auxiliary_loss_mlp": 0.01030205,
      "balance_loss_clip": 1.05362582,
      "balance_loss_mlp": 1.02131844,
      "epoch": 0.5955630373354175,
      "flos": 18551848026240.0,
      "grad_norm": 2.01830012230911,
      "language_loss": 0.86277592,
      "learning_rate": 1.4839923563872598e-06,
      "loss": 0.88488662,
      "num_input_tokens_seen": 106826650,
      "step": 4953,
      "time_per_iteration": 2.605459451675415
    },
    {
      "auxiliary_loss_clip": 0.0118511,
      "auxiliary_loss_mlp": 0.01038781,
      "balance_loss_clip": 1.05703497,
      "balance_loss_mlp": 1.0299058,
      "epoch": 0.5956832802260567,
      "flos": 19791699730560.0,
      "grad_norm": 2.206747617648311,
      "language_loss": 0.76033324,
      "learning_rate": 1.483239796842997e-06,
      "loss": 0.78257215,
      "num_input_tokens_seen": 106844680,
      "step": 4954,
      "time_per_iteration": 2.5763614177703857
    },
    {
      "auxiliary_loss_clip": 0.01154374,
      "auxiliary_loss_mlp": 0.01025762,
      "balance_loss_clip": 1.05388474,
      "balance_loss_mlp": 1.01751852,
      "epoch": 0.5958035231166957,
      "flos": 19750868945280.0,
      "grad_norm": 1.6744611169508756,
      "language_loss": 0.83994967,
      "learning_rate": 1.4824873156908462e-06,
      "loss": 0.86175108,
      "num_input_tokens_seen": 106862605,
      "step": 4955,
      "time_per_iteration": 3.6561193466186523
    },
    {
      "auxiliary_loss_clip": 0.01189655,
      "auxiliary_loss_mlp": 0.01034771,
      "balance_loss_clip": 1.05878115,
      "balance_loss_mlp": 1.02598524,
      "epoch": 0.5959237660073348,
      "flos": 21652806090240.0,
      "grad_norm": 2.107690321828789,
      "language_loss": 0.75715309,
      "learning_rate": 1.4817349130449584e-06,
      "loss": 0.77939737,
      "num_input_tokens_seen": 106882325,
      "step": 4956,
      "time_per_iteration": 4.535318851470947
    },
    {
      "auxiliary_loss_clip": 0.01167181,
      "auxiliary_loss_mlp": 0.01033125,
      "balance_loss_clip": 1.05422926,
      "balance_loss_mlp": 1.02453041,
      "epoch": 0.5960440088979739,
      "flos": 21171207513600.0,
      "grad_norm": 2.242525602050163,
      "language_loss": 0.83120692,
      "learning_rate": 1.4809825890194717e-06,
      "loss": 0.85320997,
      "num_input_tokens_seen": 106900995,
      "step": 4957,
      "time_per_iteration": 2.6134467124938965
    },
    {
      "auxiliary_loss_clip": 0.01152379,
      "auxiliary_loss_mlp": 0.01031491,
      "balance_loss_clip": 1.05337203,
      "balance_loss_mlp": 1.02343249,
      "epoch": 0.596164251788613,
      "flos": 14757526753920.0,
      "grad_norm": 1.8505580401814952,
      "language_loss": 0.77873999,
      "learning_rate": 1.4802303437285139e-06,
      "loss": 0.80057871,
      "num_input_tokens_seen": 106918265,
      "step": 4958,
      "time_per_iteration": 2.5732614994049072
    },
    {
      "auxiliary_loss_clip": 0.01140032,
      "auxiliary_loss_mlp": 0.01036713,
      "balance_loss_clip": 1.0479027,
      "balance_loss_mlp": 1.02824283,
      "epoch": 0.596284494679252,
      "flos": 20485924865280.0,
      "grad_norm": 2.468837109135881,
      "language_loss": 0.80656344,
      "learning_rate": 1.4794781772861994e-06,
      "loss": 0.82833093,
      "num_input_tokens_seen": 106934760,
      "step": 4959,
      "time_per_iteration": 2.7209224700927734
    },
    {
      "auxiliary_loss_clip": 0.01183842,
      "auxiliary_loss_mlp": 0.01036189,
      "balance_loss_clip": 1.0575459,
      "balance_loss_mlp": 1.0279994,
      "epoch": 0.5964047375698912,
      "flos": 31212262108800.0,
      "grad_norm": 2.456725146448249,
      "language_loss": 0.66933459,
      "learning_rate": 1.4787260898066324e-06,
      "loss": 0.69153488,
      "num_input_tokens_seen": 106954760,
      "step": 4960,
      "time_per_iteration": 2.6841471195220947
    },
    {
      "auxiliary_loss_clip": 0.01167542,
      "auxiliary_loss_mlp": 0.01031068,
      "balance_loss_clip": 1.05423188,
      "balance_loss_mlp": 1.02303886,
      "epoch": 0.5965249804605303,
      "flos": 27483620855040.0,
      "grad_norm": 2.6513669951976278,
      "language_loss": 0.86351198,
      "learning_rate": 1.4779740814039023e-06,
      "loss": 0.88549811,
      "num_input_tokens_seen": 106974845,
      "step": 4961,
      "time_per_iteration": 2.7387590408325195
    },
    {
      "auxiliary_loss_clip": 0.01183044,
      "auxiliary_loss_mlp": 0.01031782,
      "balance_loss_clip": 1.05510402,
      "balance_loss_mlp": 1.02332461,
      "epoch": 0.5966452233511693,
      "flos": 30773936442240.0,
      "grad_norm": 1.9273190079525118,
      "language_loss": 0.68639517,
      "learning_rate": 1.4772221521920894e-06,
      "loss": 0.70854342,
      "num_input_tokens_seen": 106994870,
      "step": 4962,
      "time_per_iteration": 2.651170253753662
    },
    {
      "auxiliary_loss_clip": 0.0118379,
      "auxiliary_loss_mlp": 0.01032261,
      "balance_loss_clip": 1.05742741,
      "balance_loss_mlp": 1.02387404,
      "epoch": 0.5967654662418085,
      "flos": 25481170477440.0,
      "grad_norm": 2.078509751128197,
      "language_loss": 0.7455225,
      "learning_rate": 1.4764703022852598e-06,
      "loss": 0.76768303,
      "num_input_tokens_seen": 107015390,
      "step": 4963,
      "time_per_iteration": 2.6666460037231445
    },
    {
      "auxiliary_loss_clip": 0.01143226,
      "auxiliary_loss_mlp": 0.01026901,
      "balance_loss_clip": 1.0473938,
      "balance_loss_mlp": 1.01843715,
      "epoch": 0.5968857091324475,
      "flos": 19099126621440.0,
      "grad_norm": 1.7619650206955437,
      "language_loss": 0.77010763,
      "learning_rate": 1.4757185317974696e-06,
      "loss": 0.79180896,
      "num_input_tokens_seen": 107033775,
      "step": 4964,
      "time_per_iteration": 2.6534531116485596
    },
    {
      "auxiliary_loss_clip": 0.01168506,
      "auxiliary_loss_mlp": 0.01026875,
      "balance_loss_clip": 1.05440497,
      "balance_loss_mlp": 1.01814342,
      "epoch": 0.5970059520230866,
      "flos": 23692711374720.0,
      "grad_norm": 2.350987021621597,
      "language_loss": 0.71113485,
      "learning_rate": 1.474966840842761e-06,
      "loss": 0.73308867,
      "num_input_tokens_seen": 107053355,
      "step": 4965,
      "time_per_iteration": 2.7501749992370605
    },
    {
      "auxiliary_loss_clip": 0.01156501,
      "auxiliary_loss_mlp": 0.01033006,
      "balance_loss_clip": 1.05063462,
      "balance_loss_mlp": 1.02516162,
      "epoch": 0.5971261949137258,
      "flos": 23185545292800.0,
      "grad_norm": 1.7983355062264108,
      "language_loss": 0.86807168,
      "learning_rate": 1.4742152295351655e-06,
      "loss": 0.88996673,
      "num_input_tokens_seen": 107072510,
      "step": 4966,
      "time_per_iteration": 3.810215473175049
    },
    {
      "auxiliary_loss_clip": 0.01175333,
      "auxiliary_loss_mlp": 0.01035626,
      "balance_loss_clip": 1.05301929,
      "balance_loss_mlp": 1.02738845,
      "epoch": 0.5972464378043648,
      "flos": 20557710195840.0,
      "grad_norm": 2.6673711225525643,
      "language_loss": 0.64229536,
      "learning_rate": 1.4734636979887016e-06,
      "loss": 0.66440499,
      "num_input_tokens_seen": 107089970,
      "step": 4967,
      "time_per_iteration": 2.6436045169830322
    },
    {
      "auxiliary_loss_clip": 0.01165536,
      "auxiliary_loss_mlp": 0.01033871,
      "balance_loss_clip": 1.05318308,
      "balance_loss_mlp": 1.02525187,
      "epoch": 0.5973666806950039,
      "flos": 29387030457600.0,
      "grad_norm": 2.864027277689752,
      "language_loss": 0.90557146,
      "learning_rate": 1.4727122463173755e-06,
      "loss": 0.92756552,
      "num_input_tokens_seen": 107108500,
      "step": 4968,
      "time_per_iteration": 2.7103254795074463
    },
    {
      "auxiliary_loss_clip": 0.01186497,
      "auxiliary_loss_mlp": 0.01026988,
      "balance_loss_clip": 1.05750704,
      "balance_loss_mlp": 1.01861358,
      "epoch": 0.597486923585643,
      "flos": 22273522041600.0,
      "grad_norm": 1.7377229394440392,
      "language_loss": 0.6438216,
      "learning_rate": 1.471960874635183e-06,
      "loss": 0.66595638,
      "num_input_tokens_seen": 107128060,
      "step": 4969,
      "time_per_iteration": 2.569364070892334
    },
    {
      "auxiliary_loss_clip": 0.0118532,
      "auxiliary_loss_mlp": 0.01037966,
      "balance_loss_clip": 1.05455995,
      "balance_loss_mlp": 1.02863812,
      "epoch": 0.5976071664762821,
      "flos": 13772461196160.0,
      "grad_norm": 2.8419796175737764,
      "language_loss": 0.70804036,
      "learning_rate": 1.4712095830561055e-06,
      "loss": 0.73027325,
      "num_input_tokens_seen": 107146550,
      "step": 4970,
      "time_per_iteration": 2.651193857192993
    },
    {
      "auxiliary_loss_clip": 0.01168787,
      "auxiliary_loss_mlp": 0.01029924,
      "balance_loss_clip": 1.0538317,
      "balance_loss_mlp": 1.02185333,
      "epoch": 0.5977274093669211,
      "flos": 19098623831040.0,
      "grad_norm": 2.8618888703675,
      "language_loss": 0.80824488,
      "learning_rate": 1.4704583716941147e-06,
      "loss": 0.83023196,
      "num_input_tokens_seen": 107165415,
      "step": 4971,
      "time_per_iteration": 2.6162734031677246
    },
    {
      "auxiliary_loss_clip": 0.01147848,
      "auxiliary_loss_mlp": 0.01031564,
      "balance_loss_clip": 1.05211163,
      "balance_loss_mlp": 1.0228796,
      "epoch": 0.5978476522575603,
      "flos": 20376002269440.0,
      "grad_norm": 1.6338349350958519,
      "language_loss": 0.72066808,
      "learning_rate": 1.4697072406631672e-06,
      "loss": 0.74246216,
      "num_input_tokens_seen": 107185320,
      "step": 4972,
      "time_per_iteration": 2.699069023132324
    },
    {
      "auxiliary_loss_clip": 0.0114516,
      "auxiliary_loss_mlp": 0.01032056,
      "balance_loss_clip": 1.05532813,
      "balance_loss_mlp": 1.02325249,
      "epoch": 0.5979678951481994,
      "flos": 29023147728000.0,
      "grad_norm": 1.9666559120523364,
      "language_loss": 0.73012292,
      "learning_rate": 1.4689561900772097e-06,
      "loss": 0.75189513,
      "num_input_tokens_seen": 107205380,
      "step": 4973,
      "time_per_iteration": 2.8081367015838623
    },
    {
      "auxiliary_loss_clip": 0.01170966,
      "auxiliary_loss_mlp": 0.01030158,
      "balance_loss_clip": 1.05373549,
      "balance_loss_mlp": 1.02203393,
      "epoch": 0.5980881380388384,
      "flos": 17967689141760.0,
      "grad_norm": 2.2241780825559396,
      "language_loss": 0.72857523,
      "learning_rate": 1.4682052200501758e-06,
      "loss": 0.75058651,
      "num_input_tokens_seen": 107222585,
      "step": 4974,
      "time_per_iteration": 2.596470355987549
    },
    {
      "auxiliary_loss_clip": 0.01166771,
      "auxiliary_loss_mlp": 0.01031872,
      "balance_loss_clip": 1.05275655,
      "balance_loss_mlp": 1.0234592,
      "epoch": 0.5982083809294776,
      "flos": 22962827013120.0,
      "grad_norm": 2.756523691004193,
      "language_loss": 0.80118263,
      "learning_rate": 1.4674543306959876e-06,
      "loss": 0.82316911,
      "num_input_tokens_seen": 107242055,
      "step": 4975,
      "time_per_iteration": 2.5850257873535156
    },
    {
      "auxiliary_loss_clip": 0.01162526,
      "auxiliary_loss_mlp": 0.0103519,
      "balance_loss_clip": 1.05307388,
      "balance_loss_mlp": 1.02661252,
      "epoch": 0.5983286238201166,
      "flos": 20991941712000.0,
      "grad_norm": 2.4947569250065693,
      "language_loss": 0.8458004,
      "learning_rate": 1.4667035221285535e-06,
      "loss": 0.86777759,
      "num_input_tokens_seen": 107259695,
      "step": 4976,
      "time_per_iteration": 2.7443766593933105
    },
    {
      "auxiliary_loss_clip": 0.01187125,
      "auxiliary_loss_mlp": 0.01031159,
      "balance_loss_clip": 1.05826652,
      "balance_loss_mlp": 1.02276039,
      "epoch": 0.5984488667107557,
      "flos": 28183448511360.0,
      "grad_norm": 1.7455178592089613,
      "language_loss": 0.74123293,
      "learning_rate": 1.4659527944617715e-06,
      "loss": 0.76341575,
      "num_input_tokens_seen": 107279640,
      "step": 4977,
      "time_per_iteration": 2.5974833965301514
    },
    {
      "auxiliary_loss_clip": 0.01184716,
      "auxiliary_loss_mlp": 0.01033966,
      "balance_loss_clip": 1.05604184,
      "balance_loss_mlp": 1.02606821,
      "epoch": 0.5985691096013949,
      "flos": 16471794314880.0,
      "grad_norm": 1.8284737786223977,
      "language_loss": 0.76172316,
      "learning_rate": 1.465202147809526e-06,
      "loss": 0.78390992,
      "num_input_tokens_seen": 107298135,
      "step": 4978,
      "time_per_iteration": 2.6511547565460205
    },
    {
      "auxiliary_loss_clip": 0.01134093,
      "auxiliary_loss_mlp": 0.01033931,
      "balance_loss_clip": 1.04872334,
      "balance_loss_mlp": 1.02589083,
      "epoch": 0.5986893524920339,
      "flos": 26719046933760.0,
      "grad_norm": 2.104317031893652,
      "language_loss": 0.7692734,
      "learning_rate": 1.4644515822856888e-06,
      "loss": 0.79095364,
      "num_input_tokens_seen": 107316570,
      "step": 4979,
      "time_per_iteration": 2.7570960521698
    },
    {
      "auxiliary_loss_clip": 0.01052007,
      "auxiliary_loss_mlp": 0.01002376,
      "balance_loss_clip": 1.01166224,
      "balance_loss_mlp": 1.00128496,
      "epoch": 0.598809595382673,
      "flos": 61608061100160.0,
      "grad_norm": 0.752215583709229,
      "language_loss": 0.56460446,
      "learning_rate": 1.4637010980041215e-06,
      "loss": 0.58514833,
      "num_input_tokens_seen": 107378680,
      "step": 4980,
      "time_per_iteration": 3.218928813934326
    },
    {
      "auxiliary_loss_clip": 0.0115502,
      "auxiliary_loss_mlp": 0.01037033,
      "balance_loss_clip": 1.05635905,
      "balance_loss_mlp": 1.02792561,
      "epoch": 0.5989298382733121,
      "flos": 11801719549440.0,
      "grad_norm": 2.1793851164852684,
      "language_loss": 0.89791793,
      "learning_rate": 1.4629506950786707e-06,
      "loss": 0.91983843,
      "num_input_tokens_seen": 107394860,
      "step": 4981,
      "time_per_iteration": 3.579226016998291
    },
    {
      "auxiliary_loss_clip": 0.01059805,
      "auxiliary_loss_mlp": 0.01001006,
      "balance_loss_clip": 1.00954127,
      "balance_loss_mlp": 0.99997443,
      "epoch": 0.5990500811639512,
      "flos": 60025800021120.0,
      "grad_norm": 0.8093181224447104,
      "language_loss": 0.56032115,
      "learning_rate": 1.4622003736231733e-06,
      "loss": 0.58092922,
      "num_input_tokens_seen": 107453850,
      "step": 4982,
      "time_per_iteration": 4.178200960159302
    },
    {
      "auxiliary_loss_clip": 0.01170338,
      "auxiliary_loss_mlp": 0.01037411,
      "balance_loss_clip": 1.05285513,
      "balance_loss_mlp": 1.0285356,
      "epoch": 0.5991703240545903,
      "flos": 18222726683520.0,
      "grad_norm": 1.9180166327166956,
      "language_loss": 0.81231046,
      "learning_rate": 1.461450133751451e-06,
      "loss": 0.8343879,
      "num_input_tokens_seen": 107471920,
      "step": 4983,
      "time_per_iteration": 3.5798757076263428
    },
    {
      "auxiliary_loss_clip": 0.01171988,
      "auxiliary_loss_mlp": 0.01030944,
      "balance_loss_clip": 1.05501759,
      "balance_loss_mlp": 1.02217638,
      "epoch": 0.5992905669452293,
      "flos": 27709894581120.0,
      "grad_norm": 3.1503256156016795,
      "language_loss": 0.76309645,
      "learning_rate": 1.4606999755773153e-06,
      "loss": 0.78512573,
      "num_input_tokens_seen": 107493125,
      "step": 4984,
      "time_per_iteration": 2.7619199752807617
    },
    {
      "auxiliary_loss_clip": 0.01185476,
      "auxiliary_loss_mlp": 0.01033216,
      "balance_loss_clip": 1.05817342,
      "balance_loss_mlp": 1.02543139,
      "epoch": 0.5994108098358685,
      "flos": 20449008662400.0,
      "grad_norm": 2.0923151123598474,
      "language_loss": 0.82511479,
      "learning_rate": 1.4599498992145643e-06,
      "loss": 0.84730172,
      "num_input_tokens_seen": 107513150,
      "step": 4985,
      "time_per_iteration": 2.711608648300171
    },
    {
      "auxiliary_loss_clip": 0.01148416,
      "auxiliary_loss_mlp": 0.01031272,
      "balance_loss_clip": 1.05342495,
      "balance_loss_mlp": 1.02304101,
      "epoch": 0.5995310527265075,
      "flos": 22269966595200.0,
      "grad_norm": 5.119064261137596,
      "language_loss": 0.71027368,
      "learning_rate": 1.4591999047769846e-06,
      "loss": 0.73207057,
      "num_input_tokens_seen": 107532005,
      "step": 4986,
      "time_per_iteration": 2.6754424571990967
    },
    {
      "auxiliary_loss_clip": 0.01187253,
      "auxiliary_loss_mlp": 0.01032009,
      "balance_loss_clip": 1.05793786,
      "balance_loss_mlp": 1.02405822,
      "epoch": 0.5996512956171466,
      "flos": 18916951818240.0,
      "grad_norm": 1.7388199446383377,
      "language_loss": 0.75715685,
      "learning_rate": 1.4584499923783486e-06,
      "loss": 0.77934951,
      "num_input_tokens_seen": 107550585,
      "step": 4987,
      "time_per_iteration": 2.6320254802703857
    },
    {
      "auxiliary_loss_clip": 0.01183043,
      "auxiliary_loss_mlp": 0.00762792,
      "balance_loss_clip": 1.05627215,
      "balance_loss_mlp": 1.00099325,
      "epoch": 0.5997715385077858,
      "flos": 15370916330880.0,
      "grad_norm": 1.9118637669457346,
      "language_loss": 0.76180232,
      "learning_rate": 1.457700162132419e-06,
      "loss": 0.78126067,
      "num_input_tokens_seen": 107567575,
      "step": 4988,
      "time_per_iteration": 2.599231481552124
    },
    {
      "auxiliary_loss_clip": 0.01184333,
      "auxiliary_loss_mlp": 0.01026825,
      "balance_loss_clip": 1.05600512,
      "balance_loss_mlp": 1.01758599,
      "epoch": 0.5998917813984248,
      "flos": 25264844818560.0,
      "grad_norm": 2.27405423416835,
      "language_loss": 0.72450292,
      "learning_rate": 1.4569504141529433e-06,
      "loss": 0.74661446,
      "num_input_tokens_seen": 107585410,
      "step": 4989,
      "time_per_iteration": 2.6532070636749268
    },
    {
      "auxiliary_loss_clip": 0.0117276,
      "auxiliary_loss_mlp": 0.01034485,
      "balance_loss_clip": 1.05192709,
      "balance_loss_mlp": 1.02578855,
      "epoch": 0.6000120242890639,
      "flos": 22054502862720.0,
      "grad_norm": 2.52512561345973,
      "language_loss": 0.72120953,
      "learning_rate": 1.456200748553658e-06,
      "loss": 0.74328196,
      "num_input_tokens_seen": 107603405,
      "step": 4990,
      "time_per_iteration": 2.6225292682647705
    },
    {
      "auxiliary_loss_clip": 0.01188382,
      "auxiliary_loss_mlp": 0.01035416,
      "balance_loss_clip": 1.05808496,
      "balance_loss_mlp": 1.02673221,
      "epoch": 0.600132267179703,
      "flos": 29863421562240.0,
      "grad_norm": 1.6904750493208924,
      "language_loss": 0.78958285,
      "learning_rate": 1.455451165448287e-06,
      "loss": 0.81182086,
      "num_input_tokens_seen": 107626060,
      "step": 4991,
      "time_per_iteration": 2.7046449184417725
    },
    {
      "auxiliary_loss_clip": 0.01184813,
      "auxiliary_loss_mlp": 0.01030532,
      "balance_loss_clip": 1.05593359,
      "balance_loss_mlp": 1.02137029,
      "epoch": 0.6002525100703421,
      "flos": 25045358762880.0,
      "grad_norm": 4.813941482379739,
      "language_loss": 0.73789299,
      "learning_rate": 1.4547016649505407e-06,
      "loss": 0.76004636,
      "num_input_tokens_seen": 107644070,
      "step": 4992,
      "time_per_iteration": 3.5520758628845215
    },
    {
      "auxiliary_loss_clip": 0.01186074,
      "auxiliary_loss_mlp": 0.01029725,
      "balance_loss_clip": 1.05713332,
      "balance_loss_mlp": 1.02086151,
      "epoch": 0.6003727529609811,
      "flos": 20849592113280.0,
      "grad_norm": 3.721079510239833,
      "language_loss": 0.84570742,
      "learning_rate": 1.4539522471741193e-06,
      "loss": 0.86786544,
      "num_input_tokens_seen": 107661495,
      "step": 4993,
      "time_per_iteration": 2.8748393058776855
    },
    {
      "auxiliary_loss_clip": 0.01189686,
      "auxiliary_loss_mlp": 0.01035941,
      "balance_loss_clip": 1.0566926,
      "balance_loss_mlp": 1.02645779,
      "epoch": 0.6004929958516203,
      "flos": 15594604277760.0,
      "grad_norm": 2.4175178238589212,
      "language_loss": 0.7081989,
      "learning_rate": 1.4532029122327067e-06,
      "loss": 0.73045516,
      "num_input_tokens_seen": 107678280,
      "step": 4994,
      "time_per_iteration": 2.700225830078125
    },
    {
      "auxiliary_loss_clip": 0.01161885,
      "auxiliary_loss_mlp": 0.0103491,
      "balance_loss_clip": 1.05125523,
      "balance_loss_mlp": 1.02665508,
      "epoch": 0.6006132387422594,
      "flos": 21763267390080.0,
      "grad_norm": 2.8473055804624856,
      "language_loss": 0.75658774,
      "learning_rate": 1.4524536602399783e-06,
      "loss": 0.77855575,
      "num_input_tokens_seen": 107697370,
      "step": 4995,
      "time_per_iteration": 2.733262300491333
    },
    {
      "auxiliary_loss_clip": 0.01185537,
      "auxiliary_loss_mlp": 0.01027965,
      "balance_loss_clip": 1.05963397,
      "balance_loss_mlp": 1.01974583,
      "epoch": 0.6007334816328984,
      "flos": 22858542852480.0,
      "grad_norm": 1.7240901194814846,
      "language_loss": 0.77404988,
      "learning_rate": 1.4517044913095938e-06,
      "loss": 0.7961849,
      "num_input_tokens_seen": 107717790,
      "step": 4996,
      "time_per_iteration": 2.6803126335144043
    },
    {
      "auxiliary_loss_clip": 0.01133897,
      "auxiliary_loss_mlp": 0.01036813,
      "balance_loss_clip": 1.05068529,
      "balance_loss_mlp": 1.02772367,
      "epoch": 0.6008537245235376,
      "flos": 28324577047680.0,
      "grad_norm": 1.939384667438972,
      "language_loss": 0.82067132,
      "learning_rate": 1.4509554055552022e-06,
      "loss": 0.84237844,
      "num_input_tokens_seen": 107738020,
      "step": 4997,
      "time_per_iteration": 2.8573694229125977
    },
    {
      "auxiliary_loss_clip": 0.01153834,
      "auxiliary_loss_mlp": 0.01029653,
      "balance_loss_clip": 1.05697346,
      "balance_loss_mlp": 1.02084947,
      "epoch": 0.6009739674141766,
      "flos": 20886113266560.0,
      "grad_norm": 2.2853337706646717,
      "language_loss": 0.84149015,
      "learning_rate": 1.450206403090439e-06,
      "loss": 0.86332512,
      "num_input_tokens_seen": 107756215,
      "step": 4998,
      "time_per_iteration": 2.7252867221832275
    },
    {
      "auxiliary_loss_clip": 0.0114381,
      "auxiliary_loss_mlp": 0.01032238,
      "balance_loss_clip": 1.055022,
      "balance_loss_mlp": 1.02413821,
      "epoch": 0.6010942103048157,
      "flos": 20481004702080.0,
      "grad_norm": 2.523060822129674,
      "language_loss": 0.87058705,
      "learning_rate": 1.4494574840289274e-06,
      "loss": 0.89234751,
      "num_input_tokens_seen": 107773330,
      "step": 4999,
      "time_per_iteration": 2.7863614559173584
    },
    {
      "auxiliary_loss_clip": 0.01170881,
      "auxiliary_loss_mlp": 0.010345,
      "balance_loss_clip": 1.05360389,
      "balance_loss_mlp": 1.02482009,
      "epoch": 0.6012144531954549,
      "flos": 23805973935360.0,
      "grad_norm": 1.6318386717270823,
      "language_loss": 0.73884058,
      "learning_rate": 1.4487086484842782e-06,
      "loss": 0.76089442,
      "num_input_tokens_seen": 107791975,
      "step": 5000,
      "time_per_iteration": 2.7496700286865234
    },
    {
      "auxiliary_loss_clip": 0.01168484,
      "auxiliary_loss_mlp": 0.01030097,
      "balance_loss_clip": 1.05226231,
      "balance_loss_mlp": 1.0220685,
      "epoch": 0.6013346960860939,
      "flos": 18988378012800.0,
      "grad_norm": 1.9254552807979528,
      "language_loss": 0.60339558,
      "learning_rate": 1.4479598965700878e-06,
      "loss": 0.62538141,
      "num_input_tokens_seen": 107809240,
      "step": 5001,
      "time_per_iteration": 2.8383679389953613
    },
    {
      "auxiliary_loss_clip": 0.0116716,
      "auxiliary_loss_mlp": 0.01027327,
      "balance_loss_clip": 1.05392587,
      "balance_loss_mlp": 1.01925671,
      "epoch": 0.601454938976733,
      "flos": 24025316336640.0,
      "grad_norm": 2.5647007757612754,
      "language_loss": 0.6932686,
      "learning_rate": 1.4472112283999427e-06,
      "loss": 0.71521348,
      "num_input_tokens_seen": 107827895,
      "step": 5002,
      "time_per_iteration": 2.6867001056671143
    },
    {
      "auxiliary_loss_clip": 0.0115094,
      "auxiliary_loss_mlp": 0.01029403,
      "balance_loss_clip": 1.05097818,
      "balance_loss_mlp": 1.02110577,
      "epoch": 0.6015751818673721,
      "flos": 26427129102720.0,
      "grad_norm": 2.4005063935547697,
      "language_loss": 0.69524604,
      "learning_rate": 1.4464626440874143e-06,
      "loss": 0.71704948,
      "num_input_tokens_seen": 107847010,
      "step": 5003,
      "time_per_iteration": 2.7661941051483154
    },
    {
      "auxiliary_loss_clip": 0.01186568,
      "auxiliary_loss_mlp": 0.01026657,
      "balance_loss_clip": 1.05652547,
      "balance_loss_mlp": 1.0174247,
      "epoch": 0.6016954247580112,
      "flos": 13115260005120.0,
      "grad_norm": 2.349284289161247,
      "language_loss": 0.74977911,
      "learning_rate": 1.4457141437460636e-06,
      "loss": 0.77191132,
      "num_input_tokens_seen": 107864235,
      "step": 5004,
      "time_per_iteration": 2.681804656982422
    },
    {
      "auxiliary_loss_clip": 0.01143874,
      "auxiliary_loss_mlp": 0.01029727,
      "balance_loss_clip": 1.04838967,
      "balance_loss_mlp": 1.02068472,
      "epoch": 0.6018156676486502,
      "flos": 23768447201280.0,
      "grad_norm": 1.7519984073077606,
      "language_loss": 0.73602325,
      "learning_rate": 1.444965727489436e-06,
      "loss": 0.75775927,
      "num_input_tokens_seen": 107883680,
      "step": 5005,
      "time_per_iteration": 2.8014421463012695
    },
    {
      "auxiliary_loss_clip": 0.01170725,
      "auxiliary_loss_mlp": 0.01029156,
      "balance_loss_clip": 1.05591488,
      "balance_loss_mlp": 1.02083492,
      "epoch": 0.6019359105392894,
      "flos": 26469360518400.0,
      "grad_norm": 1.7769991681609174,
      "language_loss": 0.633201,
      "learning_rate": 1.444217395431066e-06,
      "loss": 0.65519983,
      "num_input_tokens_seen": 107906220,
      "step": 5006,
      "time_per_iteration": 3.745478868484497
    },
    {
      "auxiliary_loss_clip": 0.01069719,
      "auxiliary_loss_mlp": 0.01006446,
      "balance_loss_clip": 1.01082599,
      "balance_loss_mlp": 1.00545108,
      "epoch": 0.6020561534299285,
      "flos": 69190849728000.0,
      "grad_norm": 1.157645456242424,
      "language_loss": 0.55728823,
      "learning_rate": 1.4434691476844755e-06,
      "loss": 0.5780499,
      "num_input_tokens_seen": 107967195,
      "step": 5007,
      "time_per_iteration": 4.059453725814819
    },
    {
      "auxiliary_loss_clip": 0.01182858,
      "auxiliary_loss_mlp": 0.01030343,
      "balance_loss_clip": 1.05758119,
      "balance_loss_mlp": 1.02229095,
      "epoch": 0.6021763963205675,
      "flos": 21835304115840.0,
      "grad_norm": 5.063942793830602,
      "language_loss": 0.67212307,
      "learning_rate": 1.4427209843631729e-06,
      "loss": 0.69425511,
      "num_input_tokens_seen": 107984245,
      "step": 5008,
      "time_per_iteration": 3.6767051219940186
    },
    {
      "auxiliary_loss_clip": 0.01183633,
      "auxiliary_loss_mlp": 0.01028243,
      "balance_loss_clip": 1.05722308,
      "balance_loss_mlp": 1.01985669,
      "epoch": 0.6022966392112067,
      "flos": 26578636669440.0,
      "grad_norm": 1.9174808595617228,
      "language_loss": 0.81087929,
      "learning_rate": 1.4419729055806534e-06,
      "loss": 0.8329981,
      "num_input_tokens_seen": 108003680,
      "step": 5009,
      "time_per_iteration": 2.723377227783203
    },
    {
      "auxiliary_loss_clip": 0.01156385,
      "auxiliary_loss_mlp": 0.01029326,
      "balance_loss_clip": 1.05211854,
      "balance_loss_mlp": 1.02159846,
      "epoch": 0.6024168821018457,
      "flos": 20703722981760.0,
      "grad_norm": 1.8141955608771037,
      "language_loss": 0.82434464,
      "learning_rate": 1.441224911450401e-06,
      "loss": 0.84620178,
      "num_input_tokens_seen": 108019635,
      "step": 5010,
      "time_per_iteration": 2.7207186222076416
    },
    {
      "auxiliary_loss_clip": 0.01169508,
      "auxiliary_loss_mlp": 0.01036134,
      "balance_loss_clip": 1.05467904,
      "balance_loss_mlp": 1.0276041,
      "epoch": 0.6025371249924848,
      "flos": 24680973242880.0,
      "grad_norm": 1.643570356616967,
      "language_loss": 0.82431698,
      "learning_rate": 1.4404770020858851e-06,
      "loss": 0.84637332,
      "num_input_tokens_seen": 108039120,
      "step": 5011,
      "time_per_iteration": 2.7509076595306396
    },
    {
      "auxiliary_loss_clip": 0.01179147,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.0529151,
      "balance_loss_mlp": 1.01953626,
      "epoch": 0.602657367883124,
      "flos": 25955801815680.0,
      "grad_norm": 1.6233596798505516,
      "language_loss": 0.86049688,
      "learning_rate": 1.439729177600563e-06,
      "loss": 0.88256556,
      "num_input_tokens_seen": 108059615,
      "step": 5012,
      "time_per_iteration": 2.735398530960083
    },
    {
      "auxiliary_loss_clip": 0.01172793,
      "auxiliary_loss_mlp": 0.01035823,
      "balance_loss_clip": 1.05670106,
      "balance_loss_mlp": 1.02740681,
      "epoch": 0.602777610773763,
      "flos": 16690633925760.0,
      "grad_norm": 2.484109661617271,
      "language_loss": 0.72966051,
      "learning_rate": 1.4389814381078793e-06,
      "loss": 0.75174677,
      "num_input_tokens_seen": 108078855,
      "step": 5013,
      "time_per_iteration": 2.621655225753784
    },
    {
      "auxiliary_loss_clip": 0.01188314,
      "auxiliary_loss_mlp": 0.01033997,
      "balance_loss_clip": 1.05978799,
      "balance_loss_mlp": 1.02617073,
      "epoch": 0.6028978536644021,
      "flos": 13334243270400.0,
      "grad_norm": 2.2538699206082877,
      "language_loss": 0.80494678,
      "learning_rate": 1.438233783721265e-06,
      "loss": 0.82716984,
      "num_input_tokens_seen": 108095020,
      "step": 5014,
      "time_per_iteration": 2.5887598991394043
    },
    {
      "auxiliary_loss_clip": 0.01184788,
      "auxiliary_loss_mlp": 0.01034044,
      "balance_loss_clip": 1.0578022,
      "balance_loss_mlp": 1.02643299,
      "epoch": 0.6030180965550412,
      "flos": 19644825018240.0,
      "grad_norm": 2.5594188067845147,
      "language_loss": 0.77921498,
      "learning_rate": 1.43748621455414e-06,
      "loss": 0.80140334,
      "num_input_tokens_seen": 108111455,
      "step": 5015,
      "time_per_iteration": 2.5683040618896484
    },
    {
      "auxiliary_loss_clip": 0.01168163,
      "auxiliary_loss_mlp": 0.01030657,
      "balance_loss_clip": 1.05431199,
      "balance_loss_mlp": 1.02222347,
      "epoch": 0.6031383394456803,
      "flos": 14458390289280.0,
      "grad_norm": 2.7759280136728792,
      "language_loss": 0.81040168,
      "learning_rate": 1.4367387307199082e-06,
      "loss": 0.83238983,
      "num_input_tokens_seen": 108128305,
      "step": 5016,
      "time_per_iteration": 2.644742965698242
    },
    {
      "auxiliary_loss_clip": 0.01170576,
      "auxiliary_loss_mlp": 0.01026532,
      "balance_loss_clip": 1.05475998,
      "balance_loss_mlp": 1.01771629,
      "epoch": 0.6032585823363193,
      "flos": 13917791623680.0,
      "grad_norm": 1.877624823064509,
      "language_loss": 0.8253848,
      "learning_rate": 1.4359913323319632e-06,
      "loss": 0.84735584,
      "num_input_tokens_seen": 108145475,
      "step": 5017,
      "time_per_iteration": 2.630903959274292
    },
    {
      "auxiliary_loss_clip": 0.01170209,
      "auxiliary_loss_mlp": 0.01028942,
      "balance_loss_clip": 1.05226719,
      "balance_loss_mlp": 1.02023971,
      "epoch": 0.6033788252269584,
      "flos": 24353252530560.0,
      "grad_norm": 2.1464767956960706,
      "language_loss": 0.77900672,
      "learning_rate": 1.4352440195036847e-06,
      "loss": 0.80099821,
      "num_input_tokens_seen": 108165650,
      "step": 5018,
      "time_per_iteration": 3.613201856613159
    },
    {
      "auxiliary_loss_clip": 0.01162594,
      "auxiliary_loss_mlp": 0.01031299,
      "balance_loss_clip": 1.05165362,
      "balance_loss_mlp": 1.02300811,
      "epoch": 0.6034990681175976,
      "flos": 25521247077120.0,
      "grad_norm": 2.7573811093849674,
      "language_loss": 0.80132633,
      "learning_rate": 1.4344967923484395e-06,
      "loss": 0.82326519,
      "num_input_tokens_seen": 108187620,
      "step": 5019,
      "time_per_iteration": 2.689009189605713
    },
    {
      "auxiliary_loss_clip": 0.01184999,
      "auxiliary_loss_mlp": 0.01030422,
      "balance_loss_clip": 1.05530035,
      "balance_loss_mlp": 1.02173805,
      "epoch": 0.6036193110082366,
      "flos": 25958387594880.0,
      "grad_norm": 3.7344038664110224,
      "language_loss": 0.72329795,
      "learning_rate": 1.433749650979581e-06,
      "loss": 0.74545217,
      "num_input_tokens_seen": 108207605,
      "step": 5020,
      "time_per_iteration": 2.833907127380371
    },
    {
      "auxiliary_loss_clip": 0.01157066,
      "auxiliary_loss_mlp": 0.01028511,
      "balance_loss_clip": 1.05190182,
      "balance_loss_mlp": 1.01997566,
      "epoch": 0.6037395538988757,
      "flos": 25593427457280.0,
      "grad_norm": 2.3086244866068086,
      "language_loss": 0.6819585,
      "learning_rate": 1.433002595510451e-06,
      "loss": 0.70381427,
      "num_input_tokens_seen": 108226385,
      "step": 5021,
      "time_per_iteration": 2.7186760902404785
    },
    {
      "auxiliary_loss_clip": 0.01169531,
      "auxiliary_loss_mlp": 0.01032248,
      "balance_loss_clip": 1.05421591,
      "balance_loss_mlp": 1.02338481,
      "epoch": 0.6038597967895148,
      "flos": 17816253402240.0,
      "grad_norm": 2.0873552308630905,
      "language_loss": 0.7184062,
      "learning_rate": 1.4322556260543757e-06,
      "loss": 0.74042404,
      "num_input_tokens_seen": 108242960,
      "step": 5022,
      "time_per_iteration": 2.6599233150482178
    },
    {
      "auxiliary_loss_clip": 0.0107022,
      "auxiliary_loss_mlp": 0.0100218,
      "balance_loss_clip": 1.01164913,
      "balance_loss_mlp": 1.00119686,
      "epoch": 0.6039800396801539,
      "flos": 65169213235200.0,
      "grad_norm": 0.9493112986624318,
      "language_loss": 0.62654871,
      "learning_rate": 1.4315087427246703e-06,
      "loss": 0.64727277,
      "num_input_tokens_seen": 108296785,
      "step": 5023,
      "time_per_iteration": 3.122488021850586
    },
    {
      "auxiliary_loss_clip": 0.010697,
      "auxiliary_loss_mlp": 0.01003476,
      "balance_loss_clip": 1.01122928,
      "balance_loss_mlp": 1.0024451,
      "epoch": 0.604100282570793,
      "flos": 67386409073280.0,
      "grad_norm": 0.8714350694284797,
      "language_loss": 0.58365035,
      "learning_rate": 1.4307619456346372e-06,
      "loss": 0.60438216,
      "num_input_tokens_seen": 108341090,
      "step": 5024,
      "time_per_iteration": 2.9481844902038574
    },
    {
      "auxiliary_loss_clip": 0.01108171,
      "auxiliary_loss_mlp": 0.01032589,
      "balance_loss_clip": 1.04144168,
      "balance_loss_mlp": 1.02402961,
      "epoch": 0.6042205254614321,
      "flos": 35297495631360.0,
      "grad_norm": 2.291517975012678,
      "language_loss": 0.74440098,
      "learning_rate": 1.430015234897564e-06,
      "loss": 0.76580858,
      "num_input_tokens_seen": 108364370,
      "step": 5025,
      "time_per_iteration": 2.8374369144439697
    },
    {
      "auxiliary_loss_clip": 0.01170673,
      "auxiliary_loss_mlp": 0.0102824,
      "balance_loss_clip": 1.05148017,
      "balance_loss_mlp": 1.01938272,
      "epoch": 0.6043407683520712,
      "flos": 45658262206080.0,
      "grad_norm": 2.527907599029869,
      "language_loss": 0.66268671,
      "learning_rate": 1.4292686106267274e-06,
      "loss": 0.68467587,
      "num_input_tokens_seen": 108387220,
      "step": 5026,
      "time_per_iteration": 2.8635642528533936
    },
    {
      "auxiliary_loss_clip": 0.01187894,
      "auxiliary_loss_mlp": 0.01031983,
      "balance_loss_clip": 1.05785728,
      "balance_loss_mlp": 1.02377534,
      "epoch": 0.6044610112427102,
      "flos": 16180020138240.0,
      "grad_norm": 1.8296051566979512,
      "language_loss": 0.77447152,
      "learning_rate": 1.4285220729353876e-06,
      "loss": 0.79667026,
      "num_input_tokens_seen": 108405760,
      "step": 5027,
      "time_per_iteration": 2.679541826248169
    },
    {
      "auxiliary_loss_clip": 0.01183129,
      "auxiliary_loss_mlp": 0.01033501,
      "balance_loss_clip": 1.05635428,
      "balance_loss_mlp": 1.02467382,
      "epoch": 0.6045812541333494,
      "flos": 13804062186240.0,
      "grad_norm": 2.1395628515329146,
      "language_loss": 0.78101945,
      "learning_rate": 1.4277756219367957e-06,
      "loss": 0.80318576,
      "num_input_tokens_seen": 108422785,
      "step": 5028,
      "time_per_iteration": 2.591649293899536
    },
    {
      "auxiliary_loss_clip": 0.01186564,
      "auxiliary_loss_mlp": 0.01031043,
      "balance_loss_clip": 1.05670714,
      "balance_loss_mlp": 1.02243066,
      "epoch": 0.6047014970239885,
      "flos": 19975059682560.0,
      "grad_norm": 2.3683827439181755,
      "language_loss": 0.80024081,
      "learning_rate": 1.4270292577441864e-06,
      "loss": 0.8224169,
      "num_input_tokens_seen": 108442290,
      "step": 5029,
      "time_per_iteration": 2.6285693645477295
    },
    {
      "auxiliary_loss_clip": 0.01184375,
      "auxiliary_loss_mlp": 0.01031964,
      "balance_loss_clip": 1.05548453,
      "balance_loss_mlp": 1.02302301,
      "epoch": 0.6048217399146275,
      "flos": 25337097025920.0,
      "grad_norm": 1.5441353309053636,
      "language_loss": 0.7229836,
      "learning_rate": 1.4262829804707836e-06,
      "loss": 0.74514693,
      "num_input_tokens_seen": 108464280,
      "step": 5030,
      "time_per_iteration": 2.667088747024536
    },
    {
      "auxiliary_loss_clip": 0.01164894,
      "auxiliary_loss_mlp": 0.01030151,
      "balance_loss_clip": 1.0516181,
      "balance_loss_mlp": 1.02089441,
      "epoch": 0.6049419828052667,
      "flos": 26030819370240.0,
      "grad_norm": 1.4184611130723583,
      "language_loss": 0.69975913,
      "learning_rate": 1.4255367902297958e-06,
      "loss": 0.72170961,
      "num_input_tokens_seen": 108485610,
      "step": 5031,
      "time_per_iteration": 2.68791127204895
    },
    {
      "auxiliary_loss_clip": 0.01181754,
      "auxiliary_loss_mlp": 0.01034113,
      "balance_loss_clip": 1.05558097,
      "balance_loss_mlp": 1.02595901,
      "epoch": 0.6050622256959057,
      "flos": 14648106948480.0,
      "grad_norm": 2.3042577163395603,
      "language_loss": 0.7960155,
      "learning_rate": 1.4247906871344215e-06,
      "loss": 0.81817424,
      "num_input_tokens_seen": 108501005,
      "step": 5032,
      "time_per_iteration": 2.617284059524536
    },
    {
      "auxiliary_loss_clip": 0.01168288,
      "auxiliary_loss_mlp": 0.01026066,
      "balance_loss_clip": 1.05123639,
      "balance_loss_mlp": 1.01748908,
      "epoch": 0.6051824685865448,
      "flos": 23331450337920.0,
      "grad_norm": 2.4366833378899093,
      "language_loss": 0.75664061,
      "learning_rate": 1.4240446712978415e-06,
      "loss": 0.77858412,
      "num_input_tokens_seen": 108519990,
      "step": 5033,
      "time_per_iteration": 3.621671199798584
    },
    {
      "auxiliary_loss_clip": 0.01157092,
      "auxiliary_loss_mlp": 0.01039547,
      "balance_loss_clip": 1.05340791,
      "balance_loss_mlp": 1.03040969,
      "epoch": 0.605302711477184,
      "flos": 27563307177600.0,
      "grad_norm": 2.3498884152586945,
      "language_loss": 0.7423743,
      "learning_rate": 1.423298742833227e-06,
      "loss": 0.7643407,
      "num_input_tokens_seen": 108538650,
      "step": 5034,
      "time_per_iteration": 3.685621738433838
    },
    {
      "auxiliary_loss_clip": 0.0115483,
      "auxiliary_loss_mlp": 0.01031181,
      "balance_loss_clip": 1.05289507,
      "balance_loss_mlp": 1.02271688,
      "epoch": 0.605422954367823,
      "flos": 15154698412800.0,
      "grad_norm": 1.9322674712881371,
      "language_loss": 0.71662664,
      "learning_rate": 1.4225529018537352e-06,
      "loss": 0.73848677,
      "num_input_tokens_seen": 108554155,
      "step": 5035,
      "time_per_iteration": 3.6675922870635986
    },
    {
      "auxiliary_loss_clip": 0.01182311,
      "auxiliary_loss_mlp": 0.01037832,
      "balance_loss_clip": 1.05511451,
      "balance_loss_mlp": 1.02986884,
      "epoch": 0.6055431972584621,
      "flos": 27673912131840.0,
      "grad_norm": 1.5368910431393668,
      "language_loss": 0.77837938,
      "learning_rate": 1.4218071484725082e-06,
      "loss": 0.80058086,
      "num_input_tokens_seen": 108576275,
      "step": 5036,
      "time_per_iteration": 2.6620285511016846
    },
    {
      "auxiliary_loss_clip": 0.0113803,
      "auxiliary_loss_mlp": 0.01039712,
      "balance_loss_clip": 1.05079651,
      "balance_loss_mlp": 1.03080678,
      "epoch": 0.6056634401491012,
      "flos": 19387489006080.0,
      "grad_norm": 2.892554507327908,
      "language_loss": 0.76482832,
      "learning_rate": 1.4210614828026786e-06,
      "loss": 0.78660572,
      "num_input_tokens_seen": 108594125,
      "step": 5037,
      "time_per_iteration": 2.7551519870758057
    },
    {
      "auxiliary_loss_clip": 0.01158882,
      "auxiliary_loss_mlp": 0.01033158,
      "balance_loss_clip": 1.05093634,
      "balance_loss_mlp": 1.02504563,
      "epoch": 0.6057836830397403,
      "flos": 24789459294720.0,
      "grad_norm": 1.5919464041358742,
      "language_loss": 0.74755108,
      "learning_rate": 1.4203159049573605e-06,
      "loss": 0.76947153,
      "num_input_tokens_seen": 108615360,
      "step": 5038,
      "time_per_iteration": 2.623789072036743
    },
    {
      "auxiliary_loss_clip": 0.0118394,
      "auxiliary_loss_mlp": 0.01027767,
      "balance_loss_clip": 1.05507028,
      "balance_loss_mlp": 1.01890993,
      "epoch": 0.6059039259303793,
      "flos": 20558248899840.0,
      "grad_norm": 2.07633181754223,
      "language_loss": 0.87186444,
      "learning_rate": 1.4195704150496593e-06,
      "loss": 0.89398158,
      "num_input_tokens_seen": 108633075,
      "step": 5039,
      "time_per_iteration": 2.6483209133148193
    },
    {
      "auxiliary_loss_clip": 0.01174107,
      "auxiliary_loss_mlp": 0.01033993,
      "balance_loss_clip": 1.05540824,
      "balance_loss_mlp": 1.02552342,
      "epoch": 0.6060241688210185,
      "flos": 21069724613760.0,
      "grad_norm": 29.978843409412292,
      "language_loss": 0.73768592,
      "learning_rate": 1.4188250131926639e-06,
      "loss": 0.75976694,
      "num_input_tokens_seen": 108651875,
      "step": 5040,
      "time_per_iteration": 2.6382932662963867
    },
    {
      "auxiliary_loss_clip": 0.01186537,
      "auxiliary_loss_mlp": 0.01035577,
      "balance_loss_clip": 1.05570102,
      "balance_loss_mlp": 1.02708375,
      "epoch": 0.6061444117116576,
      "flos": 16361081619840.0,
      "grad_norm": 1.9908496008019374,
      "language_loss": 0.80737913,
      "learning_rate": 1.4180796994994525e-06,
      "loss": 0.82960027,
      "num_input_tokens_seen": 108669290,
      "step": 5041,
      "time_per_iteration": 2.6855039596557617
    },
    {
      "auxiliary_loss_clip": 0.01171077,
      "auxiliary_loss_mlp": 0.01030391,
      "balance_loss_clip": 1.0516789,
      "balance_loss_mlp": 1.0222013,
      "epoch": 0.6062646546022966,
      "flos": 21507296094720.0,
      "grad_norm": 2.0780982935550467,
      "language_loss": 0.72322619,
      "learning_rate": 1.4173344740830877e-06,
      "loss": 0.74524087,
      "num_input_tokens_seen": 108688420,
      "step": 5042,
      "time_per_iteration": 2.6655049324035645
    },
    {
      "auxiliary_loss_clip": 0.01187832,
      "auxiliary_loss_mlp": 0.01029971,
      "balance_loss_clip": 1.05901408,
      "balance_loss_mlp": 1.02160215,
      "epoch": 0.6063848974929358,
      "flos": 38983151283840.0,
      "grad_norm": 1.573053247882989,
      "language_loss": 0.70800471,
      "learning_rate": 1.4165893370566206e-06,
      "loss": 0.73018271,
      "num_input_tokens_seen": 108712175,
      "step": 5043,
      "time_per_iteration": 2.7524197101593018
    },
    {
      "auxiliary_loss_clip": 0.01184191,
      "auxiliary_loss_mlp": 0.01031928,
      "balance_loss_clip": 1.0551784,
      "balance_loss_mlp": 1.02334523,
      "epoch": 0.6065051403835748,
      "flos": 19646584784640.0,
      "grad_norm": 1.7200701377607797,
      "language_loss": 0.77955151,
      "learning_rate": 1.4158442885330865e-06,
      "loss": 0.80171263,
      "num_input_tokens_seen": 108730745,
      "step": 5044,
      "time_per_iteration": 3.6571388244628906
    },
    {
      "auxiliary_loss_clip": 0.0118451,
      "auxiliary_loss_mlp": 0.01030318,
      "balance_loss_clip": 1.05586362,
      "balance_loss_mlp": 1.02162194,
      "epoch": 0.6066253832742139,
      "flos": 23513086437120.0,
      "grad_norm": 1.9968727427193482,
      "language_loss": 0.79211044,
      "learning_rate": 1.4150993286255094e-06,
      "loss": 0.81425875,
      "num_input_tokens_seen": 108749995,
      "step": 5045,
      "time_per_iteration": 2.629703998565674
    },
    {
      "auxiliary_loss_clip": 0.01181135,
      "auxiliary_loss_mlp": 0.01026554,
      "balance_loss_clip": 1.05433035,
      "balance_loss_mlp": 1.01876974,
      "epoch": 0.6067456261648531,
      "flos": 19133708440320.0,
      "grad_norm": 2.4968290045779464,
      "language_loss": 0.800753,
      "learning_rate": 1.4143544574468993e-06,
      "loss": 0.8228299,
      "num_input_tokens_seen": 108768355,
      "step": 5046,
      "time_per_iteration": 2.6738390922546387
    },
    {
      "auxiliary_loss_clip": 0.01163997,
      "auxiliary_loss_mlp": 0.01032689,
      "balance_loss_clip": 1.05199242,
      "balance_loss_mlp": 1.02426136,
      "epoch": 0.6068658690554921,
      "flos": 20520614424960.0,
      "grad_norm": 2.0807478098476477,
      "language_loss": 0.82643795,
      "learning_rate": 1.4136096751102523e-06,
      "loss": 0.84840482,
      "num_input_tokens_seen": 108786685,
      "step": 5047,
      "time_per_iteration": 2.640016794204712
    },
    {
      "auxiliary_loss_clip": 0.0116474,
      "auxiliary_loss_mlp": 0.01028066,
      "balance_loss_clip": 1.05207872,
      "balance_loss_mlp": 1.01980495,
      "epoch": 0.6069861119461312,
      "flos": 27374560185600.0,
      "grad_norm": 2.0585443263790952,
      "language_loss": 0.83256698,
      "learning_rate": 1.4128649817285516e-06,
      "loss": 0.85449505,
      "num_input_tokens_seen": 108804820,
      "step": 5048,
      "time_per_iteration": 2.718752861022949
    },
    {
      "auxiliary_loss_clip": 0.01182967,
      "auxiliary_loss_mlp": 0.01036902,
      "balance_loss_clip": 1.05652094,
      "balance_loss_mlp": 1.02868891,
      "epoch": 0.6071063548367702,
      "flos": 25626500904960.0,
      "grad_norm": 7.374868257564917,
      "language_loss": 0.63288391,
      "learning_rate": 1.412120377414766e-06,
      "loss": 0.65508258,
      "num_input_tokens_seen": 108825010,
      "step": 5049,
      "time_per_iteration": 2.659851312637329
    },
    {
      "auxiliary_loss_clip": 0.01167302,
      "auxiliary_loss_mlp": 0.01029813,
      "balance_loss_clip": 1.05516863,
      "balance_loss_mlp": 1.02134967,
      "epoch": 0.6072265977274094,
      "flos": 24460517520000.0,
      "grad_norm": 2.0377694963588047,
      "language_loss": 0.71418685,
      "learning_rate": 1.4113758622818522e-06,
      "loss": 0.73615801,
      "num_input_tokens_seen": 108845075,
      "step": 5050,
      "time_per_iteration": 2.7040963172912598
    },
    {
      "auxiliary_loss_clip": 0.01167823,
      "auxiliary_loss_mlp": 0.01030608,
      "balance_loss_clip": 1.05293012,
      "balance_loss_mlp": 1.0224067,
      "epoch": 0.6073468406180484,
      "flos": 18149253413760.0,
      "grad_norm": 4.9316892863320305,
      "language_loss": 0.83281147,
      "learning_rate": 1.410631436442751e-06,
      "loss": 0.85479581,
      "num_input_tokens_seen": 108863870,
      "step": 5051,
      "time_per_iteration": 2.611189842224121
    },
    {
      "auxiliary_loss_clip": 0.01169507,
      "auxiliary_loss_mlp": 0.01030472,
      "balance_loss_clip": 1.05443215,
      "balance_loss_mlp": 1.02184129,
      "epoch": 0.6074670835086875,
      "flos": 20697617669760.0,
      "grad_norm": 2.025550624849886,
      "language_loss": 0.86760569,
      "learning_rate": 1.4098871000103936e-06,
      "loss": 0.88960552,
      "num_input_tokens_seen": 108882470,
      "step": 5052,
      "time_per_iteration": 2.626204490661621
    },
    {
      "auxiliary_loss_clip": 0.01166137,
      "auxiliary_loss_mlp": 0.01030093,
      "balance_loss_clip": 1.05380893,
      "balance_loss_mlp": 1.02190876,
      "epoch": 0.6075873263993267,
      "flos": 23769955572480.0,
      "grad_norm": 3.2012348783595646,
      "language_loss": 0.82515633,
      "learning_rate": 1.409142853097693e-06,
      "loss": 0.84711862,
      "num_input_tokens_seen": 108902710,
      "step": 5053,
      "time_per_iteration": 2.6703076362609863
    },
    {
      "auxiliary_loss_clip": 0.01185907,
      "auxiliary_loss_mlp": 0.01027938,
      "balance_loss_clip": 1.05666816,
      "balance_loss_mlp": 1.01964724,
      "epoch": 0.6077075692899657,
      "flos": 24454484035200.0,
      "grad_norm": 2.3479930000939406,
      "language_loss": 0.80008376,
      "learning_rate": 1.408398695817553e-06,
      "loss": 0.82222223,
      "num_input_tokens_seen": 108919935,
      "step": 5054,
      "time_per_iteration": 2.596787691116333
    },
    {
      "auxiliary_loss_clip": 0.01186603,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.05510569,
      "balance_loss_mlp": 1.01641798,
      "epoch": 0.6078278121806048,
      "flos": 27382102041600.0,
      "grad_norm": 2.4544179251446105,
      "language_loss": 0.70444691,
      "learning_rate": 1.4076546282828593e-06,
      "loss": 0.72657382,
      "num_input_tokens_seen": 108942790,
      "step": 5055,
      "time_per_iteration": 2.7096121311187744
    },
    {
      "auxiliary_loss_clip": 0.0115431,
      "auxiliary_loss_mlp": 0.01037458,
      "balance_loss_clip": 1.04884934,
      "balance_loss_mlp": 1.02861321,
      "epoch": 0.6079480550712439,
      "flos": 38436447306240.0,
      "grad_norm": 4.452235598184443,
      "language_loss": 0.65879416,
      "learning_rate": 1.4069106506064874e-06,
      "loss": 0.68071187,
      "num_input_tokens_seen": 108964215,
      "step": 5056,
      "time_per_iteration": 2.814908027648926
    },
    {
      "auxiliary_loss_clip": 0.01181966,
      "auxiliary_loss_mlp": 0.01033846,
      "balance_loss_clip": 1.05594754,
      "balance_loss_mlp": 1.02628195,
      "epoch": 0.608068297961883,
      "flos": 25336271013120.0,
      "grad_norm": 1.816554546772838,
      "language_loss": 0.78701973,
      "learning_rate": 1.4061667629012989e-06,
      "loss": 0.80917782,
      "num_input_tokens_seen": 108984885,
      "step": 5057,
      "time_per_iteration": 2.688389778137207
    },
    {
      "auxiliary_loss_clip": 0.01166883,
      "auxiliary_loss_mlp": 0.01031619,
      "balance_loss_clip": 1.0518775,
      "balance_loss_mlp": 1.02342629,
      "epoch": 0.608188540852522,
      "flos": 24202463235840.0,
      "grad_norm": 2.1540608201052125,
      "language_loss": 0.83309501,
      "learning_rate": 1.40542296528014e-06,
      "loss": 0.85508001,
      "num_input_tokens_seen": 109004545,
      "step": 5058,
      "time_per_iteration": 2.6417183876037598
    },
    {
      "auxiliary_loss_clip": 0.01176575,
      "auxiliary_loss_mlp": 0.01035128,
      "balance_loss_clip": 1.05568027,
      "balance_loss_mlp": 1.02660418,
      "epoch": 0.6083087837431612,
      "flos": 21284146851840.0,
      "grad_norm": 2.188969657804427,
      "language_loss": 0.76139355,
      "learning_rate": 1.4046792578558452e-06,
      "loss": 0.78351057,
      "num_input_tokens_seen": 109022440,
      "step": 5059,
      "time_per_iteration": 3.603344202041626
    },
    {
      "auxiliary_loss_clip": 0.01184004,
      "auxiliary_loss_mlp": 0.01031724,
      "balance_loss_clip": 1.05709732,
      "balance_loss_mlp": 1.02334404,
      "epoch": 0.6084290266338003,
      "flos": 16471435178880.0,
      "grad_norm": 2.932141095667198,
      "language_loss": 0.76056093,
      "learning_rate": 1.4039356407412325e-06,
      "loss": 0.78271818,
      "num_input_tokens_seen": 109035680,
      "step": 5060,
      "time_per_iteration": 4.672404766082764
    },
    {
      "auxiliary_loss_clip": 0.01058839,
      "auxiliary_loss_mlp": 0.01005234,
      "balance_loss_clip": 1.00954568,
      "balance_loss_mlp": 1.004179,
      "epoch": 0.6085492695244393,
      "flos": 66443574931200.0,
      "grad_norm": 0.7886741991788709,
      "language_loss": 0.57081699,
      "learning_rate": 1.40319211404911e-06,
      "loss": 0.59145772,
      "num_input_tokens_seen": 109090680,
      "step": 5061,
      "time_per_iteration": 3.199079751968384
    },
    {
      "auxiliary_loss_clip": 0.01181739,
      "auxiliary_loss_mlp": 0.01034144,
      "balance_loss_clip": 1.05416441,
      "balance_loss_mlp": 1.02629721,
      "epoch": 0.6086695124150785,
      "flos": 23618986709760.0,
      "grad_norm": 2.213064195843356,
      "language_loss": 0.90578854,
      "learning_rate": 1.4024486778922691e-06,
      "loss": 0.9279474,
      "num_input_tokens_seen": 109108995,
      "step": 5062,
      "time_per_iteration": 2.6463217735290527
    },
    {
      "auxiliary_loss_clip": 0.0118161,
      "auxiliary_loss_mlp": 0.01030778,
      "balance_loss_clip": 1.05400074,
      "balance_loss_mlp": 1.02233791,
      "epoch": 0.6087897553057176,
      "flos": 20157054917760.0,
      "grad_norm": 2.005604209169058,
      "language_loss": 0.77735221,
      "learning_rate": 1.4017053323834884e-06,
      "loss": 0.79947615,
      "num_input_tokens_seen": 109128825,
      "step": 5063,
      "time_per_iteration": 2.6727261543273926
    },
    {
      "auxiliary_loss_clip": 0.01147066,
      "auxiliary_loss_mlp": 0.01026685,
      "balance_loss_clip": 1.04760075,
      "balance_loss_mlp": 1.01855481,
      "epoch": 0.6089099981963566,
      "flos": 25482535194240.0,
      "grad_norm": 2.065173146165536,
      "language_loss": 0.76120681,
      "learning_rate": 1.4009620776355333e-06,
      "loss": 0.78294432,
      "num_input_tokens_seen": 109150425,
      "step": 5064,
      "time_per_iteration": 2.7241880893707275
    },
    {
      "auxiliary_loss_clip": 0.01149241,
      "auxiliary_loss_mlp": 0.01030432,
      "balance_loss_clip": 1.05245376,
      "balance_loss_mlp": 1.02226663,
      "epoch": 0.6090302410869958,
      "flos": 25332895134720.0,
      "grad_norm": 1.6717345328426014,
      "language_loss": 0.79409271,
      "learning_rate": 1.4002189137611553e-06,
      "loss": 0.81588948,
      "num_input_tokens_seen": 109169765,
      "step": 5065,
      "time_per_iteration": 2.7237019538879395
    },
    {
      "auxiliary_loss_clip": 0.01169122,
      "auxiliary_loss_mlp": 0.01030074,
      "balance_loss_clip": 1.05184555,
      "balance_loss_mlp": 1.02195621,
      "epoch": 0.6091504839776348,
      "flos": 23987358639360.0,
      "grad_norm": 1.7750919166774457,
      "language_loss": 0.69489348,
      "learning_rate": 1.3994758408730901e-06,
      "loss": 0.71688545,
      "num_input_tokens_seen": 109188950,
      "step": 5066,
      "time_per_iteration": 2.73829984664917
    },
    {
      "auxiliary_loss_clip": 0.01186342,
      "auxiliary_loss_mlp": 0.01030068,
      "balance_loss_clip": 1.05604792,
      "balance_loss_mlp": 1.02127004,
      "epoch": 0.6092707268682739,
      "flos": 29643037666560.0,
      "grad_norm": 2.182895320156567,
      "language_loss": 0.76846099,
      "learning_rate": 1.3987328590840629e-06,
      "loss": 0.7906251,
      "num_input_tokens_seen": 109209895,
      "step": 5067,
      "time_per_iteration": 2.7003607749938965
    },
    {
      "auxiliary_loss_clip": 0.01183085,
      "auxiliary_loss_mlp": 0.0102907,
      "balance_loss_clip": 1.05405068,
      "balance_loss_mlp": 1.0207907,
      "epoch": 0.609390969758913,
      "flos": 24024957200640.0,
      "grad_norm": 1.8660640775138821,
      "language_loss": 0.86005652,
      "learning_rate": 1.397989968506783e-06,
      "loss": 0.88217807,
      "num_input_tokens_seen": 109228905,
      "step": 5068,
      "time_per_iteration": 2.5975823402404785
    },
    {
      "auxiliary_loss_clip": 0.01180007,
      "auxiliary_loss_mlp": 0.01027183,
      "balance_loss_clip": 1.05813169,
      "balance_loss_mlp": 1.01886845,
      "epoch": 0.6095112126495521,
      "flos": 11102143288320.0,
      "grad_norm": 2.131669940074338,
      "language_loss": 0.72591364,
      "learning_rate": 1.3972471692539458e-06,
      "loss": 0.74798554,
      "num_input_tokens_seen": 109243620,
      "step": 5069,
      "time_per_iteration": 3.6255099773406982
    },
    {
      "auxiliary_loss_clip": 0.01184687,
      "auxiliary_loss_mlp": 0.01036123,
      "balance_loss_clip": 1.05675805,
      "balance_loss_mlp": 1.02818322,
      "epoch": 0.6096314555401912,
      "flos": 17265491187840.0,
      "grad_norm": 1.9441566551532155,
      "language_loss": 0.75884914,
      "learning_rate": 1.3965044614382348e-06,
      "loss": 0.78105724,
      "num_input_tokens_seen": 109259070,
      "step": 5070,
      "time_per_iteration": 2.62312650680542
    },
    {
      "auxiliary_loss_clip": 0.01173144,
      "auxiliary_loss_mlp": 0.01031813,
      "balance_loss_clip": 1.0550406,
      "balance_loss_mlp": 1.02240133,
      "epoch": 0.6097516984308303,
      "flos": 21645910679040.0,
      "grad_norm": 2.8873499395323403,
      "language_loss": 0.75438583,
      "learning_rate": 1.3957618451723162e-06,
      "loss": 0.77643538,
      "num_input_tokens_seen": 109275100,
      "step": 5071,
      "time_per_iteration": 2.6514813899993896
    },
    {
      "auxiliary_loss_clip": 0.01183154,
      "auxiliary_loss_mlp": 0.01029468,
      "balance_loss_clip": 1.05660558,
      "balance_loss_mlp": 1.02064049,
      "epoch": 0.6098719413214694,
      "flos": 27199208966400.0,
      "grad_norm": 3.659238789583596,
      "language_loss": 0.71592021,
      "learning_rate": 1.3950193205688457e-06,
      "loss": 0.73804641,
      "num_input_tokens_seen": 109294825,
      "step": 5072,
      "time_per_iteration": 2.701292037963867
    },
    {
      "auxiliary_loss_clip": 0.01169165,
      "auxiliary_loss_mlp": 0.01030761,
      "balance_loss_clip": 1.05464625,
      "balance_loss_mlp": 1.02304816,
      "epoch": 0.6099921842121084,
      "flos": 20412954385920.0,
      "grad_norm": 1.8315493847443878,
      "language_loss": 0.83981353,
      "learning_rate": 1.3942768877404627e-06,
      "loss": 0.86181283,
      "num_input_tokens_seen": 109313790,
      "step": 5073,
      "time_per_iteration": 2.6427226066589355
    },
    {
      "auxiliary_loss_clip": 0.01163067,
      "auxiliary_loss_mlp": 0.01030168,
      "balance_loss_clip": 1.05114794,
      "balance_loss_mlp": 1.0220263,
      "epoch": 0.6101124271027476,
      "flos": 23366139897600.0,
      "grad_norm": 2.250492196799083,
      "language_loss": 0.73954606,
      "learning_rate": 1.393534546799795e-06,
      "loss": 0.76147842,
      "num_input_tokens_seen": 109333490,
      "step": 5074,
      "time_per_iteration": 2.680690050125122
    },
    {
      "auxiliary_loss_clip": 0.011671,
      "auxiliary_loss_mlp": 0.01031099,
      "balance_loss_clip": 1.05600715,
      "balance_loss_mlp": 1.02273011,
      "epoch": 0.6102326699933867,
      "flos": 26687840993280.0,
      "grad_norm": 1.809189660000823,
      "language_loss": 0.67597055,
      "learning_rate": 1.3927922978594536e-06,
      "loss": 0.69795251,
      "num_input_tokens_seen": 109354575,
      "step": 5075,
      "time_per_iteration": 2.6973812580108643
    },
    {
      "auxiliary_loss_clip": 0.01068462,
      "auxiliary_loss_mlp": 0.01004413,
      "balance_loss_clip": 1.00998688,
      "balance_loss_mlp": 1.00332189,
      "epoch": 0.6103529128840257,
      "flos": 60644612551680.0,
      "grad_norm": 0.826412261081395,
      "language_loss": 0.57425427,
      "learning_rate": 1.3920501410320387e-06,
      "loss": 0.59498298,
      "num_input_tokens_seen": 109410690,
      "step": 5076,
      "time_per_iteration": 3.1833107471466064
    },
    {
      "auxiliary_loss_clip": 0.01172559,
      "auxiliary_loss_mlp": 0.01024568,
      "balance_loss_clip": 1.05228639,
      "balance_loss_mlp": 1.01581788,
      "epoch": 0.6104731557746649,
      "flos": 19021307806080.0,
      "grad_norm": 2.328542354301618,
      "language_loss": 0.76746988,
      "learning_rate": 1.3913080764301333e-06,
      "loss": 0.78944117,
      "num_input_tokens_seen": 109427650,
      "step": 5077,
      "time_per_iteration": 2.6885786056518555
    },
    {
      "auxiliary_loss_clip": 0.01125801,
      "auxiliary_loss_mlp": 0.01034635,
      "balance_loss_clip": 1.05241275,
      "balance_loss_mlp": 1.0254674,
      "epoch": 0.6105933986653039,
      "flos": 23366894083200.0,
      "grad_norm": 1.9813411262468343,
      "language_loss": 0.71572036,
      "learning_rate": 1.3905661041663085e-06,
      "loss": 0.73732471,
      "num_input_tokens_seen": 109448835,
      "step": 5078,
      "time_per_iteration": 2.726712465286255
    },
    {
      "auxiliary_loss_clip": 0.01147411,
      "auxiliary_loss_mlp": 0.01029376,
      "balance_loss_clip": 1.04980671,
      "balance_loss_mlp": 1.02022648,
      "epoch": 0.610713641555943,
      "flos": 34637565006720.0,
      "grad_norm": 2.003928583644579,
      "language_loss": 0.65411305,
      "learning_rate": 1.389824224353122e-06,
      "loss": 0.67588091,
      "num_input_tokens_seen": 109470425,
      "step": 5079,
      "time_per_iteration": 2.8238909244537354
    },
    {
      "auxiliary_loss_clip": 0.0118259,
      "auxiliary_loss_mlp": 0.01031867,
      "balance_loss_clip": 1.05612445,
      "balance_loss_mlp": 1.02297711,
      "epoch": 0.610833884446582,
      "flos": 26646471504000.0,
      "grad_norm": 1.7291888107850195,
      "language_loss": 0.76988721,
      "learning_rate": 1.389082437103115e-06,
      "loss": 0.79203176,
      "num_input_tokens_seen": 109489695,
      "step": 5080,
      "time_per_iteration": 2.6442134380340576
    },
    {
      "auxiliary_loss_clip": 0.0118472,
      "auxiliary_loss_mlp": 0.0103116,
      "balance_loss_clip": 1.05457473,
      "balance_loss_mlp": 1.02283359,
      "epoch": 0.6109541273372212,
      "flos": 21215126868480.0,
      "grad_norm": 6.879424646441445,
      "language_loss": 0.78346235,
      "learning_rate": 1.3883407425288172e-06,
      "loss": 0.80562115,
      "num_input_tokens_seen": 109510030,
      "step": 5081,
      "time_per_iteration": 2.646186113357544
    },
    {
      "auxiliary_loss_clip": 0.01180954,
      "auxiliary_loss_mlp": 0.01023942,
      "balance_loss_clip": 1.0533483,
      "balance_loss_mlp": 1.01492405,
      "epoch": 0.6110743702278603,
      "flos": 20084084438400.0,
      "grad_norm": 156.48605645566374,
      "language_loss": 0.79601038,
      "learning_rate": 1.3875991407427417e-06,
      "loss": 0.81805933,
      "num_input_tokens_seen": 109528255,
      "step": 5082,
      "time_per_iteration": 2.6713366508483887
    },
    {
      "auxiliary_loss_clip": 0.01057845,
      "auxiliary_loss_mlp": 0.01001874,
      "balance_loss_clip": 1.00997138,
      "balance_loss_mlp": 1.00086105,
      "epoch": 0.6111946131184993,
      "flos": 68302957438080.0,
      "grad_norm": 0.7685826723812141,
      "language_loss": 0.58128101,
      "learning_rate": 1.38685763185739e-06,
      "loss": 0.60187823,
      "num_input_tokens_seen": 109581915,
      "step": 5083,
      "time_per_iteration": 3.2705068588256836
    },
    {
      "auxiliary_loss_clip": 0.01165804,
      "auxiliary_loss_mlp": 0.01034225,
      "balance_loss_clip": 1.05454075,
      "balance_loss_mlp": 1.02586222,
      "epoch": 0.6113148560091385,
      "flos": 19937676602880.0,
      "grad_norm": 5.965977648063091,
      "language_loss": 0.67786157,
      "learning_rate": 1.3861162159852476e-06,
      "loss": 0.69986194,
      "num_input_tokens_seen": 109600050,
      "step": 5084,
      "time_per_iteration": 2.6005806922912598
    },
    {
      "auxiliary_loss_clip": 0.01171179,
      "auxiliary_loss_mlp": 0.0102887,
      "balance_loss_clip": 1.05561614,
      "balance_loss_mlp": 1.02049589,
      "epoch": 0.6114350988997775,
      "flos": 23731854220800.0,
      "grad_norm": 1.8547005295147734,
      "language_loss": 0.80207574,
      "learning_rate": 1.3853748932387875e-06,
      "loss": 0.82407624,
      "num_input_tokens_seen": 109620690,
      "step": 5085,
      "time_per_iteration": 3.593073606491089
    },
    {
      "auxiliary_loss_clip": 0.01153782,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.05172634,
      "balance_loss_mlp": 1.018448,
      "epoch": 0.6115553417904166,
      "flos": 24023700224640.0,
      "grad_norm": 2.681309494830352,
      "language_loss": 0.75699162,
      "learning_rate": 1.3846336637304671e-06,
      "loss": 0.77879846,
      "num_input_tokens_seen": 109638960,
      "step": 5086,
      "time_per_iteration": 2.730273485183716
    },
    {
      "auxiliary_loss_clip": 0.01166326,
      "auxiliary_loss_mlp": 0.01027227,
      "balance_loss_clip": 1.05371213,
      "balance_loss_mlp": 1.01857233,
      "epoch": 0.6116755846810558,
      "flos": 23733542160000.0,
      "grad_norm": 2.3209172432264,
      "language_loss": 0.83516526,
      "learning_rate": 1.3838925275727316e-06,
      "loss": 0.85710078,
      "num_input_tokens_seen": 109659700,
      "step": 5087,
      "time_per_iteration": 4.639319181442261
    },
    {
      "auxiliary_loss_clip": 0.01184333,
      "auxiliary_loss_mlp": 0.01031631,
      "balance_loss_clip": 1.05724239,
      "balance_loss_mlp": 1.02341187,
      "epoch": 0.6117958275716948,
      "flos": 18661626967680.0,
      "grad_norm": 2.289749125859352,
      "language_loss": 0.79265153,
      "learning_rate": 1.3831514848780089e-06,
      "loss": 0.81481111,
      "num_input_tokens_seen": 109679275,
      "step": 5088,
      "time_per_iteration": 2.6791274547576904
    },
    {
      "auxiliary_loss_clip": 0.01140367,
      "auxiliary_loss_mlp": 0.01027203,
      "balance_loss_clip": 1.0508548,
      "balance_loss_mlp": 1.01895952,
      "epoch": 0.6119160704623339,
      "flos": 16471183783680.0,
      "grad_norm": 3.04617699379088,
      "language_loss": 0.92143905,
      "learning_rate": 1.3824105357587152e-06,
      "loss": 0.94311476,
      "num_input_tokens_seen": 109696380,
      "step": 5089,
      "time_per_iteration": 2.820699691772461
    },
    {
      "auxiliary_loss_clip": 0.01166492,
      "auxiliary_loss_mlp": 0.01026051,
      "balance_loss_clip": 1.04992092,
      "balance_loss_mlp": 1.01868975,
      "epoch": 0.612036313352973,
      "flos": 23915465568000.0,
      "grad_norm": 1.5835676504548584,
      "language_loss": 0.82888699,
      "learning_rate": 1.381669680327253e-06,
      "loss": 0.85081238,
      "num_input_tokens_seen": 109718060,
      "step": 5090,
      "time_per_iteration": 2.6885340213775635
    },
    {
      "auxiliary_loss_clip": 0.0115842,
      "auxiliary_loss_mlp": 0.01030585,
      "balance_loss_clip": 1.05146432,
      "balance_loss_mlp": 1.02255058,
      "epoch": 0.6121565562436121,
      "flos": 26974766833920.0,
      "grad_norm": 3.6924550431474517,
      "language_loss": 0.71301585,
      "learning_rate": 1.380928918696008e-06,
      "loss": 0.7349059,
      "num_input_tokens_seen": 109736830,
      "step": 5091,
      "time_per_iteration": 2.725785493850708
    },
    {
      "auxiliary_loss_clip": 0.01163405,
      "auxiliary_loss_mlp": 0.01030976,
      "balance_loss_clip": 1.05350876,
      "balance_loss_mlp": 1.02205348,
      "epoch": 0.6122767991342511,
      "flos": 15668867646720.0,
      "grad_norm": 2.8058325644675475,
      "language_loss": 0.71703917,
      "learning_rate": 1.3801882509773548e-06,
      "loss": 0.73898298,
      "num_input_tokens_seen": 109754690,
      "step": 5092,
      "time_per_iteration": 2.6518898010253906
    },
    {
      "auxiliary_loss_clip": 0.01149178,
      "auxiliary_loss_mlp": 0.01026652,
      "balance_loss_clip": 1.04915774,
      "balance_loss_mlp": 1.01848638,
      "epoch": 0.6123970420248903,
      "flos": 27964321591680.0,
      "grad_norm": 1.772580406289133,
      "language_loss": 0.81853175,
      "learning_rate": 1.3794476772836503e-06,
      "loss": 0.84029007,
      "num_input_tokens_seen": 109775790,
      "step": 5093,
      "time_per_iteration": 2.833937168121338
    },
    {
      "auxiliary_loss_clip": 0.01169428,
      "auxiliary_loss_mlp": 0.01024953,
      "balance_loss_clip": 1.05367148,
      "balance_loss_mlp": 1.01617312,
      "epoch": 0.6125172849155294,
      "flos": 21468727866240.0,
      "grad_norm": 1.7864466259283311,
      "language_loss": 0.84822297,
      "learning_rate": 1.3787071977272402e-06,
      "loss": 0.87016684,
      "num_input_tokens_seen": 109795050,
      "step": 5094,
      "time_per_iteration": 2.6433637142181396
    },
    {
      "auxiliary_loss_clip": 0.01170135,
      "auxiliary_loss_mlp": 0.01029599,
      "balance_loss_clip": 1.05337238,
      "balance_loss_mlp": 1.02142084,
      "epoch": 0.6126375278061684,
      "flos": 16248321849600.0,
      "grad_norm": 3.2402905531984167,
      "language_loss": 0.72733259,
      "learning_rate": 1.3779668124204535e-06,
      "loss": 0.74932992,
      "num_input_tokens_seen": 109811465,
      "step": 5095,
      "time_per_iteration": 3.5971827507019043
    },
    {
      "auxiliary_loss_clip": 0.01176611,
      "auxiliary_loss_mlp": 0.01029012,
      "balance_loss_clip": 1.05289531,
      "balance_loss_mlp": 1.02115571,
      "epoch": 0.6127577706968076,
      "flos": 20448865008000.0,
      "grad_norm": 2.051544146947343,
      "language_loss": 0.81183898,
      "learning_rate": 1.3772265214756074e-06,
      "loss": 0.83389521,
      "num_input_tokens_seen": 109831225,
      "step": 5096,
      "time_per_iteration": 2.6395103931427
    },
    {
      "auxiliary_loss_clip": 0.01181875,
      "auxiliary_loss_mlp": 0.01025071,
      "balance_loss_clip": 1.05318809,
      "balance_loss_mlp": 1.01620805,
      "epoch": 0.6128780135874466,
      "flos": 18260397072000.0,
      "grad_norm": 2.7029428302259664,
      "language_loss": 0.75398719,
      "learning_rate": 1.3764863250050025e-06,
      "loss": 0.77605665,
      "num_input_tokens_seen": 109849465,
      "step": 5097,
      "time_per_iteration": 2.634547710418701
    },
    {
      "auxiliary_loss_clip": 0.01179883,
      "auxiliary_loss_mlp": 0.01031548,
      "balance_loss_clip": 1.05456471,
      "balance_loss_mlp": 1.02376032,
      "epoch": 0.6129982564780857,
      "flos": 24937088192640.0,
      "grad_norm": 1.9533280258882255,
      "language_loss": 0.80886805,
      "learning_rate": 1.3757462231209272e-06,
      "loss": 0.83098233,
      "num_input_tokens_seen": 109869770,
      "step": 5098,
      "time_per_iteration": 2.6908481121063232
    },
    {
      "auxiliary_loss_clip": 0.0117167,
      "auxiliary_loss_mlp": 0.01028557,
      "balance_loss_clip": 1.05317974,
      "balance_loss_mlp": 1.0196048,
      "epoch": 0.6131184993687249,
      "flos": 22492038430080.0,
      "grad_norm": 2.05126186738585,
      "language_loss": 0.88792855,
      "learning_rate": 1.3750062159356525e-06,
      "loss": 0.90993083,
      "num_input_tokens_seen": 109889120,
      "step": 5099,
      "time_per_iteration": 2.672349691390991
    },
    {
      "auxiliary_loss_clip": 0.01166979,
      "auxiliary_loss_mlp": 0.01024949,
      "balance_loss_clip": 1.05010557,
      "balance_loss_mlp": 1.01704597,
      "epoch": 0.6132387422593639,
      "flos": 15885839750400.0,
      "grad_norm": 1.8814440083312773,
      "language_loss": 0.8320688,
      "learning_rate": 1.3742663035614382e-06,
      "loss": 0.85398811,
      "num_input_tokens_seen": 109906490,
      "step": 5100,
      "time_per_iteration": 2.7044198513031006
    },
    {
      "auxiliary_loss_clip": 0.01181821,
      "auxiliary_loss_mlp": 0.01035446,
      "balance_loss_clip": 1.05408955,
      "balance_loss_mlp": 1.02674341,
      "epoch": 0.613358985150003,
      "flos": 25411539962880.0,
      "grad_norm": 2.75292137048986,
      "language_loss": 0.80268037,
      "learning_rate": 1.3735264861105283e-06,
      "loss": 0.82485306,
      "num_input_tokens_seen": 109927130,
      "step": 5101,
      "time_per_iteration": 2.6635794639587402
    },
    {
      "auxiliary_loss_clip": 0.01145035,
      "auxiliary_loss_mlp": 0.01031407,
      "balance_loss_clip": 1.0480938,
      "balance_loss_mlp": 1.02355075,
      "epoch": 0.6134792280406421,
      "flos": 21361283308800.0,
      "grad_norm": 2.028720791679057,
      "language_loss": 0.78957272,
      "learning_rate": 1.372786763695152e-06,
      "loss": 0.81133711,
      "num_input_tokens_seen": 109945890,
      "step": 5102,
      "time_per_iteration": 2.695117473602295
    },
    {
      "auxiliary_loss_clip": 0.01170902,
      "auxiliary_loss_mlp": 0.01031506,
      "balance_loss_clip": 1.05516469,
      "balance_loss_mlp": 1.02289307,
      "epoch": 0.6135994709312812,
      "flos": 21211248199680.0,
      "grad_norm": 1.7120767482433958,
      "language_loss": 0.77385867,
      "learning_rate": 1.3720471364275257e-06,
      "loss": 0.7958827,
      "num_input_tokens_seen": 109965535,
      "step": 5103,
      "time_per_iteration": 2.6203012466430664
    },
    {
      "auxiliary_loss_clip": 0.01169767,
      "auxiliary_loss_mlp": 0.01032737,
      "balance_loss_clip": 1.05369723,
      "balance_loss_mlp": 1.0245235,
      "epoch": 0.6137197138219203,
      "flos": 14794047907200.0,
      "grad_norm": 2.0263703906651096,
      "language_loss": 0.78180158,
      "learning_rate": 1.3713076044198486e-06,
      "loss": 0.80382663,
      "num_input_tokens_seen": 109982345,
      "step": 5104,
      "time_per_iteration": 2.708949327468872
    },
    {
      "auxiliary_loss_clip": 0.01159209,
      "auxiliary_loss_mlp": 0.01031476,
      "balance_loss_clip": 1.05001712,
      "balance_loss_mlp": 1.02296996,
      "epoch": 0.6138399567125594,
      "flos": 20084515401600.0,
      "grad_norm": 2.329075451949086,
      "language_loss": 0.80839133,
      "learning_rate": 1.3705681677843086e-06,
      "loss": 0.83029813,
      "num_input_tokens_seen": 110000940,
      "step": 5105,
      "time_per_iteration": 2.650890827178955
    },
    {
      "auxiliary_loss_clip": 0.01056922,
      "auxiliary_loss_mlp": 0.01002393,
      "balance_loss_clip": 1.01084995,
      "balance_loss_mlp": 1.00136757,
      "epoch": 0.6139601996031985,
      "flos": 60123838193280.0,
      "grad_norm": 0.7695247170693529,
      "language_loss": 0.60493183,
      "learning_rate": 1.3698288266330768e-06,
      "loss": 0.625525,
      "num_input_tokens_seen": 110061565,
      "step": 5106,
      "time_per_iteration": 3.297022819519043
    },
    {
      "auxiliary_loss_clip": 0.01181473,
      "auxiliary_loss_mlp": 0.01030591,
      "balance_loss_clip": 1.05621219,
      "balance_loss_mlp": 1.02288365,
      "epoch": 0.6140804424938375,
      "flos": 23586703361280.0,
      "grad_norm": 5.163406191280955,
      "language_loss": 0.72432959,
      "learning_rate": 1.3690895810783113e-06,
      "loss": 0.74645019,
      "num_input_tokens_seen": 110080360,
      "step": 5107,
      "time_per_iteration": 2.660722494125366
    },
    {
      "auxiliary_loss_clip": 0.01149121,
      "auxiliary_loss_mlp": 0.01027755,
      "balance_loss_clip": 1.0470103,
      "balance_loss_mlp": 1.01971388,
      "epoch": 0.6142006853844767,
      "flos": 21398199511680.0,
      "grad_norm": 2.5097145080639556,
      "language_loss": 0.71806467,
      "learning_rate": 1.3683504312321543e-06,
      "loss": 0.73983341,
      "num_input_tokens_seen": 110100695,
      "step": 5108,
      "time_per_iteration": 2.6633992195129395
    },
    {
      "auxiliary_loss_clip": 0.01151424,
      "auxiliary_loss_mlp": 0.01032872,
      "balance_loss_clip": 1.05278945,
      "balance_loss_mlp": 1.02480781,
      "epoch": 0.6143209282751158,
      "flos": 12057367622400.0,
      "grad_norm": 4.676068514866488,
      "language_loss": 0.80529869,
      "learning_rate": 1.3676113772067355e-06,
      "loss": 0.82714164,
      "num_input_tokens_seen": 110117750,
      "step": 5109,
      "time_per_iteration": 2.6553409099578857
    },
    {
      "auxiliary_loss_clip": 0.011659,
      "auxiliary_loss_mlp": 0.01035385,
      "balance_loss_clip": 1.05378366,
      "balance_loss_mlp": 1.0270462,
      "epoch": 0.6144411711657548,
      "flos": 25082274965760.0,
      "grad_norm": 2.236710680219997,
      "language_loss": 0.73026133,
      "learning_rate": 1.3668724191141671e-06,
      "loss": 0.75227416,
      "num_input_tokens_seen": 110137020,
      "step": 5110,
      "time_per_iteration": 2.68037486076355
    },
    {
      "auxiliary_loss_clip": 0.01184201,
      "auxiliary_loss_mlp": 0.01031484,
      "balance_loss_clip": 1.05747712,
      "balance_loss_mlp": 1.02318072,
      "epoch": 0.6145614140563939,
      "flos": 20114069316480.0,
      "grad_norm": 9.03934811641936,
      "language_loss": 0.66682756,
      "learning_rate": 1.3661335570665493e-06,
      "loss": 0.68898439,
      "num_input_tokens_seen": 110154930,
      "step": 5111,
      "time_per_iteration": 2.6789941787719727
    },
    {
      "auxiliary_loss_clip": 0.01183572,
      "auxiliary_loss_mlp": 0.01038993,
      "balance_loss_clip": 1.05569458,
      "balance_loss_mlp": 1.0299567,
      "epoch": 0.614681656947033,
      "flos": 16800376953600.0,
      "grad_norm": 3.293663269079689,
      "language_loss": 0.69728673,
      "learning_rate": 1.3653947911759676e-06,
      "loss": 0.7195124,
      "num_input_tokens_seen": 110172480,
      "step": 5112,
      "time_per_iteration": 5.489833116531372
    },
    {
      "auxiliary_loss_clip": 0.01172378,
      "auxiliary_loss_mlp": 0.01033203,
      "balance_loss_clip": 1.05286765,
      "balance_loss_mlp": 1.02516794,
      "epoch": 0.6148018998376721,
      "flos": 38801587011840.0,
      "grad_norm": 1.9741443028272116,
      "language_loss": 0.74885988,
      "learning_rate": 1.3646561215544904e-06,
      "loss": 0.77091575,
      "num_input_tokens_seen": 110197120,
      "step": 5113,
      "time_per_iteration": 2.7841954231262207
    },
    {
      "auxiliary_loss_clip": 0.01130796,
      "auxiliary_loss_mlp": 0.01028687,
      "balance_loss_clip": 1.04625273,
      "balance_loss_mlp": 1.02002013,
      "epoch": 0.6149221427283111,
      "flos": 23327032965120.0,
      "grad_norm": 2.395529870326532,
      "language_loss": 0.79372168,
      "learning_rate": 1.363917548314176e-06,
      "loss": 0.8153165,
      "num_input_tokens_seen": 110216385,
      "step": 5114,
      "time_per_iteration": 2.750429153442383
    },
    {
      "auxiliary_loss_clip": 0.01182204,
      "auxiliary_loss_mlp": 0.00762582,
      "balance_loss_clip": 1.05344689,
      "balance_loss_mlp": 1.00088191,
      "epoch": 0.6150423856189503,
      "flos": 22379494141440.0,
      "grad_norm": 2.2405189918933135,
      "language_loss": 0.73305804,
      "learning_rate": 1.3631790715670626e-06,
      "loss": 0.7525059,
      "num_input_tokens_seen": 110234790,
      "step": 5115,
      "time_per_iteration": 2.626767635345459
    },
    {
      "auxiliary_loss_clip": 0.01162952,
      "auxiliary_loss_mlp": 0.01035211,
      "balance_loss_clip": 1.0515902,
      "balance_loss_mlp": 1.02718234,
      "epoch": 0.6151626285095894,
      "flos": 18692078722560.0,
      "grad_norm": 4.763235721554086,
      "language_loss": 0.85621136,
      "learning_rate": 1.3624406914251783e-06,
      "loss": 0.87819296,
      "num_input_tokens_seen": 110251910,
      "step": 5116,
      "time_per_iteration": 2.669412136077881
    },
    {
      "auxiliary_loss_clip": 0.01179368,
      "auxiliary_loss_mlp": 0.01027255,
      "balance_loss_clip": 1.05488253,
      "balance_loss_mlp": 1.01932788,
      "epoch": 0.6152828714002284,
      "flos": 15851688894720.0,
      "grad_norm": 1.8748976680726124,
      "language_loss": 0.88255519,
      "learning_rate": 1.3617024080005335e-06,
      "loss": 0.90462136,
      "num_input_tokens_seen": 110268810,
      "step": 5117,
      "time_per_iteration": 2.586557388305664
    },
    {
      "auxiliary_loss_clip": 0.01179286,
      "auxiliary_loss_mlp": 0.01031443,
      "balance_loss_clip": 1.0525986,
      "balance_loss_mlp": 1.02314568,
      "epoch": 0.6154031142908676,
      "flos": 24869792062080.0,
      "grad_norm": 1.7150118400222805,
      "language_loss": 0.74353737,
      "learning_rate": 1.3609642214051266e-06,
      "loss": 0.76564467,
      "num_input_tokens_seen": 110293035,
      "step": 5118,
      "time_per_iteration": 2.737977981567383
    },
    {
      "auxiliary_loss_clip": 0.01152559,
      "auxiliary_loss_mlp": 0.01034023,
      "balance_loss_clip": 1.05396247,
      "balance_loss_mlp": 1.02597022,
      "epoch": 0.6155233571815066,
      "flos": 19244744357760.0,
      "grad_norm": 1.9078596292825758,
      "language_loss": 0.6641106,
      "learning_rate": 1.3602261317509385e-06,
      "loss": 0.68597639,
      "num_input_tokens_seen": 110309695,
      "step": 5119,
      "time_per_iteration": 2.7607064247131348
    },
    {
      "auxiliary_loss_clip": 0.0116761,
      "auxiliary_loss_mlp": 0.01030735,
      "balance_loss_clip": 1.05297232,
      "balance_loss_mlp": 1.02192509,
      "epoch": 0.6156436000721457,
      "flos": 18770077105920.0,
      "grad_norm": 2.498611986952097,
      "language_loss": 0.83305973,
      "learning_rate": 1.3594881391499387e-06,
      "loss": 0.85504317,
      "num_input_tokens_seen": 110328610,
      "step": 5120,
      "time_per_iteration": 2.615342140197754
    },
    {
      "auxiliary_loss_clip": 0.01151587,
      "auxiliary_loss_mlp": 0.01023903,
      "balance_loss_clip": 1.05207801,
      "balance_loss_mlp": 1.01593399,
      "epoch": 0.6157638429627849,
      "flos": 18041198325120.0,
      "grad_norm": 1.7312842644894324,
      "language_loss": 0.79456544,
      "learning_rate": 1.3587502437140778e-06,
      "loss": 0.81632036,
      "num_input_tokens_seen": 110346775,
      "step": 5121,
      "time_per_iteration": 3.652045965194702
    },
    {
      "auxiliary_loss_clip": 0.01169874,
      "auxiliary_loss_mlp": 0.01034505,
      "balance_loss_clip": 1.05209517,
      "balance_loss_mlp": 1.02633333,
      "epoch": 0.6158840858534239,
      "flos": 25556726736000.0,
      "grad_norm": 7.080095344553312,
      "language_loss": 0.8523584,
      "learning_rate": 1.3580124455552952e-06,
      "loss": 0.87440223,
      "num_input_tokens_seen": 110366140,
      "step": 5122,
      "time_per_iteration": 2.689767360687256
    },
    {
      "auxiliary_loss_clip": 0.01170422,
      "auxiliary_loss_mlp": 0.01028987,
      "balance_loss_clip": 1.05341947,
      "balance_loss_mlp": 1.02125621,
      "epoch": 0.616004328744063,
      "flos": 24640788902400.0,
      "grad_norm": 1.804209505992755,
      "language_loss": 0.87534058,
      "learning_rate": 1.3572747447855148e-06,
      "loss": 0.89733469,
      "num_input_tokens_seen": 110386550,
      "step": 5123,
      "time_per_iteration": 2.6880979537963867
    },
    {
      "auxiliary_loss_clip": 0.01182262,
      "auxiliary_loss_mlp": 0.01028794,
      "balance_loss_clip": 1.0549345,
      "balance_loss_mlp": 1.02011514,
      "epoch": 0.6161245716347021,
      "flos": 21689686379520.0,
      "grad_norm": 2.1419975151147663,
      "language_loss": 0.69319028,
      "learning_rate": 1.356537141516644e-06,
      "loss": 0.7153008,
      "num_input_tokens_seen": 110403970,
      "step": 5124,
      "time_per_iteration": 2.659597635269165
    },
    {
      "auxiliary_loss_clip": 0.01181046,
      "auxiliary_loss_mlp": 0.01025125,
      "balance_loss_clip": 1.05595565,
      "balance_loss_mlp": 1.01715636,
      "epoch": 0.6162448145253412,
      "flos": 35189225061120.0,
      "grad_norm": 2.56532794969349,
      "language_loss": 0.61826766,
      "learning_rate": 1.3557996358605775e-06,
      "loss": 0.64032936,
      "num_input_tokens_seen": 110423890,
      "step": 5125,
      "time_per_iteration": 2.7194390296936035
    },
    {
      "auxiliary_loss_clip": 0.01161972,
      "auxiliary_loss_mlp": 0.01029331,
      "balance_loss_clip": 1.05110002,
      "balance_loss_mlp": 1.02090907,
      "epoch": 0.6163650574159802,
      "flos": 21615279356160.0,
      "grad_norm": 2.0968374420658473,
      "language_loss": 0.70802456,
      "learning_rate": 1.3550622279291941e-06,
      "loss": 0.72993755,
      "num_input_tokens_seen": 110442035,
      "step": 5126,
      "time_per_iteration": 2.6578621864318848
    },
    {
      "auxiliary_loss_clip": 0.01166295,
      "auxiliary_loss_mlp": 0.01025196,
      "balance_loss_clip": 1.05457783,
      "balance_loss_mlp": 1.01731634,
      "epoch": 0.6164853003066194,
      "flos": 24572163968640.0,
      "grad_norm": 1.4336159657985224,
      "language_loss": 0.83502555,
      "learning_rate": 1.354324917834358e-06,
      "loss": 0.85694045,
      "num_input_tokens_seen": 110463280,
      "step": 5127,
      "time_per_iteration": 2.6856002807617188
    },
    {
      "auxiliary_loss_clip": 0.01177473,
      "auxiliary_loss_mlp": 0.01027719,
      "balance_loss_clip": 1.05180418,
      "balance_loss_mlp": 1.02021527,
      "epoch": 0.6166055431972585,
      "flos": 21835986474240.0,
      "grad_norm": 2.089864203107315,
      "language_loss": 0.77294946,
      "learning_rate": 1.353587705687918e-06,
      "loss": 0.79500139,
      "num_input_tokens_seen": 110481455,
      "step": 5128,
      "time_per_iteration": 2.6647145748138428
    },
    {
      "auxiliary_loss_clip": 0.01157905,
      "auxiliary_loss_mlp": 0.01029455,
      "balance_loss_clip": 1.05335069,
      "balance_loss_mlp": 1.02128315,
      "epoch": 0.6167257860878975,
      "flos": 17785262943360.0,
      "grad_norm": 2.408721961948649,
      "language_loss": 0.72575659,
      "learning_rate": 1.3528505916017096e-06,
      "loss": 0.74763018,
      "num_input_tokens_seen": 110499155,
      "step": 5129,
      "time_per_iteration": 2.6139705181121826
    },
    {
      "auxiliary_loss_clip": 0.01171549,
      "auxiliary_loss_mlp": 0.01026932,
      "balance_loss_clip": 1.05290437,
      "balance_loss_mlp": 1.01830721,
      "epoch": 0.6168460289785367,
      "flos": 23214811898880.0,
      "grad_norm": 2.800677774058534,
      "language_loss": 0.88693297,
      "learning_rate": 1.3521135756875514e-06,
      "loss": 0.90891778,
      "num_input_tokens_seen": 110515470,
      "step": 5130,
      "time_per_iteration": 2.698082685470581
    },
    {
      "auxiliary_loss_clip": 0.01148304,
      "auxiliary_loss_mlp": 0.01025262,
      "balance_loss_clip": 1.050699,
      "balance_loss_mlp": 1.01806736,
      "epoch": 0.6169662718691757,
      "flos": 26213281482240.0,
      "grad_norm": 1.6945214534008843,
      "language_loss": 0.86284631,
      "learning_rate": 1.3513766580572496e-06,
      "loss": 0.88458198,
      "num_input_tokens_seen": 110538290,
      "step": 5131,
      "time_per_iteration": 2.7309107780456543
    },
    {
      "auxiliary_loss_clip": 0.01176235,
      "auxiliary_loss_mlp": 0.01026804,
      "balance_loss_clip": 1.05195916,
      "balance_loss_mlp": 1.0195173,
      "epoch": 0.6170865147598148,
      "flos": 19026120228480.0,
      "grad_norm": 2.8559201038964432,
      "language_loss": 0.77124292,
      "learning_rate": 1.3506398388225924e-06,
      "loss": 0.79327333,
      "num_input_tokens_seen": 110555610,
      "step": 5132,
      "time_per_iteration": 2.688955545425415
    },
    {
      "auxiliary_loss_clip": 0.01179483,
      "auxiliary_loss_mlp": 0.01029877,
      "balance_loss_clip": 1.05389297,
      "balance_loss_mlp": 1.02180672,
      "epoch": 0.617206757650454,
      "flos": 18260361158400.0,
      "grad_norm": 2.4941013821352622,
      "language_loss": 0.72070843,
      "learning_rate": 1.349903118095355e-06,
      "loss": 0.74280202,
      "num_input_tokens_seen": 110574745,
      "step": 5133,
      "time_per_iteration": 2.619152784347534
    },
    {
      "auxiliary_loss_clip": 0.01181074,
      "auxiliary_loss_mlp": 0.01029439,
      "balance_loss_clip": 1.05534434,
      "balance_loss_mlp": 1.0218159,
      "epoch": 0.617327000541093,
      "flos": 18186959715840.0,
      "grad_norm": 2.491460913182146,
      "language_loss": 0.73739296,
      "learning_rate": 1.349166495987298e-06,
      "loss": 0.75949812,
      "num_input_tokens_seen": 110593310,
      "step": 5134,
      "time_per_iteration": 2.744075298309326
    },
    {
      "auxiliary_loss_clip": 0.01058441,
      "auxiliary_loss_mlp": 0.00752293,
      "balance_loss_clip": 1.00980604,
      "balance_loss_mlp": 1.00015533,
      "epoch": 0.6174472434317321,
      "flos": 61833796122240.0,
      "grad_norm": 0.8219619155360625,
      "language_loss": 0.60872191,
      "learning_rate": 1.348429972610166e-06,
      "loss": 0.62682927,
      "num_input_tokens_seen": 110657615,
      "step": 5135,
      "time_per_iteration": 3.3156704902648926
    },
    {
      "auxiliary_loss_clip": 0.01067189,
      "auxiliary_loss_mlp": 0.01000258,
      "balance_loss_clip": 1.00912511,
      "balance_loss_mlp": 0.99928015,
      "epoch": 0.6175674863223712,
      "flos": 71230970494080.0,
      "grad_norm": 0.8496998169377713,
      "language_loss": 0.57804084,
      "learning_rate": 1.3476935480756897e-06,
      "loss": 0.59871531,
      "num_input_tokens_seen": 110714365,
      "step": 5136,
      "time_per_iteration": 3.0518391132354736
    },
    {
      "auxiliary_loss_clip": 0.0118299,
      "auxiliary_loss_mlp": 0.01031803,
      "balance_loss_clip": 1.05569386,
      "balance_loss_mlp": 1.02331531,
      "epoch": 0.6176877292130103,
      "flos": 21835447770240.0,
      "grad_norm": 2.340839369586786,
      "language_loss": 0.75749409,
      "learning_rate": 1.346957222495583e-06,
      "loss": 0.77964199,
      "num_input_tokens_seen": 110732160,
      "step": 5137,
      "time_per_iteration": 3.5527632236480713
    },
    {
      "auxiliary_loss_clip": 0.01158378,
      "auxiliary_loss_mlp": 0.01029511,
      "balance_loss_clip": 1.05340505,
      "balance_loss_mlp": 1.02155352,
      "epoch": 0.6178079721036493,
      "flos": 17741738638080.0,
      "grad_norm": 2.8540169007480247,
      "language_loss": 0.71007335,
      "learning_rate": 1.3462209959815466e-06,
      "loss": 0.73195219,
      "num_input_tokens_seen": 110746900,
      "step": 5138,
      "time_per_iteration": 2.6966257095336914
    },
    {
      "auxiliary_loss_clip": 0.01164228,
      "auxiliary_loss_mlp": 0.0103035,
      "balance_loss_clip": 1.05295312,
      "balance_loss_mlp": 1.02217507,
      "epoch": 0.6179282149942885,
      "flos": 22633131052800.0,
      "grad_norm": 1.9547474874818989,
      "language_loss": 0.74392819,
      "learning_rate": 1.345484868645265e-06,
      "loss": 0.76587403,
      "num_input_tokens_seen": 110765710,
      "step": 5139,
      "time_per_iteration": 3.591825485229492
    },
    {
      "auxiliary_loss_clip": 0.0118006,
      "auxiliary_loss_mlp": 0.01025061,
      "balance_loss_clip": 1.05352449,
      "balance_loss_mlp": 1.0172832,
      "epoch": 0.6180484578849276,
      "flos": 22310330503680.0,
      "grad_norm": 2.245124743866712,
      "language_loss": 0.78859341,
      "learning_rate": 1.3447488405984088e-06,
      "loss": 0.81064463,
      "num_input_tokens_seen": 110783970,
      "step": 5140,
      "time_per_iteration": 2.696514129638672
    },
    {
      "auxiliary_loss_clip": 0.01181661,
      "auxiliary_loss_mlp": 0.01035117,
      "balance_loss_clip": 1.05430841,
      "balance_loss_mlp": 1.02685547,
      "epoch": 0.6181687007755666,
      "flos": 35225458905600.0,
      "grad_norm": 2.596132545407717,
      "language_loss": 0.7016139,
      "learning_rate": 1.3440129119526322e-06,
      "loss": 0.7237817,
      "num_input_tokens_seen": 110806395,
      "step": 5141,
      "time_per_iteration": 2.6989095211029053
    },
    {
      "auxiliary_loss_clip": 0.01067851,
      "auxiliary_loss_mlp": 0.01001432,
      "balance_loss_clip": 1.00952148,
      "balance_loss_mlp": 1.00041258,
      "epoch": 0.6182889436662057,
      "flos": 61547370094080.0,
      "grad_norm": 0.8116880522670531,
      "language_loss": 0.51117295,
      "learning_rate": 1.3432770828195762e-06,
      "loss": 0.53186578,
      "num_input_tokens_seen": 110867380,
      "step": 5142,
      "time_per_iteration": 3.4114677906036377
    },
    {
      "auxiliary_loss_clip": 0.01149497,
      "auxiliary_loss_mlp": 0.0103462,
      "balance_loss_clip": 1.05144215,
      "balance_loss_mlp": 1.02614379,
      "epoch": 0.6184091865568448,
      "flos": 19609991804160.0,
      "grad_norm": 2.306742202267275,
      "language_loss": 0.70618629,
      "learning_rate": 1.3425413533108635e-06,
      "loss": 0.72802746,
      "num_input_tokens_seen": 110885980,
      "step": 5143,
      "time_per_iteration": 2.628544807434082
    },
    {
      "auxiliary_loss_clip": 0.01164577,
      "auxiliary_loss_mlp": 0.01034172,
      "balance_loss_clip": 1.05563712,
      "balance_loss_mlp": 1.0260179,
      "epoch": 0.6185294294474839,
      "flos": 23586882929280.0,
      "grad_norm": 3.1781263540058,
      "language_loss": 0.70832968,
      "learning_rate": 1.341805723538105e-06,
      "loss": 0.73031723,
      "num_input_tokens_seen": 110906085,
      "step": 5144,
      "time_per_iteration": 2.7393887042999268
    },
    {
      "auxiliary_loss_clip": 0.01166171,
      "auxiliary_loss_mlp": 0.01023362,
      "balance_loss_clip": 1.05430543,
      "balance_loss_mlp": 1.01520264,
      "epoch": 0.618649672338123,
      "flos": 26762032535040.0,
      "grad_norm": 1.5858084706447104,
      "language_loss": 0.77589941,
      "learning_rate": 1.3410701936128948e-06,
      "loss": 0.7977947,
      "num_input_tokens_seen": 110928865,
      "step": 5145,
      "time_per_iteration": 2.827998638153076
    },
    {
      "auxiliary_loss_clip": 0.01180755,
      "auxiliary_loss_mlp": 0.01025873,
      "balance_loss_clip": 1.05517507,
      "balance_loss_mlp": 1.01781428,
      "epoch": 0.6187699152287621,
      "flos": 14456630522880.0,
      "grad_norm": 4.00691869388837,
      "language_loss": 0.85359788,
      "learning_rate": 1.340334763646812e-06,
      "loss": 0.87566411,
      "num_input_tokens_seen": 110943000,
      "step": 5146,
      "time_per_iteration": 2.5926477909088135
    },
    {
      "auxiliary_loss_clip": 0.01152672,
      "auxiliary_loss_mlp": 0.01030543,
      "balance_loss_clip": 1.05108905,
      "balance_loss_mlp": 1.02262402,
      "epoch": 0.6188901581194012,
      "flos": 20084766796800.0,
      "grad_norm": 1.9468730854120788,
      "language_loss": 0.74557209,
      "learning_rate": 1.3395994337514218e-06,
      "loss": 0.76740426,
      "num_input_tokens_seen": 110963170,
      "step": 5147,
      "time_per_iteration": 2.740541696548462
    },
    {
      "auxiliary_loss_clip": 0.01132606,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.04569817,
      "balance_loss_mlp": 1.02119303,
      "epoch": 0.6190104010100402,
      "flos": 25700728360320.0,
      "grad_norm": 1.7359720072746074,
      "language_loss": 0.78719461,
      "learning_rate": 1.3388642040382725e-06,
      "loss": 0.80881202,
      "num_input_tokens_seen": 110983595,
      "step": 5148,
      "time_per_iteration": 3.644094705581665
    },
    {
      "auxiliary_loss_clip": 0.01163751,
      "auxiliary_loss_mlp": 0.01029643,
      "balance_loss_clip": 1.05312586,
      "balance_loss_mlp": 1.02170932,
      "epoch": 0.6191306439006794,
      "flos": 30442372974720.0,
      "grad_norm": 1.6743675938322762,
      "language_loss": 0.84572107,
      "learning_rate": 1.3381290746188975e-06,
      "loss": 0.86765504,
      "num_input_tokens_seen": 111002965,
      "step": 5149,
      "time_per_iteration": 2.7215332984924316
    },
    {
      "auxiliary_loss_clip": 0.01173537,
      "auxiliary_loss_mlp": 0.0103336,
      "balance_loss_clip": 1.05650544,
      "balance_loss_mlp": 1.02522993,
      "epoch": 0.6192508867913185,
      "flos": 26685793918080.0,
      "grad_norm": 2.0273845069718783,
      "language_loss": 0.67390132,
      "learning_rate": 1.3373940456048152e-06,
      "loss": 0.6959703,
      "num_input_tokens_seen": 111022990,
      "step": 5150,
      "time_per_iteration": 2.692631721496582
    },
    {
      "auxiliary_loss_clip": 0.01179891,
      "auxiliary_loss_mlp": 0.01023931,
      "balance_loss_clip": 1.05431724,
      "balance_loss_mlp": 1.01603293,
      "epoch": 0.6193711296819575,
      "flos": 36722036090880.0,
      "grad_norm": 1.8433506300346802,
      "language_loss": 0.59720433,
      "learning_rate": 1.3366591171075299e-06,
      "loss": 0.61924255,
      "num_input_tokens_seen": 111046495,
      "step": 5151,
      "time_per_iteration": 2.735368490219116
    },
    {
      "auxiliary_loss_clip": 0.01179858,
      "auxiliary_loss_mlp": 0.01030268,
      "balance_loss_clip": 1.05422258,
      "balance_loss_mlp": 1.02193487,
      "epoch": 0.6194913725725967,
      "flos": 25192556697600.0,
      "grad_norm": 1.899570901554,
      "language_loss": 0.91012686,
      "learning_rate": 1.335924289238529e-06,
      "loss": 0.93222809,
      "num_input_tokens_seen": 111065705,
      "step": 5152,
      "time_per_iteration": 2.6385293006896973
    },
    {
      "auxiliary_loss_clip": 0.01165239,
      "auxiliary_loss_mlp": 0.01028631,
      "balance_loss_clip": 1.0547483,
      "balance_loss_mlp": 1.0207994,
      "epoch": 0.6196116154632357,
      "flos": 21178821196800.0,
      "grad_norm": 1.652384651967056,
      "language_loss": 0.76848567,
      "learning_rate": 1.3351895621092859e-06,
      "loss": 0.79042435,
      "num_input_tokens_seen": 111086050,
      "step": 5153,
      "time_per_iteration": 2.6574409008026123
    },
    {
      "auxiliary_loss_clip": 0.01177009,
      "auxiliary_loss_mlp": 0.01025806,
      "balance_loss_clip": 1.05355334,
      "balance_loss_mlp": 1.01787877,
      "epoch": 0.6197318583538748,
      "flos": 16253744803200.0,
      "grad_norm": 7.172994599092179,
      "language_loss": 0.76436931,
      "learning_rate": 1.3344549358312567e-06,
      "loss": 0.78639746,
      "num_input_tokens_seen": 111104450,
      "step": 5154,
      "time_per_iteration": 2.638472557067871
    },
    {
      "auxiliary_loss_clip": 0.01169241,
      "auxiliary_loss_mlp": 0.01032895,
      "balance_loss_clip": 1.05261672,
      "balance_loss_mlp": 1.02478874,
      "epoch": 0.619852101244514,
      "flos": 24425612478720.0,
      "grad_norm": 3.471818718496727,
      "language_loss": 0.78314602,
      "learning_rate": 1.3337204105158852e-06,
      "loss": 0.80516738,
      "num_input_tokens_seen": 111123320,
      "step": 5155,
      "time_per_iteration": 2.6720426082611084
    },
    {
      "auxiliary_loss_clip": 0.01148365,
      "auxiliary_loss_mlp": 0.01028181,
      "balance_loss_clip": 1.04921675,
      "balance_loss_mlp": 1.02016449,
      "epoch": 0.619972344135153,
      "flos": 16727298733440.0,
      "grad_norm": 2.128985527509844,
      "language_loss": 0.7296809,
      "learning_rate": 1.332985986274597e-06,
      "loss": 0.75144637,
      "num_input_tokens_seen": 111140950,
      "step": 5156,
      "time_per_iteration": 2.7879533767700195
    },
    {
      "auxiliary_loss_clip": 0.0117927,
      "auxiliary_loss_mlp": 0.01028403,
      "balance_loss_clip": 1.0564115,
      "balance_loss_mlp": 1.02139664,
      "epoch": 0.6200925870257921,
      "flos": 12495190498560.0,
      "grad_norm": 1.9561229718808173,
      "language_loss": 0.75171119,
      "learning_rate": 1.3322516632188047e-06,
      "loss": 0.77378786,
      "num_input_tokens_seen": 111157845,
      "step": 5157,
      "time_per_iteration": 2.601982593536377
    },
    {
      "auxiliary_loss_clip": 0.01154414,
      "auxiliary_loss_mlp": 0.0102968,
      "balance_loss_clip": 1.05106199,
      "balance_loss_mlp": 1.02183294,
      "epoch": 0.6202128299164312,
      "flos": 26539350168960.0,
      "grad_norm": 1.988629427916436,
      "language_loss": 0.67131346,
      "learning_rate": 1.3315174414599045e-06,
      "loss": 0.69315439,
      "num_input_tokens_seen": 111179165,
      "step": 5158,
      "time_per_iteration": 2.6988589763641357
    },
    {
      "auxiliary_loss_clip": 0.01167098,
      "auxiliary_loss_mlp": 0.01029019,
      "balance_loss_clip": 1.05058432,
      "balance_loss_mlp": 1.02118123,
      "epoch": 0.6203330728070703,
      "flos": 18770508069120.0,
      "grad_norm": 3.7392939473394686,
      "language_loss": 0.7578125,
      "learning_rate": 1.3307833211092768e-06,
      "loss": 0.77977365,
      "num_input_tokens_seen": 111197830,
      "step": 5159,
      "time_per_iteration": 2.5992143154144287
    },
    {
      "auxiliary_loss_clip": 0.01159683,
      "auxiliary_loss_mlp": 0.01033839,
      "balance_loss_clip": 1.05421221,
      "balance_loss_mlp": 1.02578092,
      "epoch": 0.6204533156977093,
      "flos": 20629782835200.0,
      "grad_norm": 1.6977391292215847,
      "language_loss": 0.75673366,
      "learning_rate": 1.3300493022782873e-06,
      "loss": 0.77866882,
      "num_input_tokens_seen": 111218400,
      "step": 5160,
      "time_per_iteration": 2.720677375793457
    },
    {
      "auxiliary_loss_clip": 0.01167253,
      "auxiliary_loss_mlp": 0.00762714,
      "balance_loss_clip": 1.05054283,
      "balance_loss_mlp": 1.000911,
      "epoch": 0.6205735585883485,
      "flos": 17348050598400.0,
      "grad_norm": 3.323644948132784,
      "language_loss": 0.72701681,
      "learning_rate": 1.3293153850782855e-06,
      "loss": 0.74631655,
      "num_input_tokens_seen": 111236720,
      "step": 5161,
      "time_per_iteration": 2.6293489933013916
    },
    {
      "auxiliary_loss_clip": 0.01183504,
      "auxiliary_loss_mlp": 0.01030174,
      "balance_loss_clip": 1.0544064,
      "balance_loss_mlp": 1.02152538,
      "epoch": 0.6206938014789876,
      "flos": 22965017742720.0,
      "grad_norm": 1.728527406793376,
      "language_loss": 0.71367931,
      "learning_rate": 1.3285815696206069e-06,
      "loss": 0.73581606,
      "num_input_tokens_seen": 111258265,
      "step": 5162,
      "time_per_iteration": 2.790384531021118
    },
    {
      "auxiliary_loss_clip": 0.01181358,
      "auxiliary_loss_mlp": 0.01026161,
      "balance_loss_clip": 1.05262446,
      "balance_loss_mlp": 1.01762605,
      "epoch": 0.6208140443696266,
      "flos": 23983192661760.0,
      "grad_norm": 2.082131363013137,
      "language_loss": 0.77120847,
      "learning_rate": 1.32784785601657e-06,
      "loss": 0.7932837,
      "num_input_tokens_seen": 111277675,
      "step": 5163,
      "time_per_iteration": 2.612699508666992
    },
    {
      "auxiliary_loss_clip": 0.01178641,
      "auxiliary_loss_mlp": 0.01029211,
      "balance_loss_clip": 1.05369699,
      "balance_loss_mlp": 1.02058673,
      "epoch": 0.6209342872602658,
      "flos": 35077291303680.0,
      "grad_norm": 1.7101500496459494,
      "language_loss": 0.73830783,
      "learning_rate": 1.3271142443774798e-06,
      "loss": 0.76038635,
      "num_input_tokens_seen": 111299910,
      "step": 5164,
      "time_per_iteration": 3.666594982147217
    },
    {
      "auxiliary_loss_clip": 0.01177969,
      "auxiliary_loss_mlp": 0.01027836,
      "balance_loss_clip": 1.05363142,
      "balance_loss_mlp": 1.02005744,
      "epoch": 0.6210545301509048,
      "flos": 26979327861120.0,
      "grad_norm": 1.8738326567999481,
      "language_loss": 0.81793308,
      "learning_rate": 1.3263807348146228e-06,
      "loss": 0.83999109,
      "num_input_tokens_seen": 111319765,
      "step": 5165,
      "time_per_iteration": 3.561316967010498
    },
    {
      "auxiliary_loss_clip": 0.01137975,
      "auxiliary_loss_mlp": 0.0102906,
      "balance_loss_clip": 1.04689229,
      "balance_loss_mlp": 1.02083492,
      "epoch": 0.6211747730415439,
      "flos": 33618240852480.0,
      "grad_norm": 2.199761922853238,
      "language_loss": 0.73615354,
      "learning_rate": 1.3256473274392733e-06,
      "loss": 0.75782388,
      "num_input_tokens_seen": 111341110,
      "step": 5166,
      "time_per_iteration": 2.778846025466919
    },
    {
      "auxiliary_loss_clip": 0.01181353,
      "auxiliary_loss_mlp": 0.01030642,
      "balance_loss_clip": 1.05583525,
      "balance_loss_mlp": 1.0224458,
      "epoch": 0.6212950159321831,
      "flos": 34167099646080.0,
      "grad_norm": 1.7510527358567,
      "language_loss": 0.70435542,
      "learning_rate": 1.3249140223626873e-06,
      "loss": 0.72647536,
      "num_input_tokens_seen": 111362730,
      "step": 5167,
      "time_per_iteration": 2.684755802154541
    },
    {
      "auxiliary_loss_clip": 0.01147828,
      "auxiliary_loss_mlp": 0.01027567,
      "balance_loss_clip": 1.0490334,
      "balance_loss_mlp": 1.01962185,
      "epoch": 0.6214152588228221,
      "flos": 27965758135680.0,
      "grad_norm": 1.5626602820276467,
      "language_loss": 0.75684476,
      "learning_rate": 1.3241808196961077e-06,
      "loss": 0.77859867,
      "num_input_tokens_seen": 111383855,
      "step": 5168,
      "time_per_iteration": 2.7364513874053955
    },
    {
      "auxiliary_loss_clip": 0.01174438,
      "auxiliary_loss_mlp": 0.01026841,
      "balance_loss_clip": 1.05243444,
      "balance_loss_mlp": 1.0198462,
      "epoch": 0.6215355017134612,
      "flos": 20230204965120.0,
      "grad_norm": 2.2985506501820616,
      "language_loss": 0.70833975,
      "learning_rate": 1.3234477195507608e-06,
      "loss": 0.73035252,
      "num_input_tokens_seen": 111402685,
      "step": 5169,
      "time_per_iteration": 2.6244895458221436
    },
    {
      "auxiliary_loss_clip": 0.01180872,
      "auxiliary_loss_mlp": 0.01029687,
      "balance_loss_clip": 1.05525851,
      "balance_loss_mlp": 1.02207565,
      "epoch": 0.6216557446041003,
      "flos": 41428129219200.0,
      "grad_norm": 2.0984956548020746,
      "language_loss": 0.62547219,
      "learning_rate": 1.322714722037857e-06,
      "loss": 0.64757776,
      "num_input_tokens_seen": 111424130,
      "step": 5170,
      "time_per_iteration": 2.763031482696533
    },
    {
      "auxiliary_loss_clip": 0.01171483,
      "auxiliary_loss_mlp": 0.01028907,
      "balance_loss_clip": 1.05260015,
      "balance_loss_mlp": 1.02086616,
      "epoch": 0.6217759874947394,
      "flos": 27928770105600.0,
      "grad_norm": 2.2621623674527087,
      "language_loss": 0.77685058,
      "learning_rate": 1.321981827268591e-06,
      "loss": 0.79885453,
      "num_input_tokens_seen": 111444785,
      "step": 5171,
      "time_per_iteration": 2.687647819519043
    },
    {
      "auxiliary_loss_clip": 0.01177752,
      "auxiliary_loss_mlp": 0.01027565,
      "balance_loss_clip": 1.05213785,
      "balance_loss_mlp": 1.02006698,
      "epoch": 0.6218962303853784,
      "flos": 21765673601280.0,
      "grad_norm": 1.7443840668240662,
      "language_loss": 0.81664729,
      "learning_rate": 1.3212490353541426e-06,
      "loss": 0.83870047,
      "num_input_tokens_seen": 111467045,
      "step": 5172,
      "time_per_iteration": 2.620314359664917
    },
    {
      "auxiliary_loss_clip": 0.01166963,
      "auxiliary_loss_mlp": 0.01031723,
      "balance_loss_clip": 1.05142093,
      "balance_loss_mlp": 1.02327752,
      "epoch": 0.6220164732760175,
      "flos": 21246260981760.0,
      "grad_norm": 1.985457449121283,
      "language_loss": 0.80720508,
      "learning_rate": 1.3205163464056762e-06,
      "loss": 0.82919198,
      "num_input_tokens_seen": 111483650,
      "step": 5173,
      "time_per_iteration": 2.6151859760284424
    },
    {
      "auxiliary_loss_clip": 0.01149149,
      "auxiliary_loss_mlp": 0.01028066,
      "balance_loss_clip": 1.04820991,
      "balance_loss_mlp": 1.0198946,
      "epoch": 0.6221367161666567,
      "flos": 26136360506880.0,
      "grad_norm": 2.690531576366629,
      "language_loss": 0.72793537,
      "learning_rate": 1.319783760534339e-06,
      "loss": 0.74970752,
      "num_input_tokens_seen": 111502895,
      "step": 5174,
      "time_per_iteration": 3.659036874771118
    },
    {
      "auxiliary_loss_clip": 0.01152749,
      "auxiliary_loss_mlp": 0.01026886,
      "balance_loss_clip": 1.0534184,
      "balance_loss_mlp": 1.01903641,
      "epoch": 0.6222569590572957,
      "flos": 16284196558080.0,
      "grad_norm": 3.054738274532835,
      "language_loss": 0.7545889,
      "learning_rate": 1.319051277851266e-06,
      "loss": 0.77638531,
      "num_input_tokens_seen": 111519180,
      "step": 5175,
      "time_per_iteration": 2.554499626159668
    },
    {
      "auxiliary_loss_clip": 0.0115978,
      "auxiliary_loss_mlp": 0.01029215,
      "balance_loss_clip": 1.04911292,
      "balance_loss_mlp": 1.02170467,
      "epoch": 0.6223772019479348,
      "flos": 18223840005120.0,
      "grad_norm": 2.0109344091366435,
      "language_loss": 0.84497029,
      "learning_rate": 1.3183188984675716e-06,
      "loss": 0.86686021,
      "num_input_tokens_seen": 111537545,
      "step": 5176,
      "time_per_iteration": 2.5372869968414307
    },
    {
      "auxiliary_loss_clip": 0.01142799,
      "auxiliary_loss_mlp": 0.0103031,
      "balance_loss_clip": 1.04877305,
      "balance_loss_mlp": 1.02250767,
      "epoch": 0.6224974448385739,
      "flos": 27489797994240.0,
      "grad_norm": 2.576548309500183,
      "language_loss": 0.71422523,
      "learning_rate": 1.3175866224943586e-06,
      "loss": 0.73595631,
      "num_input_tokens_seen": 111556265,
      "step": 5177,
      "time_per_iteration": 2.568159341812134
    },
    {
      "auxiliary_loss_clip": 0.01181696,
      "auxiliary_loss_mlp": 0.01028341,
      "balance_loss_clip": 1.05426526,
      "balance_loss_mlp": 1.0199132,
      "epoch": 0.622617687729213,
      "flos": 19791951125760.0,
      "grad_norm": 2.691315121665777,
      "language_loss": 0.73286867,
      "learning_rate": 1.316854450042712e-06,
      "loss": 0.75496906,
      "num_input_tokens_seen": 111574205,
      "step": 5178,
      "time_per_iteration": 2.55387282371521
    },
    {
      "auxiliary_loss_clip": 0.0113453,
      "auxiliary_loss_mlp": 0.01031346,
      "balance_loss_clip": 1.04680097,
      "balance_loss_mlp": 1.02282238,
      "epoch": 0.622737930619852,
      "flos": 23038886062080.0,
      "grad_norm": 2.737635844261491,
      "language_loss": 0.74728268,
      "learning_rate": 1.3161223812237024e-06,
      "loss": 0.76894146,
      "num_input_tokens_seen": 111593560,
      "step": 5179,
      "time_per_iteration": 2.6686999797821045
    },
    {
      "auxiliary_loss_clip": 0.01163884,
      "auxiliary_loss_mlp": 0.01029325,
      "balance_loss_clip": 1.05137491,
      "balance_loss_mlp": 1.02117705,
      "epoch": 0.6228581735104912,
      "flos": 12634271959680.0,
      "grad_norm": 4.508281021540286,
      "language_loss": 0.85646051,
      "learning_rate": 1.3153904161483842e-06,
      "loss": 0.87839258,
      "num_input_tokens_seen": 111608860,
      "step": 5180,
      "time_per_iteration": 2.556626319885254
    },
    {
      "auxiliary_loss_clip": 0.01182375,
      "auxiliary_loss_mlp": 0.01028813,
      "balance_loss_clip": 1.05435777,
      "balance_loss_mlp": 1.02052188,
      "epoch": 0.6229784164011303,
      "flos": 23802813538560.0,
      "grad_norm": 2.1261417331852206,
      "language_loss": 0.85282493,
      "learning_rate": 1.3146585549277953e-06,
      "loss": 0.87493688,
      "num_input_tokens_seen": 111627500,
      "step": 5181,
      "time_per_iteration": 2.581618547439575
    },
    {
      "auxiliary_loss_clip": 0.01182205,
      "auxiliary_loss_mlp": 0.01028441,
      "balance_loss_clip": 1.05544972,
      "balance_loss_mlp": 1.02050757,
      "epoch": 0.6230986592917693,
      "flos": 22414219614720.0,
      "grad_norm": 2.2250471342155147,
      "language_loss": 0.78510529,
      "learning_rate": 1.3139267976729591e-06,
      "loss": 0.80721176,
      "num_input_tokens_seen": 111647690,
      "step": 5182,
      "time_per_iteration": 2.6198787689208984
    },
    {
      "auxiliary_loss_clip": 0.01168629,
      "auxiliary_loss_mlp": 0.01026593,
      "balance_loss_clip": 1.05190349,
      "balance_loss_mlp": 1.01877928,
      "epoch": 0.6232189021824085,
      "flos": 34528217028480.0,
      "grad_norm": 2.5598817653897648,
      "language_loss": 0.71875924,
      "learning_rate": 1.3131951444948815e-06,
      "loss": 0.74071145,
      "num_input_tokens_seen": 111667090,
      "step": 5183,
      "time_per_iteration": 2.7013885974884033
    },
    {
      "auxiliary_loss_clip": 0.01167976,
      "auxiliary_loss_mlp": 0.01025282,
      "balance_loss_clip": 1.05262208,
      "balance_loss_mlp": 1.01743174,
      "epoch": 0.6233391450730476,
      "flos": 22237000888320.0,
      "grad_norm": 2.195343801536805,
      "language_loss": 0.76793861,
      "learning_rate": 1.3124635955045546e-06,
      "loss": 0.78987122,
      "num_input_tokens_seen": 111686905,
      "step": 5184,
      "time_per_iteration": 2.644277572631836
    },
    {
      "auxiliary_loss_clip": 0.01164432,
      "auxiliary_loss_mlp": 0.0102887,
      "balance_loss_clip": 1.05378294,
      "balance_loss_mlp": 1.02096093,
      "epoch": 0.6234593879636866,
      "flos": 20332693445760.0,
      "grad_norm": 2.781256305495995,
      "language_loss": 0.84574413,
      "learning_rate": 1.3117321508129537e-06,
      "loss": 0.86767721,
      "num_input_tokens_seen": 111704985,
      "step": 5185,
      "time_per_iteration": 2.576003074645996
    },
    {
      "auxiliary_loss_clip": 0.01180808,
      "auxiliary_loss_mlp": 0.01033057,
      "balance_loss_clip": 1.05522466,
      "balance_loss_mlp": 1.02498019,
      "epoch": 0.6235796308543258,
      "flos": 20664903358080.0,
      "grad_norm": 1.6736630387881213,
      "language_loss": 0.76492238,
      "learning_rate": 1.3110008105310388e-06,
      "loss": 0.78706104,
      "num_input_tokens_seen": 111724805,
      "step": 5186,
      "time_per_iteration": 2.505511999130249
    },
    {
      "auxiliary_loss_clip": 0.01116682,
      "auxiliary_loss_mlp": 0.01026411,
      "balance_loss_clip": 1.04710627,
      "balance_loss_mlp": 1.01785207,
      "epoch": 0.6236998737449648,
      "flos": 26618641441920.0,
      "grad_norm": 2.2234451400795536,
      "language_loss": 0.78105241,
      "learning_rate": 1.3102695747697526e-06,
      "loss": 0.80248338,
      "num_input_tokens_seen": 111747675,
      "step": 5187,
      "time_per_iteration": 2.772880792617798
    },
    {
      "auxiliary_loss_clip": 0.01159076,
      "auxiliary_loss_mlp": 0.01037604,
      "balance_loss_clip": 1.04998314,
      "balance_loss_mlp": 1.02933669,
      "epoch": 0.6238201166356039,
      "flos": 12674599954560.0,
      "grad_norm": 2.6150340044808833,
      "language_loss": 0.90806502,
      "learning_rate": 1.3095384436400237e-06,
      "loss": 0.93003178,
      "num_input_tokens_seen": 111759205,
      "step": 5188,
      "time_per_iteration": 2.585071325302124
    },
    {
      "auxiliary_loss_clip": 0.01181052,
      "auxiliary_loss_mlp": 0.01036101,
      "balance_loss_clip": 1.05395174,
      "balance_loss_mlp": 1.02820599,
      "epoch": 0.623940359526243,
      "flos": 10452160730880.0,
      "grad_norm": 2.9932173243547795,
      "language_loss": 0.82352209,
      "learning_rate": 1.3088074172527633e-06,
      "loss": 0.84569359,
      "num_input_tokens_seen": 111776335,
      "step": 5189,
      "time_per_iteration": 2.571542501449585
    },
    {
      "auxiliary_loss_clip": 0.01168591,
      "auxiliary_loss_mlp": 0.0102844,
      "balance_loss_clip": 1.04991138,
      "balance_loss_mlp": 1.01977384,
      "epoch": 0.6240606024168821,
      "flos": 29059525226880.0,
      "grad_norm": 2.779695177809754,
      "language_loss": 0.71615517,
      "learning_rate": 1.3080764957188684e-06,
      "loss": 0.73812544,
      "num_input_tokens_seen": 111796580,
      "step": 5190,
      "time_per_iteration": 3.66239333152771
    },
    {
      "auxiliary_loss_clip": 0.0118292,
      "auxiliary_loss_mlp": 0.01031435,
      "balance_loss_clip": 1.05414438,
      "balance_loss_mlp": 1.02379966,
      "epoch": 0.6241808453075212,
      "flos": 22018089450240.0,
      "grad_norm": 1.944422876755998,
      "language_loss": 0.71130502,
      "learning_rate": 1.3073456791492192e-06,
      "loss": 0.73344862,
      "num_input_tokens_seen": 111816290,
      "step": 5191,
      "time_per_iteration": 4.3933727741241455
    },
    {
      "auxiliary_loss_clip": 0.01177468,
      "auxiliary_loss_mlp": 0.01028769,
      "balance_loss_clip": 1.0522939,
      "balance_loss_mlp": 1.02075863,
      "epoch": 0.6243010881981603,
      "flos": 21138708683520.0,
      "grad_norm": 1.941064107555537,
      "language_loss": 0.78352314,
      "learning_rate": 1.3066149676546801e-06,
      "loss": 0.8055855,
      "num_input_tokens_seen": 111834470,
      "step": 5192,
      "time_per_iteration": 2.674586772918701
    },
    {
      "auxiliary_loss_clip": 0.01179856,
      "auxiliary_loss_mlp": 0.01033187,
      "balance_loss_clip": 1.05484831,
      "balance_loss_mlp": 1.02499104,
      "epoch": 0.6244213310887994,
      "flos": 22344948236160.0,
      "grad_norm": 1.7138361651805856,
      "language_loss": 0.66280508,
      "learning_rate": 1.3058843613460985e-06,
      "loss": 0.68493551,
      "num_input_tokens_seen": 111852410,
      "step": 5193,
      "time_per_iteration": 2.608544111251831
    },
    {
      "auxiliary_loss_clip": 0.01164943,
      "auxiliary_loss_mlp": 0.01031139,
      "balance_loss_clip": 1.05241096,
      "balance_loss_mlp": 1.0226984,
      "epoch": 0.6245415739794384,
      "flos": 15231978524160.0,
      "grad_norm": 2.029382879901601,
      "language_loss": 0.74497235,
      "learning_rate": 1.3051538603343075e-06,
      "loss": 0.7669332,
      "num_input_tokens_seen": 111870340,
      "step": 5194,
      "time_per_iteration": 2.6453211307525635
    },
    {
      "auxiliary_loss_clip": 0.01147335,
      "auxiliary_loss_mlp": 0.01029409,
      "balance_loss_clip": 1.05212855,
      "balance_loss_mlp": 1.02146959,
      "epoch": 0.6246618168700776,
      "flos": 18879891960960.0,
      "grad_norm": 2.068608224340944,
      "language_loss": 0.67937475,
      "learning_rate": 1.3044234647301235e-06,
      "loss": 0.70114219,
      "num_input_tokens_seen": 111888365,
      "step": 5195,
      "time_per_iteration": 2.6240131855010986
    },
    {
      "auxiliary_loss_clip": 0.01162742,
      "auxiliary_loss_mlp": 0.01025067,
      "balance_loss_clip": 1.05151796,
      "balance_loss_mlp": 1.0174017,
      "epoch": 0.6247820597607167,
      "flos": 14319201087360.0,
      "grad_norm": 1.8220988723657943,
      "language_loss": 0.72655445,
      "learning_rate": 1.303693174644347e-06,
      "loss": 0.74843258,
      "num_input_tokens_seen": 111905840,
      "step": 5196,
      "time_per_iteration": 2.663182020187378
    },
    {
      "auxiliary_loss_clip": 0.01181953,
      "auxiliary_loss_mlp": 0.01025705,
      "balance_loss_clip": 1.05384028,
      "balance_loss_mlp": 1.01800978,
      "epoch": 0.6249023026513557,
      "flos": 22637979388800.0,
      "grad_norm": 1.922356274541232,
      "language_loss": 0.80553353,
      "learning_rate": 1.3029629901877625e-06,
      "loss": 0.82761014,
      "num_input_tokens_seen": 111925215,
      "step": 5197,
      "time_per_iteration": 2.589386463165283
    },
    {
      "auxiliary_loss_clip": 0.01168184,
      "auxiliary_loss_mlp": 0.0102418,
      "balance_loss_clip": 1.05491757,
      "balance_loss_mlp": 1.01517963,
      "epoch": 0.6250225455419949,
      "flos": 20266690204800.0,
      "grad_norm": 2.686747500398147,
      "language_loss": 0.77661693,
      "learning_rate": 1.3022329114711376e-06,
      "loss": 0.79854059,
      "num_input_tokens_seen": 111943925,
      "step": 5198,
      "time_per_iteration": 2.635976791381836
    },
    {
      "auxiliary_loss_clip": 0.01164867,
      "auxiliary_loss_mlp": 0.01035202,
      "balance_loss_clip": 1.05424261,
      "balance_loss_mlp": 1.02667272,
      "epoch": 0.6251427884326339,
      "flos": 23437853400960.0,
      "grad_norm": 2.5602878804890326,
      "language_loss": 0.69831008,
      "learning_rate": 1.3015029386052256e-06,
      "loss": 0.72031081,
      "num_input_tokens_seen": 111964095,
      "step": 5199,
      "time_per_iteration": 2.683497905731201
    },
    {
      "auxiliary_loss_clip": 0.01182438,
      "auxiliary_loss_mlp": 0.01022948,
      "balance_loss_clip": 1.05412078,
      "balance_loss_mlp": 1.01527357,
      "epoch": 0.625263031323273,
      "flos": 31723055464320.0,
      "grad_norm": 1.9780586947051815,
      "language_loss": 0.73196816,
      "learning_rate": 1.3007730717007622e-06,
      "loss": 0.75402206,
      "num_input_tokens_seen": 111984910,
      "step": 5200,
      "time_per_iteration": 3.6187124252319336
    },
    {
      "auxiliary_loss_clip": 0.01161473,
      "auxiliary_loss_mlp": 0.00762809,
      "balance_loss_clip": 1.05046296,
      "balance_loss_mlp": 1.00072145,
      "epoch": 0.6253832742139122,
      "flos": 24134341092480.0,
      "grad_norm": 2.523270103511466,
      "language_loss": 0.75621414,
      "learning_rate": 1.3000433108684676e-06,
      "loss": 0.77545696,
      "num_input_tokens_seen": 112005410,
      "step": 5201,
      "time_per_iteration": 2.72145414352417
    },
    {
      "auxiliary_loss_clip": 0.01178982,
      "auxiliary_loss_mlp": 0.0102884,
      "balance_loss_clip": 1.05294943,
      "balance_loss_mlp": 1.02132964,
      "epoch": 0.6255035171045512,
      "flos": 27668812400640.0,
      "grad_norm": 2.4391911889487425,
      "language_loss": 0.80516118,
      "learning_rate": 1.2993136562190467e-06,
      "loss": 0.82723939,
      "num_input_tokens_seen": 112024530,
      "step": 5202,
      "time_per_iteration": 2.6005146503448486
    },
    {
      "auxiliary_loss_clip": 0.01152937,
      "auxiliary_loss_mlp": 0.01029025,
      "balance_loss_clip": 1.05091536,
      "balance_loss_mlp": 1.02073407,
      "epoch": 0.6256237599951903,
      "flos": 20227798753920.0,
      "grad_norm": 2.382466346471837,
      "language_loss": 0.70727563,
      "learning_rate": 1.2985841078631871e-06,
      "loss": 0.72909528,
      "num_input_tokens_seen": 112043850,
      "step": 5203,
      "time_per_iteration": 2.691460371017456
    },
    {
      "auxiliary_loss_clip": 0.01166782,
      "auxiliary_loss_mlp": 0.01034203,
      "balance_loss_clip": 1.0531075,
      "balance_loss_mlp": 1.02539325,
      "epoch": 0.6257440028858293,
      "flos": 24170574936960.0,
      "grad_norm": 2.0798783313938616,
      "language_loss": 0.78379667,
      "learning_rate": 1.2978546659115608e-06,
      "loss": 0.80580652,
      "num_input_tokens_seen": 112061930,
      "step": 5204,
      "time_per_iteration": 2.649643898010254
    },
    {
      "auxiliary_loss_clip": 0.01170388,
      "auxiliary_loss_mlp": 0.01026509,
      "balance_loss_clip": 1.05454803,
      "balance_loss_mlp": 1.01847398,
      "epoch": 0.6258642457764685,
      "flos": 15851940289920.0,
      "grad_norm": 1.890694244591785,
      "language_loss": 0.85546124,
      "learning_rate": 1.2971253304748228e-06,
      "loss": 0.8774302,
      "num_input_tokens_seen": 112079645,
      "step": 5205,
      "time_per_iteration": 2.6679975986480713
    },
    {
      "auxiliary_loss_clip": 0.01185229,
      "auxiliary_loss_mlp": 0.01036618,
      "balance_loss_clip": 1.05570912,
      "balance_loss_mlp": 1.02790999,
      "epoch": 0.6259844886671075,
      "flos": 11911354836480.0,
      "grad_norm": 1.7871489357728076,
      "language_loss": 0.75184417,
      "learning_rate": 1.296396101663614e-06,
      "loss": 0.77406263,
      "num_input_tokens_seen": 112096205,
      "step": 5206,
      "time_per_iteration": 2.5709009170532227
    },
    {
      "auxiliary_loss_clip": 0.01181033,
      "auxiliary_loss_mlp": 0.00761812,
      "balance_loss_clip": 1.05564022,
      "balance_loss_mlp": 1.00076365,
      "epoch": 0.6261047315577466,
      "flos": 15887958652800.0,
      "grad_norm": 4.694582454658745,
      "language_loss": 0.84275079,
      "learning_rate": 1.2956669795885565e-06,
      "loss": 0.86217928,
      "num_input_tokens_seen": 112112835,
      "step": 5207,
      "time_per_iteration": 2.5948729515075684
    },
    {
      "auxiliary_loss_clip": 0.011726,
      "auxiliary_loss_mlp": 0.01038898,
      "balance_loss_clip": 1.05550933,
      "balance_loss_mlp": 1.03067255,
      "epoch": 0.6262249744483858,
      "flos": 31248926916480.0,
      "grad_norm": 2.0767853339732247,
      "language_loss": 0.67966777,
      "learning_rate": 1.294937964360259e-06,
      "loss": 0.70178282,
      "num_input_tokens_seen": 112133105,
      "step": 5208,
      "time_per_iteration": 2.66799259185791
    },
    {
      "auxiliary_loss_clip": 0.0113306,
      "auxiliary_loss_mlp": 0.01031808,
      "balance_loss_clip": 1.04764426,
      "balance_loss_mlp": 1.02321899,
      "epoch": 0.6263452173390248,
      "flos": 27198598435200.0,
      "grad_norm": 2.2904124788401825,
      "language_loss": 0.71414876,
      "learning_rate": 1.2942090560893108e-06,
      "loss": 0.73579741,
      "num_input_tokens_seen": 112152510,
      "step": 5209,
      "time_per_iteration": 2.8031086921691895
    },
    {
      "auxiliary_loss_clip": 0.01179876,
      "auxiliary_loss_mlp": 0.0103048,
      "balance_loss_clip": 1.05420566,
      "balance_loss_mlp": 1.02261281,
      "epoch": 0.6264654602296639,
      "flos": 37342069683840.0,
      "grad_norm": 1.761007668338144,
      "language_loss": 0.60606098,
      "learning_rate": 1.2934802548862882e-06,
      "loss": 0.62816459,
      "num_input_tokens_seen": 112175295,
      "step": 5210,
      "time_per_iteration": 2.782245397567749
    },
    {
      "auxiliary_loss_clip": 0.01145024,
      "auxiliary_loss_mlp": 0.01028876,
      "balance_loss_clip": 1.04582202,
      "balance_loss_mlp": 1.02075791,
      "epoch": 0.626585703120303,
      "flos": 14756952136320.0,
      "grad_norm": 2.4420421956417147,
      "language_loss": 0.82507992,
      "learning_rate": 1.292751560861749e-06,
      "loss": 0.84681892,
      "num_input_tokens_seen": 112190200,
      "step": 5211,
      "time_per_iteration": 2.662954568862915
    },
    {
      "auxiliary_loss_clip": 0.01167733,
      "auxiliary_loss_mlp": 0.01033225,
      "balance_loss_clip": 1.05341673,
      "balance_loss_mlp": 1.0250349,
      "epoch": 0.6267059460109421,
      "flos": 22347318533760.0,
      "grad_norm": 1.773552669434702,
      "language_loss": 0.79918712,
      "learning_rate": 1.2920229741262354e-06,
      "loss": 0.82119668,
      "num_input_tokens_seen": 112208205,
      "step": 5212,
      "time_per_iteration": 2.6588072776794434
    },
    {
      "auxiliary_loss_clip": 0.01181356,
      "auxiliary_loss_mlp": 0.01029426,
      "balance_loss_clip": 1.05536175,
      "balance_loss_mlp": 1.02185655,
      "epoch": 0.6268261889015811,
      "flos": 17748813617280.0,
      "grad_norm": 2.783415061395415,
      "language_loss": 0.75528824,
      "learning_rate": 1.2912944947902739e-06,
      "loss": 0.77739608,
      "num_input_tokens_seen": 112224690,
      "step": 5213,
      "time_per_iteration": 2.617391586303711
    },
    {
      "auxiliary_loss_clip": 0.01165601,
      "auxiliary_loss_mlp": 0.01030163,
      "balance_loss_clip": 1.0523138,
      "balance_loss_mlp": 1.02150846,
      "epoch": 0.6269464317922203,
      "flos": 32846484211200.0,
      "grad_norm": 2.8148471626306555,
      "language_loss": 0.71484154,
      "learning_rate": 1.2905661229643742e-06,
      "loss": 0.73679918,
      "num_input_tokens_seen": 112244450,
      "step": 5214,
      "time_per_iteration": 2.7362234592437744
    },
    {
      "auxiliary_loss_clip": 0.01143547,
      "auxiliary_loss_mlp": 0.01029171,
      "balance_loss_clip": 1.04730344,
      "balance_loss_mlp": 1.02037907,
      "epoch": 0.6270666746828594,
      "flos": 17929192740480.0,
      "grad_norm": 2.477807783936362,
      "language_loss": 0.84324366,
      "learning_rate": 1.2898378587590299e-06,
      "loss": 0.86497086,
      "num_input_tokens_seen": 112261050,
      "step": 5215,
      "time_per_iteration": 2.6224842071533203
    },
    {
      "auxiliary_loss_clip": 0.01162947,
      "auxiliary_loss_mlp": 0.01034012,
      "balance_loss_clip": 1.05262065,
      "balance_loss_mlp": 1.02581084,
      "epoch": 0.6271869175734984,
      "flos": 17457326749440.0,
      "grad_norm": 2.942021596350266,
      "language_loss": 0.87473893,
      "learning_rate": 1.2891097022847173e-06,
      "loss": 0.89670849,
      "num_input_tokens_seen": 112278395,
      "step": 5216,
      "time_per_iteration": 3.530043601989746
    },
    {
      "auxiliary_loss_clip": 0.01182128,
      "auxiliary_loss_mlp": 0.01027509,
      "balance_loss_clip": 1.0522964,
      "balance_loss_mlp": 1.01918244,
      "epoch": 0.6273071604641376,
      "flos": 26868615166080.0,
      "grad_norm": 1.5921446417851493,
      "language_loss": 0.66922081,
      "learning_rate": 1.2883816536518978e-06,
      "loss": 0.6913172,
      "num_input_tokens_seen": 112299535,
      "step": 5217,
      "time_per_iteration": 4.5005176067352295
    },
    {
      "auxiliary_loss_clip": 0.01176555,
      "auxiliary_loss_mlp": 0.01027677,
      "balance_loss_clip": 1.0526973,
      "balance_loss_mlp": 1.02009535,
      "epoch": 0.6274274033547766,
      "flos": 26062384446720.0,
      "grad_norm": 2.1953245870549445,
      "language_loss": 0.8240062,
      "learning_rate": 1.2876537129710155e-06,
      "loss": 0.84604859,
      "num_input_tokens_seen": 112317265,
      "step": 5218,
      "time_per_iteration": 2.6108715534210205
    },
    {
      "auxiliary_loss_clip": 0.01184926,
      "auxiliary_loss_mlp": 0.01026778,
      "balance_loss_clip": 1.05585074,
      "balance_loss_mlp": 1.01802242,
      "epoch": 0.6275476462454157,
      "flos": 20266259241600.0,
      "grad_norm": 2.1734699373106516,
      "language_loss": 0.75851798,
      "learning_rate": 1.286925880352499e-06,
      "loss": 0.780635,
      "num_input_tokens_seen": 112336125,
      "step": 5219,
      "time_per_iteration": 2.5900626182556152
    },
    {
      "auxiliary_loss_clip": 0.01151319,
      "auxiliary_loss_mlp": 0.01029687,
      "balance_loss_clip": 1.05058694,
      "balance_loss_mlp": 1.0216043,
      "epoch": 0.6276678891360549,
      "flos": 26320402817280.0,
      "grad_norm": 2.2048184599547866,
      "language_loss": 0.71380389,
      "learning_rate": 1.2861981559067592e-06,
      "loss": 0.73561394,
      "num_input_tokens_seen": 112356730,
      "step": 5220,
      "time_per_iteration": 2.6552376747131348
    },
    {
      "auxiliary_loss_clip": 0.01140568,
      "auxiliary_loss_mlp": 0.01028876,
      "balance_loss_clip": 1.04802608,
      "balance_loss_mlp": 1.02040589,
      "epoch": 0.6277881320266939,
      "flos": 13912512324480.0,
      "grad_norm": 2.1102970623309125,
      "language_loss": 0.80429673,
      "learning_rate": 1.2854705397441917e-06,
      "loss": 0.82599115,
      "num_input_tokens_seen": 112372270,
      "step": 5221,
      "time_per_iteration": 2.6530239582061768
    },
    {
      "auxiliary_loss_clip": 0.01178482,
      "auxiliary_loss_mlp": 0.01027022,
      "balance_loss_clip": 1.05270267,
      "balance_loss_mlp": 1.01927364,
      "epoch": 0.627908374917333,
      "flos": 27048922462080.0,
      "grad_norm": 2.27256984955514,
      "language_loss": 0.77336365,
      "learning_rate": 1.2847430319751747e-06,
      "loss": 0.79541874,
      "num_input_tokens_seen": 112390365,
      "step": 5222,
      "time_per_iteration": 2.6053168773651123
    },
    {
      "auxiliary_loss_clip": 0.01180609,
      "auxiliary_loss_mlp": 0.01028266,
      "balance_loss_clip": 1.05570006,
      "balance_loss_mlp": 1.02007055,
      "epoch": 0.6280286178079721,
      "flos": 23769201386880.0,
      "grad_norm": 2.047538434269002,
      "language_loss": 0.67818964,
      "learning_rate": 1.2840156327100712e-06,
      "loss": 0.7002784,
      "num_input_tokens_seen": 112407490,
      "step": 5223,
      "time_per_iteration": 2.6049935817718506
    },
    {
      "auxiliary_loss_clip": 0.01151017,
      "auxiliary_loss_mlp": 0.01024802,
      "balance_loss_clip": 1.049474,
      "balance_loss_mlp": 1.01661205,
      "epoch": 0.6281488606986112,
      "flos": 26359150613760.0,
      "grad_norm": 1.9029627871451542,
      "language_loss": 0.72632062,
      "learning_rate": 1.2832883420592272e-06,
      "loss": 0.74807882,
      "num_input_tokens_seen": 112426385,
      "step": 5224,
      "time_per_iteration": 2.7282731533050537
    },
    {
      "auxiliary_loss_clip": 0.0114324,
      "auxiliary_loss_mlp": 0.01028995,
      "balance_loss_clip": 1.05056357,
      "balance_loss_mlp": 1.02072763,
      "epoch": 0.6282691035892503,
      "flos": 36137194848000.0,
      "grad_norm": 2.211651634241907,
      "language_loss": 0.64136028,
      "learning_rate": 1.282561160132972e-06,
      "loss": 0.66308266,
      "num_input_tokens_seen": 112446905,
      "step": 5225,
      "time_per_iteration": 2.80395245552063
    },
    {
      "auxiliary_loss_clip": 0.01180758,
      "auxiliary_loss_mlp": 0.01022617,
      "balance_loss_clip": 1.05394423,
      "balance_loss_mlp": 1.0144664,
      "epoch": 0.6283893464798894,
      "flos": 26537231266560.0,
      "grad_norm": 1.661878564787008,
      "language_loss": 0.80942398,
      "learning_rate": 1.2818340870416186e-06,
      "loss": 0.83145773,
      "num_input_tokens_seen": 112468040,
      "step": 5226,
      "time_per_iteration": 2.6664533615112305
    },
    {
      "auxiliary_loss_clip": 0.01183022,
      "auxiliary_loss_mlp": 0.01026129,
      "balance_loss_clip": 1.05440235,
      "balance_loss_mlp": 1.01728344,
      "epoch": 0.6285095893705285,
      "flos": 22237216369920.0,
      "grad_norm": 2.164602214842677,
      "language_loss": 0.75957412,
      "learning_rate": 1.2811071228954626e-06,
      "loss": 0.78166562,
      "num_input_tokens_seen": 112486675,
      "step": 5227,
      "time_per_iteration": 3.5416176319122314
    },
    {
      "auxiliary_loss_clip": 0.01167736,
      "auxiliary_loss_mlp": 0.01032324,
      "balance_loss_clip": 1.05177808,
      "balance_loss_mlp": 1.02449226,
      "epoch": 0.6286298322611675,
      "flos": 26542259170560.0,
      "grad_norm": 4.152825807394982,
      "language_loss": 0.81336665,
      "learning_rate": 1.2803802678047846e-06,
      "loss": 0.83536732,
      "num_input_tokens_seen": 112506825,
      "step": 5228,
      "time_per_iteration": 2.678494453430176
    },
    {
      "auxiliary_loss_clip": 0.01184238,
      "auxiliary_loss_mlp": 0.01038654,
      "balance_loss_clip": 1.05517519,
      "balance_loss_mlp": 1.02898037,
      "epoch": 0.6287500751518067,
      "flos": 21795227516160.0,
      "grad_norm": 1.9638011071815467,
      "language_loss": 0.73853582,
      "learning_rate": 1.279653521879848e-06,
      "loss": 0.76076472,
      "num_input_tokens_seen": 112526890,
      "step": 5229,
      "time_per_iteration": 2.591407537460327
    },
    {
      "auxiliary_loss_clip": 0.01175032,
      "auxiliary_loss_mlp": 0.0103124,
      "balance_loss_clip": 1.05225325,
      "balance_loss_mlp": 1.02330709,
      "epoch": 0.6288703180424458,
      "flos": 20009605587840.0,
      "grad_norm": 2.1364793270154454,
      "language_loss": 0.83770794,
      "learning_rate": 1.2789268852308997e-06,
      "loss": 0.85977066,
      "num_input_tokens_seen": 112542100,
      "step": 5230,
      "time_per_iteration": 2.631035089492798
    },
    {
      "auxiliary_loss_clip": 0.01167334,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.05476904,
      "balance_loss_mlp": 1.01961946,
      "epoch": 0.6289905609330848,
      "flos": 22124923476480.0,
      "grad_norm": 1.7993621156603132,
      "language_loss": 0.707605,
      "learning_rate": 1.2782003579681688e-06,
      "loss": 0.72955787,
      "num_input_tokens_seen": 112561630,
      "step": 5231,
      "time_per_iteration": 2.633986234664917
    },
    {
      "auxiliary_loss_clip": 0.011672,
      "auxiliary_loss_mlp": 0.01022351,
      "balance_loss_clip": 1.04962707,
      "balance_loss_mlp": 1.01417351,
      "epoch": 0.629110803823724,
      "flos": 25518481729920.0,
      "grad_norm": 1.6717514604873716,
      "language_loss": 0.7474153,
      "learning_rate": 1.2774739402018701e-06,
      "loss": 0.76931083,
      "num_input_tokens_seen": 112582465,
      "step": 5232,
      "time_per_iteration": 2.7062840461730957
    },
    {
      "auxiliary_loss_clip": 0.01147174,
      "auxiliary_loss_mlp": 0.01031379,
      "balance_loss_clip": 1.04915476,
      "balance_loss_mlp": 1.02235508,
      "epoch": 0.629231046714363,
      "flos": 20886616056960.0,
      "grad_norm": 2.100123395154449,
      "language_loss": 0.73336291,
      "learning_rate": 1.2767476320422002e-06,
      "loss": 0.75514853,
      "num_input_tokens_seen": 112602390,
      "step": 5233,
      "time_per_iteration": 2.712082862854004
    },
    {
      "auxiliary_loss_clip": 0.01046589,
      "auxiliary_loss_mlp": 0.01003158,
      "balance_loss_clip": 1.00862098,
      "balance_loss_mlp": 1.00213325,
      "epoch": 0.6293512896050021,
      "flos": 65050027908480.0,
      "grad_norm": 0.6863067041527952,
      "language_loss": 0.57144982,
      "learning_rate": 1.2760214335993392e-06,
      "loss": 0.59194732,
      "num_input_tokens_seen": 112669035,
      "step": 5234,
      "time_per_iteration": 3.3014490604400635
    },
    {
      "auxiliary_loss_clip": 0.01175594,
      "auxiliary_loss_mlp": 0.01031021,
      "balance_loss_clip": 1.05201364,
      "balance_loss_mlp": 1.02369881,
      "epoch": 0.6294715324956413,
      "flos": 34677857088000.0,
      "grad_norm": 2.1904535937305885,
      "language_loss": 0.59253442,
      "learning_rate": 1.2752953449834514e-06,
      "loss": 0.61460054,
      "num_input_tokens_seen": 112691485,
      "step": 5235,
      "time_per_iteration": 2.809359073638916
    },
    {
      "auxiliary_loss_clip": 0.01176745,
      "auxiliary_loss_mlp": 0.01030215,
      "balance_loss_clip": 1.05264664,
      "balance_loss_mlp": 1.02238905,
      "epoch": 0.6295917753862803,
      "flos": 22784207656320.0,
      "grad_norm": 1.6482544239531784,
      "language_loss": 0.80390346,
      "learning_rate": 1.2745693663046836e-06,
      "loss": 0.82597315,
      "num_input_tokens_seen": 112710555,
      "step": 5236,
      "time_per_iteration": 2.577061176300049
    },
    {
      "auxiliary_loss_clip": 0.01162738,
      "auxiliary_loss_mlp": 0.01028073,
      "balance_loss_clip": 1.05384004,
      "balance_loss_mlp": 1.02071166,
      "epoch": 0.6297120182769194,
      "flos": 20850454039680.0,
      "grad_norm": 1.7044322872542688,
      "language_loss": 0.80252743,
      "learning_rate": 1.2738434976731662e-06,
      "loss": 0.82443547,
      "num_input_tokens_seen": 112728740,
      "step": 5237,
      "time_per_iteration": 2.68979549407959
    },
    {
      "auxiliary_loss_clip": 0.01185458,
      "auxiliary_loss_mlp": 0.01032748,
      "balance_loss_clip": 1.05811226,
      "balance_loss_mlp": 1.02406347,
      "epoch": 0.6298322611675584,
      "flos": 19497662997120.0,
      "grad_norm": 1.5035754201241824,
      "language_loss": 0.75335032,
      "learning_rate": 1.2731177391990125e-06,
      "loss": 0.77553236,
      "num_input_tokens_seen": 112748665,
      "step": 5238,
      "time_per_iteration": 2.6456069946289062
    },
    {
      "auxiliary_loss_clip": 0.01161134,
      "auxiliary_loss_mlp": 0.0102654,
      "balance_loss_clip": 1.05046403,
      "balance_loss_mlp": 1.01840413,
      "epoch": 0.6299525040581976,
      "flos": 12604466649600.0,
      "grad_norm": 1.9432243012888082,
      "language_loss": 0.82292902,
      "learning_rate": 1.2723920909923203e-06,
      "loss": 0.84480578,
      "num_input_tokens_seen": 112764410,
      "step": 5239,
      "time_per_iteration": 2.6567914485931396
    },
    {
      "auxiliary_loss_clip": 0.01059329,
      "auxiliary_loss_mlp": 0.0100051,
      "balance_loss_clip": 1.01320195,
      "balance_loss_mlp": 0.9994486,
      "epoch": 0.6300727469488366,
      "flos": 57725685636480.0,
      "grad_norm": 0.8546075659097953,
      "language_loss": 0.60430038,
      "learning_rate": 1.2716665531631688e-06,
      "loss": 0.62489879,
      "num_input_tokens_seen": 112818695,
      "step": 5240,
      "time_per_iteration": 3.160109758377075
    },
    {
      "auxiliary_loss_clip": 0.01182065,
      "auxiliary_loss_mlp": 0.01032464,
      "balance_loss_clip": 1.05383611,
      "balance_loss_mlp": 1.02427483,
      "epoch": 0.6301929898394757,
      "flos": 22527302607360.0,
      "grad_norm": 2.8630849191723384,
      "language_loss": 0.77054453,
      "learning_rate": 1.270941125821623e-06,
      "loss": 0.79268986,
      "num_input_tokens_seen": 112839120,
      "step": 5241,
      "time_per_iteration": 2.6632516384124756
    },
    {
      "auxiliary_loss_clip": 0.01176582,
      "auxiliary_loss_mlp": 0.01031843,
      "balance_loss_clip": 1.05143034,
      "balance_loss_mlp": 1.02363837,
      "epoch": 0.6303132327301149,
      "flos": 28293550675200.0,
      "grad_norm": 1.6347641417688288,
      "language_loss": 0.75284171,
      "learning_rate": 1.2702158090777278e-06,
      "loss": 0.77492589,
      "num_input_tokens_seen": 112860210,
      "step": 5242,
      "time_per_iteration": 3.583233594894409
    },
    {
      "auxiliary_loss_clip": 0.01158292,
      "auxiliary_loss_mlp": 0.01025886,
      "balance_loss_clip": 1.05096698,
      "balance_loss_mlp": 1.01795268,
      "epoch": 0.6304334756207539,
      "flos": 25264521596160.0,
      "grad_norm": 2.053059663989284,
      "language_loss": 0.7496922,
      "learning_rate": 1.2694906030415148e-06,
      "loss": 0.77153403,
      "num_input_tokens_seen": 112877955,
      "step": 5243,
      "time_per_iteration": 3.574709177017212
    },
    {
      "auxiliary_loss_clip": 0.01169243,
      "auxiliary_loss_mlp": 0.01026156,
      "balance_loss_clip": 1.0518775,
      "balance_loss_mlp": 1.01693535,
      "epoch": 0.630553718511393,
      "flos": 18033548728320.0,
      "grad_norm": 2.968128460615419,
      "language_loss": 0.829862,
      "learning_rate": 1.2687655078229958e-06,
      "loss": 0.85181594,
      "num_input_tokens_seen": 112892285,
      "step": 5244,
      "time_per_iteration": 3.5184667110443115
    },
    {
      "auxiliary_loss_clip": 0.01145654,
      "auxiliary_loss_mlp": 0.01034285,
      "balance_loss_clip": 1.04884481,
      "balance_loss_mlp": 1.02594948,
      "epoch": 0.6306739614020321,
      "flos": 27304103658240.0,
      "grad_norm": 2.092936892476573,
      "language_loss": 0.69294202,
      "learning_rate": 1.2680405235321678e-06,
      "loss": 0.71474147,
      "num_input_tokens_seen": 112913620,
      "step": 5245,
      "time_per_iteration": 2.7480530738830566
    },
    {
      "auxiliary_loss_clip": 0.0116721,
      "auxiliary_loss_mlp": 0.01035211,
      "balance_loss_clip": 1.054461,
      "balance_loss_mlp": 1.02610326,
      "epoch": 0.6307942042926712,
      "flos": 15341434243200.0,
      "grad_norm": 2.5644180659407376,
      "language_loss": 0.78929377,
      "learning_rate": 1.267315650279011e-06,
      "loss": 0.81131792,
      "num_input_tokens_seen": 112932090,
      "step": 5246,
      "time_per_iteration": 2.6780216693878174
    },
    {
      "auxiliary_loss_clip": 0.01152601,
      "auxiliary_loss_mlp": 0.01030604,
      "balance_loss_clip": 1.05130935,
      "balance_loss_mlp": 1.02272439,
      "epoch": 0.6309144471833102,
      "flos": 19606400444160.0,
      "grad_norm": 1.9527812717900235,
      "language_loss": 0.73853445,
      "learning_rate": 1.2665908881734874e-06,
      "loss": 0.7603665,
      "num_input_tokens_seen": 112950925,
      "step": 5247,
      "time_per_iteration": 2.722710132598877
    },
    {
      "auxiliary_loss_clip": 0.01163656,
      "auxiliary_loss_mlp": 0.01026907,
      "balance_loss_clip": 1.05298185,
      "balance_loss_mlp": 1.01921749,
      "epoch": 0.6310346900739494,
      "flos": 17493345112320.0,
      "grad_norm": 2.8224061520917543,
      "language_loss": 0.85072017,
      "learning_rate": 1.2658662373255432e-06,
      "loss": 0.87262577,
      "num_input_tokens_seen": 112969315,
      "step": 5248,
      "time_per_iteration": 2.588752031326294
    },
    {
      "auxiliary_loss_clip": 0.01058077,
      "auxiliary_loss_mlp": 0.01004155,
      "balance_loss_clip": 1.00962257,
      "balance_loss_mlp": 1.00312376,
      "epoch": 0.6311549329645885,
      "flos": 55070164131840.0,
      "grad_norm": 0.7069270290591277,
      "language_loss": 0.52144516,
      "learning_rate": 1.2651416978451063e-06,
      "loss": 0.54206747,
      "num_input_tokens_seen": 113034700,
      "step": 5249,
      "time_per_iteration": 3.2794952392578125
    },
    {
      "auxiliary_loss_clip": 0.01184288,
      "auxiliary_loss_mlp": 0.01029382,
      "balance_loss_clip": 1.05578017,
      "balance_loss_mlp": 1.02119851,
      "epoch": 0.6312751758552275,
      "flos": 41902545075840.0,
      "grad_norm": 2.555323710709368,
      "language_loss": 0.65294731,
      "learning_rate": 1.2644172698420903e-06,
      "loss": 0.67508399,
      "num_input_tokens_seen": 113056805,
      "step": 5250,
      "time_per_iteration": 2.760822057723999
    },
    {
      "auxiliary_loss_clip": 0.01167224,
      "auxiliary_loss_mlp": 0.0102647,
      "balance_loss_clip": 1.05417204,
      "balance_loss_mlp": 1.01863813,
      "epoch": 0.6313954187458667,
      "flos": 19646800266240.0,
      "grad_norm": 3.6252640231589583,
      "language_loss": 0.84841996,
      "learning_rate": 1.2636929534263892e-06,
      "loss": 0.87035692,
      "num_input_tokens_seen": 113075790,
      "step": 5251,
      "time_per_iteration": 2.6727819442749023
    },
    {
      "auxiliary_loss_clip": 0.01154184,
      "auxiliary_loss_mlp": 0.01026526,
      "balance_loss_clip": 1.05341721,
      "balance_loss_mlp": 1.01799059,
      "epoch": 0.6315156616365057,
      "flos": 22894273906560.0,
      "grad_norm": 1.6818990133686484,
      "language_loss": 0.77484047,
      "learning_rate": 1.2629687487078821e-06,
      "loss": 0.79664755,
      "num_input_tokens_seen": 113094600,
      "step": 5252,
      "time_per_iteration": 2.7128496170043945
    },
    {
      "auxiliary_loss_clip": 0.01163846,
      "auxiliary_loss_mlp": 0.01027386,
      "balance_loss_clip": 1.05203938,
      "balance_loss_mlp": 1.01899338,
      "epoch": 0.6316359045271448,
      "flos": 23726251699200.0,
      "grad_norm": 2.841893564605264,
      "language_loss": 0.76817548,
      "learning_rate": 1.2622446557964293e-06,
      "loss": 0.79008776,
      "num_input_tokens_seen": 113112605,
      "step": 5253,
      "time_per_iteration": 3.624033212661743
    },
    {
      "auxiliary_loss_clip": 0.01177173,
      "auxiliary_loss_mlp": 0.0102752,
      "balance_loss_clip": 1.05299997,
      "balance_loss_mlp": 1.02012014,
      "epoch": 0.631756147417784,
      "flos": 33108417164160.0,
      "grad_norm": 1.8248212090206246,
      "language_loss": 0.71511281,
      "learning_rate": 1.261520674801876e-06,
      "loss": 0.73715973,
      "num_input_tokens_seen": 113133200,
      "step": 5254,
      "time_per_iteration": 2.7104270458221436
    },
    {
      "auxiliary_loss_clip": 0.01166767,
      "auxiliary_loss_mlp": 0.01025954,
      "balance_loss_clip": 1.05347943,
      "balance_loss_mlp": 1.01772809,
      "epoch": 0.631876390308423,
      "flos": 31248424126080.0,
      "grad_norm": 5.960109019135604,
      "language_loss": 0.7226851,
      "learning_rate": 1.2607968058340488e-06,
      "loss": 0.74461234,
      "num_input_tokens_seen": 113152895,
      "step": 5255,
      "time_per_iteration": 2.755009651184082
    },
    {
      "auxiliary_loss_clip": 0.01177844,
      "auxiliary_loss_mlp": 0.01026916,
      "balance_loss_clip": 1.05323255,
      "balance_loss_mlp": 1.01883399,
      "epoch": 0.6319966331990621,
      "flos": 24681152810880.0,
      "grad_norm": 3.3368319670791693,
      "language_loss": 0.73739111,
      "learning_rate": 1.2600730490027583e-06,
      "loss": 0.75943869,
      "num_input_tokens_seen": 113173135,
      "step": 5256,
      "time_per_iteration": 2.6595475673675537
    },
    {
      "auxiliary_loss_clip": 0.0117947,
      "auxiliary_loss_mlp": 0.01024956,
      "balance_loss_clip": 1.05369353,
      "balance_loss_mlp": 1.01708567,
      "epoch": 0.6321168760897012,
      "flos": 17491764913920.0,
      "grad_norm": 1.6315745732633142,
      "language_loss": 0.8048414,
      "learning_rate": 1.2593494044177984e-06,
      "loss": 0.8268857,
      "num_input_tokens_seen": 113191440,
      "step": 5257,
      "time_per_iteration": 2.5594310760498047
    },
    {
      "auxiliary_loss_clip": 0.01166489,
      "auxiliary_loss_mlp": 0.0103053,
      "balance_loss_clip": 1.04795504,
      "balance_loss_mlp": 1.0218817,
      "epoch": 0.6322371189803403,
      "flos": 18295373940480.0,
      "grad_norm": 2.587573621581484,
      "language_loss": 0.80577785,
      "learning_rate": 1.2586258721889448e-06,
      "loss": 0.827748,
      "num_input_tokens_seen": 113208790,
      "step": 5258,
      "time_per_iteration": 2.640752077102661
    },
    {
      "auxiliary_loss_clip": 0.01143146,
      "auxiliary_loss_mlp": 0.01030794,
      "balance_loss_clip": 1.04935849,
      "balance_loss_mlp": 1.02270865,
      "epoch": 0.6323573618709794,
      "flos": 20157270399360.0,
      "grad_norm": 1.9343771515539694,
      "language_loss": 0.82014406,
      "learning_rate": 1.2579024524259573e-06,
      "loss": 0.84188348,
      "num_input_tokens_seen": 113225050,
      "step": 5259,
      "time_per_iteration": 2.7304792404174805
    },
    {
      "auxiliary_loss_clip": 0.01180789,
      "auxiliary_loss_mlp": 0.01031183,
      "balance_loss_clip": 1.05444932,
      "balance_loss_mlp": 1.02260566,
      "epoch": 0.6324776047616185,
      "flos": 20042391726720.0,
      "grad_norm": 1.857204351491142,
      "language_loss": 0.91049087,
      "learning_rate": 1.2571791452385768e-06,
      "loss": 0.93261057,
      "num_input_tokens_seen": 113242315,
      "step": 5260,
      "time_per_iteration": 2.6217899322509766
    },
    {
      "auxiliary_loss_clip": 0.01179408,
      "auxiliary_loss_mlp": 0.01030412,
      "balance_loss_clip": 1.05538583,
      "balance_loss_mlp": 1.02254081,
      "epoch": 0.6325978476522576,
      "flos": 30848235724800.0,
      "grad_norm": 1.7076901537409845,
      "language_loss": 0.77721322,
      "learning_rate": 1.2564559507365301e-06,
      "loss": 0.7993114,
      "num_input_tokens_seen": 113264720,
      "step": 5261,
      "time_per_iteration": 2.740657091140747
    },
    {
      "auxiliary_loss_clip": 0.01182096,
      "auxiliary_loss_mlp": 0.01033383,
      "balance_loss_clip": 1.05398393,
      "balance_loss_mlp": 1.02469277,
      "epoch": 0.6327180905428966,
      "flos": 24535104111360.0,
      "grad_norm": 2.4986586814191982,
      "language_loss": 0.78740275,
      "learning_rate": 1.2557328690295244e-06,
      "loss": 0.8095575,
      "num_input_tokens_seen": 113282910,
      "step": 5262,
      "time_per_iteration": 2.675488233566284
    },
    {
      "auxiliary_loss_clip": 0.01179456,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 1.05384159,
      "balance_loss_mlp": 1.01979816,
      "epoch": 0.6328383334335358,
      "flos": 21575274583680.0,
      "grad_norm": 1.7191872710030465,
      "language_loss": 0.76191181,
      "learning_rate": 1.255009900227251e-06,
      "loss": 0.78397834,
      "num_input_tokens_seen": 113301935,
      "step": 5263,
      "time_per_iteration": 2.60857892036438
    },
    {
      "auxiliary_loss_clip": 0.01151006,
      "auxiliary_loss_mlp": 0.01028198,
      "balance_loss_clip": 1.05206251,
      "balance_loss_mlp": 1.02000797,
      "epoch": 0.6329585763241748,
      "flos": 22929861306240.0,
      "grad_norm": 1.886151952505908,
      "language_loss": 0.79717791,
      "learning_rate": 1.254287044439383e-06,
      "loss": 0.81896996,
      "num_input_tokens_seen": 113321540,
      "step": 5264,
      "time_per_iteration": 2.6774258613586426
    },
    {
      "auxiliary_loss_clip": 0.01019125,
      "auxiliary_loss_mlp": 0.0100395,
      "balance_loss_clip": 1.00769544,
      "balance_loss_mlp": 1.00297225,
      "epoch": 0.6330788192148139,
      "flos": 70936897847040.0,
      "grad_norm": 0.7785820639220089,
      "language_loss": 0.54387879,
      "learning_rate": 1.2535643017755776e-06,
      "loss": 0.56410956,
      "num_input_tokens_seen": 113383730,
      "step": 5265,
      "time_per_iteration": 3.4028475284576416
    },
    {
      "auxiliary_loss_clip": 0.01177337,
      "auxiliary_loss_mlp": 0.01028763,
      "balance_loss_clip": 1.05230129,
      "balance_loss_mlp": 1.02081764,
      "epoch": 0.6331990621054531,
      "flos": 21244501215360.0,
      "grad_norm": 2.304459635335986,
      "language_loss": 0.72781962,
      "learning_rate": 1.2528416723454737e-06,
      "loss": 0.74988061,
      "num_input_tokens_seen": 113400400,
      "step": 5266,
      "time_per_iteration": 2.5443708896636963
    },
    {
      "auxiliary_loss_clip": 0.01164799,
      "auxiliary_loss_mlp": 0.01022414,
      "balance_loss_clip": 1.05282772,
      "balance_loss_mlp": 1.01531851,
      "epoch": 0.6333193049960921,
      "flos": 34459412526720.0,
      "grad_norm": 1.439333791125118,
      "language_loss": 0.71407431,
      "learning_rate": 1.2521191562586945e-06,
      "loss": 0.73594642,
      "num_input_tokens_seen": 113424050,
      "step": 5267,
      "time_per_iteration": 2.7852485179901123
    },
    {
      "auxiliary_loss_clip": 0.01157797,
      "auxiliary_loss_mlp": 0.01028137,
      "balance_loss_clip": 1.05065179,
      "balance_loss_mlp": 1.02069223,
      "epoch": 0.6334395478867312,
      "flos": 18329883932160.0,
      "grad_norm": 2.239581091299141,
      "language_loss": 0.76670897,
      "learning_rate": 1.2513967536248445e-06,
      "loss": 0.78856832,
      "num_input_tokens_seen": 113440370,
      "step": 5268,
      "time_per_iteration": 3.638347625732422
    },
    {
      "auxiliary_loss_clip": 0.01164893,
      "auxiliary_loss_mlp": 0.01033002,
      "balance_loss_clip": 1.0515244,
      "balance_loss_mlp": 1.02505338,
      "epoch": 0.6335597907773702,
      "flos": 23623152687360.0,
      "grad_norm": 1.8546261004127094,
      "language_loss": 0.80875885,
      "learning_rate": 1.2506744645535117e-06,
      "loss": 0.83073783,
      "num_input_tokens_seen": 113460800,
      "step": 5269,
      "time_per_iteration": 3.6121134757995605
    },
    {
      "auxiliary_loss_clip": 0.01164817,
      "auxiliary_loss_mlp": 0.01026447,
      "balance_loss_clip": 1.05161905,
      "balance_loss_mlp": 1.01855266,
      "epoch": 0.6336800336680094,
      "flos": 22710913954560.0,
      "grad_norm": 2.027779430628669,
      "language_loss": 0.60424101,
      "learning_rate": 1.249952289154267e-06,
      "loss": 0.62615365,
      "num_input_tokens_seen": 113480840,
      "step": 5270,
      "time_per_iteration": 3.572197437286377
    },
    {
      "auxiliary_loss_clip": 0.01165477,
      "auxiliary_loss_mlp": 0.0076224,
      "balance_loss_clip": 1.05117881,
      "balance_loss_mlp": 1.00077128,
      "epoch": 0.6338002765586485,
      "flos": 23622757637760.0,
      "grad_norm": 1.8128357684014533,
      "language_loss": 0.76526308,
      "learning_rate": 1.2492302275366635e-06,
      "loss": 0.78454024,
      "num_input_tokens_seen": 113500515,
      "step": 5271,
      "time_per_iteration": 2.726893424987793
    },
    {
      "auxiliary_loss_clip": 0.01168011,
      "auxiliary_loss_mlp": 0.01029805,
      "balance_loss_clip": 1.05236816,
      "balance_loss_mlp": 1.0205127,
      "epoch": 0.6339205194492875,
      "flos": 26505450708480.0,
      "grad_norm": 2.262090516166846,
      "language_loss": 0.65816909,
      "learning_rate": 1.2485082798102377e-06,
      "loss": 0.68014729,
      "num_input_tokens_seen": 113520930,
      "step": 5272,
      "time_per_iteration": 2.6710355281829834
    },
    {
      "auxiliary_loss_clip": 0.01155744,
      "auxiliary_loss_mlp": 0.01035258,
      "balance_loss_clip": 1.05581462,
      "balance_loss_mlp": 1.02693152,
      "epoch": 0.6340407623399267,
      "flos": 18544306170240.0,
      "grad_norm": 2.2974530244664106,
      "language_loss": 0.69083679,
      "learning_rate": 1.2477864460845084e-06,
      "loss": 0.71274674,
      "num_input_tokens_seen": 113537330,
      "step": 5273,
      "time_per_iteration": 2.6453969478607178
    },
    {
      "auxiliary_loss_clip": 0.01173509,
      "auxiliary_loss_mlp": 0.01033803,
      "balance_loss_clip": 1.05593824,
      "balance_loss_mlp": 1.02477288,
      "epoch": 0.6341610052305657,
      "flos": 17712579772800.0,
      "grad_norm": 3.1483078505402955,
      "language_loss": 0.73838478,
      "learning_rate": 1.2470647264689776e-06,
      "loss": 0.76045793,
      "num_input_tokens_seen": 113555810,
      "step": 5274,
      "time_per_iteration": 2.6233742237091064
    },
    {
      "auxiliary_loss_clip": 0.01140332,
      "auxiliary_loss_mlp": 0.01029092,
      "balance_loss_clip": 1.04685283,
      "balance_loss_mlp": 1.02089643,
      "epoch": 0.6342812481212048,
      "flos": 23587026583680.0,
      "grad_norm": 16.05707362626258,
      "language_loss": 0.71111602,
      "learning_rate": 1.2463431210731282e-06,
      "loss": 0.73281026,
      "num_input_tokens_seen": 113575395,
      "step": 5275,
      "time_per_iteration": 2.7534773349761963
    },
    {
      "auxiliary_loss_clip": 0.01183658,
      "auxiliary_loss_mlp": 0.01027234,
      "balance_loss_clip": 1.05474663,
      "balance_loss_mlp": 1.01939583,
      "epoch": 0.634401491011844,
      "flos": 17821927751040.0,
      "grad_norm": 3.0587787299890787,
      "language_loss": 0.76678407,
      "learning_rate": 1.2456216300064289e-06,
      "loss": 0.78889298,
      "num_input_tokens_seen": 113592945,
      "step": 5276,
      "time_per_iteration": 2.648106336593628
    },
    {
      "auxiliary_loss_clip": 0.01171156,
      "auxiliary_loss_mlp": 0.01027085,
      "balance_loss_clip": 1.05447531,
      "balance_loss_mlp": 1.01934552,
      "epoch": 0.634521733902483,
      "flos": 21358158825600.0,
      "grad_norm": 1.8382672864132814,
      "language_loss": 0.78311509,
      "learning_rate": 1.244900253378328e-06,
      "loss": 0.80509746,
      "num_input_tokens_seen": 113613000,
      "step": 5277,
      "time_per_iteration": 2.7010719776153564
    },
    {
      "auxiliary_loss_clip": 0.01179527,
      "auxiliary_loss_mlp": 0.0102609,
      "balance_loss_clip": 1.05323195,
      "balance_loss_mlp": 1.01794767,
      "epoch": 0.6346419767931221,
      "flos": 16545052103040.0,
      "grad_norm": 2.4875911796187795,
      "language_loss": 0.69749111,
      "learning_rate": 1.2441789912982583e-06,
      "loss": 0.71954727,
      "num_input_tokens_seen": 113630085,
      "step": 5278,
      "time_per_iteration": 2.556210517883301
    },
    {
      "auxiliary_loss_clip": 0.01155971,
      "auxiliary_loss_mlp": 0.01033752,
      "balance_loss_clip": 1.05216932,
      "balance_loss_mlp": 1.02557445,
      "epoch": 0.6347622196837612,
      "flos": 24350989973760.0,
      "grad_norm": 2.6316359464888945,
      "language_loss": 0.65012562,
      "learning_rate": 1.2434578438756346e-06,
      "loss": 0.67202282,
      "num_input_tokens_seen": 113650515,
      "step": 5279,
      "time_per_iteration": 2.7042386531829834
    },
    {
      "auxiliary_loss_clip": 0.01154975,
      "auxiliary_loss_mlp": 0.01026705,
      "balance_loss_clip": 1.04701948,
      "balance_loss_mlp": 1.0191741,
      "epoch": 0.6348824625744003,
      "flos": 64523178195840.0,
      "grad_norm": 2.624610074727126,
      "language_loss": 0.78696316,
      "learning_rate": 1.242736811219855e-06,
      "loss": 0.80877995,
      "num_input_tokens_seen": 113676475,
      "step": 5280,
      "time_per_iteration": 3.993685722351074
    },
    {
      "auxiliary_loss_clip": 0.01165883,
      "auxiliary_loss_mlp": 0.01030607,
      "balance_loss_clip": 1.05149078,
      "balance_loss_mlp": 1.02314472,
      "epoch": 0.6350027054650393,
      "flos": 28622133313920.0,
      "grad_norm": 1.9307778768236663,
      "language_loss": 0.8180083,
      "learning_rate": 1.2420158934402988e-06,
      "loss": 0.83997315,
      "num_input_tokens_seen": 113697090,
      "step": 5281,
      "time_per_iteration": 2.658458948135376
    },
    {
      "auxiliary_loss_clip": 0.01148586,
      "auxiliary_loss_mlp": 0.01028871,
      "balance_loss_clip": 1.04806042,
      "balance_loss_mlp": 1.01985931,
      "epoch": 0.6351229483556785,
      "flos": 23002544476800.0,
      "grad_norm": 2.43788784776368,
      "language_loss": 0.85037488,
      "learning_rate": 1.2412950906463286e-06,
      "loss": 0.87214947,
      "num_input_tokens_seen": 113714395,
      "step": 5282,
      "time_per_iteration": 2.72418475151062
    },
    {
      "auxiliary_loss_clip": 0.01176448,
      "auxiliary_loss_mlp": 0.01028076,
      "balance_loss_clip": 1.05360162,
      "balance_loss_mlp": 1.02063739,
      "epoch": 0.6352431912463176,
      "flos": 21939300967680.0,
      "grad_norm": 1.7554300073449687,
      "language_loss": 0.89944762,
      "learning_rate": 1.2405744029472902e-06,
      "loss": 0.92149287,
      "num_input_tokens_seen": 113733880,
      "step": 5283,
      "time_per_iteration": 2.6366872787475586
    },
    {
      "auxiliary_loss_clip": 0.01178191,
      "auxiliary_loss_mlp": 0.0103695,
      "balance_loss_clip": 1.05421197,
      "balance_loss_mlp": 1.02861142,
      "epoch": 0.6353634341369566,
      "flos": 13735257684480.0,
      "grad_norm": 3.439085516834877,
      "language_loss": 0.76403558,
      "learning_rate": 1.2398538304525108e-06,
      "loss": 0.78618699,
      "num_input_tokens_seen": 113752505,
      "step": 5284,
      "time_per_iteration": 2.6121253967285156
    },
    {
      "auxiliary_loss_clip": 0.01170603,
      "auxiliary_loss_mlp": 0.01036223,
      "balance_loss_clip": 1.05112982,
      "balance_loss_mlp": 1.02663267,
      "epoch": 0.6354836770275958,
      "flos": 19316170552320.0,
      "grad_norm": 2.0906561851183114,
      "language_loss": 0.76050365,
      "learning_rate": 1.2391333732713016e-06,
      "loss": 0.78257191,
      "num_input_tokens_seen": 113770310,
      "step": 5285,
      "time_per_iteration": 2.59319806098938
    },
    {
      "auxiliary_loss_clip": 0.01167459,
      "auxiliary_loss_mlp": 0.01029604,
      "balance_loss_clip": 1.05318093,
      "balance_loss_mlp": 1.0203954,
      "epoch": 0.6356039199182348,
      "flos": 21613375935360.0,
      "grad_norm": 2.1254426844342005,
      "language_loss": 0.78761834,
      "learning_rate": 1.2384130315129543e-06,
      "loss": 0.80958891,
      "num_input_tokens_seen": 113788635,
      "step": 5286,
      "time_per_iteration": 2.681259870529175
    },
    {
      "auxiliary_loss_clip": 0.01168121,
      "auxiliary_loss_mlp": 0.01027303,
      "balance_loss_clip": 1.05293465,
      "balance_loss_mlp": 1.01865149,
      "epoch": 0.6357241628088739,
      "flos": 18111978074880.0,
      "grad_norm": 2.5695725469111044,
      "language_loss": 0.73558104,
      "learning_rate": 1.2376928052867447e-06,
      "loss": 0.75753534,
      "num_input_tokens_seen": 113807755,
      "step": 5287,
      "time_per_iteration": 2.6290066242218018
    },
    {
      "auxiliary_loss_clip": 0.01154464,
      "auxiliary_loss_mlp": 0.01030742,
      "balance_loss_clip": 1.04983473,
      "balance_loss_mlp": 1.02272558,
      "epoch": 0.6358444056995131,
      "flos": 24935256599040.0,
      "grad_norm": 2.608071912057352,
      "language_loss": 0.78144467,
      "learning_rate": 1.2369726947019299e-06,
      "loss": 0.80329669,
      "num_input_tokens_seen": 113828230,
      "step": 5288,
      "time_per_iteration": 2.697744131088257
    },
    {
      "auxiliary_loss_clip": 0.01160304,
      "auxiliary_loss_mlp": 0.01028919,
      "balance_loss_clip": 1.05142188,
      "balance_loss_mlp": 1.02072954,
      "epoch": 0.6359646485901521,
      "flos": 23293348986240.0,
      "grad_norm": 2.339026082428597,
      "language_loss": 0.67563403,
      "learning_rate": 1.2362526998677511e-06,
      "loss": 0.69752622,
      "num_input_tokens_seen": 113844595,
      "step": 5289,
      "time_per_iteration": 2.6922264099121094
    },
    {
      "auxiliary_loss_clip": 0.01178288,
      "auxiliary_loss_mlp": 0.01024922,
      "balance_loss_clip": 1.05361724,
      "balance_loss_mlp": 1.01714349,
      "epoch": 0.6360848914807912,
      "flos": 20887442069760.0,
      "grad_norm": 2.0665042118696118,
      "language_loss": 0.84394467,
      "learning_rate": 1.2355328208934301e-06,
      "loss": 0.86597681,
      "num_input_tokens_seen": 113863470,
      "step": 5290,
      "time_per_iteration": 2.660754680633545
    },
    {
      "auxiliary_loss_clip": 0.01178718,
      "auxiliary_loss_mlp": 0.01027829,
      "balance_loss_clip": 1.0528717,
      "balance_loss_mlp": 1.01980615,
      "epoch": 0.6362051343714303,
      "flos": 18479775386880.0,
      "grad_norm": 2.3529970389249026,
      "language_loss": 0.72632647,
      "learning_rate": 1.2348130578881728e-06,
      "loss": 0.74839193,
      "num_input_tokens_seen": 113881690,
      "step": 5291,
      "time_per_iteration": 2.6298882961273193
    },
    {
      "auxiliary_loss_clip": 0.01182599,
      "auxiliary_loss_mlp": 0.01030111,
      "balance_loss_clip": 1.05471981,
      "balance_loss_mlp": 1.02152753,
      "epoch": 0.6363253772620694,
      "flos": 24389594115840.0,
      "grad_norm": 1.878670652912046,
      "language_loss": 0.76222563,
      "learning_rate": 1.2340934109611664e-06,
      "loss": 0.78435278,
      "num_input_tokens_seen": 113902450,
      "step": 5292,
      "time_per_iteration": 2.6525778770446777
    },
    {
      "auxiliary_loss_clip": 0.01187551,
      "auxiliary_loss_mlp": 0.01034254,
      "balance_loss_clip": 1.05573201,
      "balance_loss_mlp": 1.02521253,
      "epoch": 0.6364456201527084,
      "flos": 25958243940480.0,
      "grad_norm": 2.5417301892452278,
      "language_loss": 0.69016963,
      "learning_rate": 1.2333738802215798e-06,
      "loss": 0.71238768,
      "num_input_tokens_seen": 113922670,
      "step": 5293,
      "time_per_iteration": 2.709240436553955
    },
    {
      "auxiliary_loss_clip": 0.01150887,
      "auxiliary_loss_mlp": 0.0076309,
      "balance_loss_clip": 1.04780221,
      "balance_loss_mlp": 1.00094056,
      "epoch": 0.6365658630433476,
      "flos": 20740711011840.0,
      "grad_norm": 2.1813736215796875,
      "language_loss": 0.81211293,
      "learning_rate": 1.2326544657785668e-06,
      "loss": 0.83125269,
      "num_input_tokens_seen": 113942360,
      "step": 5294,
      "time_per_iteration": 3.6566147804260254
    },
    {
      "auxiliary_loss_clip": 0.011843,
      "auxiliary_loss_mlp": 0.01039729,
      "balance_loss_clip": 1.05823731,
      "balance_loss_mlp": 1.03150392,
      "epoch": 0.6366861059339867,
      "flos": 21434146047360.0,
      "grad_norm": 3.7887675070150704,
      "language_loss": 0.75048256,
      "learning_rate": 1.2319351677412608e-06,
      "loss": 0.77272284,
      "num_input_tokens_seen": 113959405,
      "step": 5295,
      "time_per_iteration": 3.585949420928955
    },
    {
      "auxiliary_loss_clip": 0.01184106,
      "auxiliary_loss_mlp": 0.00762791,
      "balance_loss_clip": 1.05650282,
      "balance_loss_mlp": 1.00092387,
      "epoch": 0.6368063488246257,
      "flos": 22267093507200.0,
      "grad_norm": 1.883054491322251,
      "language_loss": 0.7448014,
      "learning_rate": 1.2312159862187796e-06,
      "loss": 0.76427037,
      "num_input_tokens_seen": 113977815,
      "step": 5296,
      "time_per_iteration": 3.5604496002197266
    },
    {
      "auxiliary_loss_clip": 0.01145341,
      "auxiliary_loss_mlp": 0.01040452,
      "balance_loss_clip": 1.04901445,
      "balance_loss_mlp": 1.03158903,
      "epoch": 0.6369265917152649,
      "flos": 22420719976320.0,
      "grad_norm": 1.8728697597555761,
      "language_loss": 0.76584601,
      "learning_rate": 1.2304969213202217e-06,
      "loss": 0.78770393,
      "num_input_tokens_seen": 113999075,
      "step": 5297,
      "time_per_iteration": 2.6513235569000244
    },
    {
      "auxiliary_loss_clip": 0.01161266,
      "auxiliary_loss_mlp": 0.01030055,
      "balance_loss_clip": 1.0520519,
      "balance_loss_mlp": 1.0219965,
      "epoch": 0.6370468346059039,
      "flos": 24718176754560.0,
      "grad_norm": 25.302956836743224,
      "language_loss": 0.79503059,
      "learning_rate": 1.2297779731546692e-06,
      "loss": 0.81694376,
      "num_input_tokens_seen": 114018170,
      "step": 5298,
      "time_per_iteration": 2.652768135070801
    },
    {
      "auxiliary_loss_clip": 0.01181233,
      "auxiliary_loss_mlp": 0.01027464,
      "balance_loss_clip": 1.05572724,
      "balance_loss_mlp": 1.01950943,
      "epoch": 0.637167077496543,
      "flos": 25296589463040.0,
      "grad_norm": 2.0072338353141923,
      "language_loss": 0.78005791,
      "learning_rate": 1.2290591418311853e-06,
      "loss": 0.80214489,
      "num_input_tokens_seen": 114035565,
      "step": 5299,
      "time_per_iteration": 2.607755661010742
    },
    {
      "auxiliary_loss_clip": 0.01182363,
      "auxiliary_loss_mlp": 0.01037412,
      "balance_loss_clip": 1.05374491,
      "balance_loss_mlp": 1.02893615,
      "epoch": 0.637287320387182,
      "flos": 27671110871040.0,
      "grad_norm": 1.8178278543325168,
      "language_loss": 0.72566843,
      "learning_rate": 1.2283404274588172e-06,
      "loss": 0.74786615,
      "num_input_tokens_seen": 114054510,
      "step": 5300,
      "time_per_iteration": 2.6358485221862793
    },
    {
      "auxiliary_loss_clip": 0.01055181,
      "auxiliary_loss_mlp": 0.01003246,
      "balance_loss_clip": 1.01197934,
      "balance_loss_mlp": 1.00228047,
      "epoch": 0.6374075632778212,
      "flos": 63173406873600.0,
      "grad_norm": 0.7425005836259906,
      "language_loss": 0.52736306,
      "learning_rate": 1.227621830146592e-06,
      "loss": 0.54794729,
      "num_input_tokens_seen": 114109875,
      "step": 5301,
      "time_per_iteration": 3.190974235534668
    },
    {
      "auxiliary_loss_clip": 0.01159979,
      "auxiliary_loss_mlp": 0.01027849,
      "balance_loss_clip": 1.05128694,
      "balance_loss_mlp": 1.01953447,
      "epoch": 0.6375278061684603,
      "flos": 25558127366400.0,
      "grad_norm": 1.9591508821606678,
      "language_loss": 0.79457772,
      "learning_rate": 1.2269033500035217e-06,
      "loss": 0.81645602,
      "num_input_tokens_seen": 114130010,
      "step": 5302,
      "time_per_iteration": 2.701634168624878
    },
    {
      "auxiliary_loss_clip": 0.01157349,
      "auxiliary_loss_mlp": 0.01032596,
      "balance_loss_clip": 1.05101275,
      "balance_loss_mlp": 1.02406073,
      "epoch": 0.6376480490590993,
      "flos": 25666362023040.0,
      "grad_norm": 2.2691427925304986,
      "language_loss": 0.73679525,
      "learning_rate": 1.2261849871385988e-06,
      "loss": 0.75869471,
      "num_input_tokens_seen": 114151115,
      "step": 5303,
      "time_per_iteration": 2.738091230392456
    },
    {
      "auxiliary_loss_clip": 0.01169032,
      "auxiliary_loss_mlp": 0.01035834,
      "balance_loss_clip": 1.05165482,
      "balance_loss_mlp": 1.02700698,
      "epoch": 0.6377682919497385,
      "flos": 31537684350720.0,
      "grad_norm": 2.099370534990893,
      "language_loss": 0.63157201,
      "learning_rate": 1.2254667416607972e-06,
      "loss": 0.6536206,
      "num_input_tokens_seen": 114172715,
      "step": 5304,
      "time_per_iteration": 2.714045286178589
    },
    {
      "auxiliary_loss_clip": 0.01145592,
      "auxiliary_loss_mlp": 0.01026945,
      "balance_loss_clip": 1.04703069,
      "balance_loss_mlp": 1.01836205,
      "epoch": 0.6378885348403776,
      "flos": 23039209284480.0,
      "grad_norm": 2.515013648261189,
      "language_loss": 0.83362043,
      "learning_rate": 1.2247486136790756e-06,
      "loss": 0.85534573,
      "num_input_tokens_seen": 114192195,
      "step": 5305,
      "time_per_iteration": 2.697391986846924
    },
    {
      "auxiliary_loss_clip": 0.01181449,
      "auxiliary_loss_mlp": 0.01034425,
      "balance_loss_clip": 1.05513573,
      "balance_loss_mlp": 1.02600276,
      "epoch": 0.6380087777310166,
      "flos": 18697070712960.0,
      "grad_norm": 2.0967113925069527,
      "language_loss": 0.80900609,
      "learning_rate": 1.2240306033023726e-06,
      "loss": 0.83116478,
      "num_input_tokens_seen": 114210020,
      "step": 5306,
      "time_per_iteration": 3.632775068283081
    },
    {
      "auxiliary_loss_clip": 0.01151088,
      "auxiliary_loss_mlp": 0.01029809,
      "balance_loss_clip": 1.05328727,
      "balance_loss_mlp": 1.02171469,
      "epoch": 0.6381290206216558,
      "flos": 23331558078720.0,
      "grad_norm": 1.7094690436377629,
      "language_loss": 0.72028971,
      "learning_rate": 1.223312710639611e-06,
      "loss": 0.74209869,
      "num_input_tokens_seen": 114228740,
      "step": 5307,
      "time_per_iteration": 2.7140674591064453
    },
    {
      "auxiliary_loss_clip": 0.01161372,
      "auxiliary_loss_mlp": 0.01026038,
      "balance_loss_clip": 1.05331278,
      "balance_loss_mlp": 1.0175209,
      "epoch": 0.6382492635122948,
      "flos": 18880466578560.0,
      "grad_norm": 1.888644355215603,
      "language_loss": 0.86992258,
      "learning_rate": 1.2225949357996928e-06,
      "loss": 0.89179671,
      "num_input_tokens_seen": 114246865,
      "step": 5308,
      "time_per_iteration": 2.6560099124908447
    },
    {
      "auxiliary_loss_clip": 0.01180001,
      "auxiliary_loss_mlp": 0.01030942,
      "balance_loss_clip": 1.05440629,
      "balance_loss_mlp": 1.02350914,
      "epoch": 0.6383695064029339,
      "flos": 27819134818560.0,
      "grad_norm": 1.487943415857413,
      "language_loss": 0.80224675,
      "learning_rate": 1.221877278891505e-06,
      "loss": 0.82435614,
      "num_input_tokens_seen": 114266120,
      "step": 5309,
      "time_per_iteration": 2.682337999343872
    },
    {
      "auxiliary_loss_clip": 0.01146566,
      "auxiliary_loss_mlp": 0.01034506,
      "balance_loss_clip": 1.0498867,
      "balance_loss_mlp": 1.02536297,
      "epoch": 0.638489749293573,
      "flos": 26395635853440.0,
      "grad_norm": 3.8947101441520826,
      "language_loss": 0.7144146,
      "learning_rate": 1.221159740023915e-06,
      "loss": 0.73622525,
      "num_input_tokens_seen": 114285950,
      "step": 5310,
      "time_per_iteration": 2.6854991912841797
    },
    {
      "auxiliary_loss_clip": 0.01172853,
      "auxiliary_loss_mlp": 0.01036261,
      "balance_loss_clip": 1.05604124,
      "balance_loss_mlp": 1.02742147,
      "epoch": 0.6386099921842121,
      "flos": 23988328306560.0,
      "grad_norm": 3.7450493418066304,
      "language_loss": 0.72445536,
      "learning_rate": 1.2204423193057735e-06,
      "loss": 0.74654651,
      "num_input_tokens_seen": 114304780,
      "step": 5311,
      "time_per_iteration": 2.6904473304748535
    },
    {
      "auxiliary_loss_clip": 0.01048652,
      "auxiliary_loss_mlp": 0.01001676,
      "balance_loss_clip": 1.01784945,
      "balance_loss_mlp": 1.000669,
      "epoch": 0.6387302350748512,
      "flos": 71731169337600.0,
      "grad_norm": 0.8501460434445461,
      "language_loss": 0.63282222,
      "learning_rate": 1.2197250168459122e-06,
      "loss": 0.6533255,
      "num_input_tokens_seen": 114361180,
      "step": 5312,
      "time_per_iteration": 3.2823963165283203
    },
    {
      "auxiliary_loss_clip": 0.01183069,
      "auxiliary_loss_mlp": 0.01030343,
      "balance_loss_clip": 1.05699503,
      "balance_loss_mlp": 1.02237415,
      "epoch": 0.6388504779654903,
      "flos": 14535778141440.0,
      "grad_norm": 2.0464294954332245,
      "language_loss": 0.7467187,
      "learning_rate": 1.2190078327531454e-06,
      "loss": 0.76885283,
      "num_input_tokens_seen": 114377425,
      "step": 5313,
      "time_per_iteration": 2.5576820373535156
    },
    {
      "auxiliary_loss_clip": 0.0117833,
      "auxiliary_loss_mlp": 0.01029734,
      "balance_loss_clip": 1.05362904,
      "balance_loss_mlp": 1.02211654,
      "epoch": 0.6389707208561294,
      "flos": 22346133384960.0,
      "grad_norm": 1.922260446154256,
      "language_loss": 0.72856712,
      "learning_rate": 1.2182907671362697e-06,
      "loss": 0.75064778,
      "num_input_tokens_seen": 114398120,
      "step": 5314,
      "time_per_iteration": 2.6378185749053955
    },
    {
      "auxiliary_loss_clip": 0.01168719,
      "auxiliary_loss_mlp": 0.01028598,
      "balance_loss_clip": 1.05503845,
      "balance_loss_mlp": 1.01997328,
      "epoch": 0.6390909637467684,
      "flos": 19426883247360.0,
      "grad_norm": 2.2265932029037074,
      "language_loss": 0.79192519,
      "learning_rate": 1.2175738201040626e-06,
      "loss": 0.81389844,
      "num_input_tokens_seen": 114415160,
      "step": 5315,
      "time_per_iteration": 2.590496301651001
    },
    {
      "auxiliary_loss_clip": 0.01181013,
      "auxiliary_loss_mlp": 0.01038093,
      "balance_loss_clip": 1.05307257,
      "balance_loss_mlp": 1.02877641,
      "epoch": 0.6392112066374076,
      "flos": 24090852700800.0,
      "grad_norm": 2.1003433198993977,
      "language_loss": 0.79010254,
      "learning_rate": 1.2168569917652855e-06,
      "loss": 0.81229359,
      "num_input_tokens_seen": 114435015,
      "step": 5316,
      "time_per_iteration": 2.6778011322021484
    },
    {
      "auxiliary_loss_clip": 0.01142687,
      "auxiliary_loss_mlp": 0.01030568,
      "balance_loss_clip": 1.04827607,
      "balance_loss_mlp": 1.02216136,
      "epoch": 0.6393314495280467,
      "flos": 26795141896320.0,
      "grad_norm": 1.615545884542252,
      "language_loss": 0.63662732,
      "learning_rate": 1.2161402822286797e-06,
      "loss": 0.65835983,
      "num_input_tokens_seen": 114455700,
      "step": 5317,
      "time_per_iteration": 2.728487014770508
    },
    {
      "auxiliary_loss_clip": 0.01184824,
      "auxiliary_loss_mlp": 0.0102536,
      "balance_loss_clip": 1.05705559,
      "balance_loss_mlp": 1.0170542,
      "epoch": 0.6394516924186857,
      "flos": 20260692633600.0,
      "grad_norm": 1.9282523508378295,
      "language_loss": 0.79226947,
      "learning_rate": 1.2154236916029703e-06,
      "loss": 0.81437123,
      "num_input_tokens_seen": 114473675,
      "step": 5318,
      "time_per_iteration": 2.608725070953369
    },
    {
      "auxiliary_loss_clip": 0.01161961,
      "auxiliary_loss_mlp": 0.01030076,
      "balance_loss_clip": 1.0491693,
      "balance_loss_mlp": 1.02173686,
      "epoch": 0.6395719353093249,
      "flos": 18368847210240.0,
      "grad_norm": 2.4795708620391146,
      "language_loss": 0.73865771,
      "learning_rate": 1.2147072199968627e-06,
      "loss": 0.76057804,
      "num_input_tokens_seen": 114492310,
      "step": 5319,
      "time_per_iteration": 2.6693923473358154
    },
    {
      "auxiliary_loss_clip": 0.01153692,
      "auxiliary_loss_mlp": 0.01023447,
      "balance_loss_clip": 1.05173945,
      "balance_loss_mlp": 1.0158298,
      "epoch": 0.6396921781999639,
      "flos": 17566315591680.0,
      "grad_norm": 1.9679534281007067,
      "language_loss": 0.72001702,
      "learning_rate": 1.2139908675190454e-06,
      "loss": 0.74178839,
      "num_input_tokens_seen": 114511520,
      "step": 5320,
      "time_per_iteration": 3.5750622749328613
    },
    {
      "auxiliary_loss_clip": 0.01157463,
      "auxiliary_loss_mlp": 0.0102854,
      "balance_loss_clip": 1.04923975,
      "balance_loss_mlp": 1.01990294,
      "epoch": 0.639812421090603,
      "flos": 21251252972160.0,
      "grad_norm": 2.51685933188557,
      "language_loss": 0.75587714,
      "learning_rate": 1.2132746342781883e-06,
      "loss": 0.77773714,
      "num_input_tokens_seen": 114532680,
      "step": 5321,
      "time_per_iteration": 3.6293604373931885
    },
    {
      "auxiliary_loss_clip": 0.0116907,
      "auxiliary_loss_mlp": 0.01030365,
      "balance_loss_clip": 1.05151522,
      "balance_loss_mlp": 1.02220559,
      "epoch": 0.6399326639812422,
      "flos": 11180967684480.0,
      "grad_norm": 2.8685414546905417,
      "language_loss": 0.80278563,
      "learning_rate": 1.2125585203829442e-06,
      "loss": 0.82477999,
      "num_input_tokens_seen": 114548320,
      "step": 5322,
      "time_per_iteration": 2.6208980083465576
    },
    {
      "auxiliary_loss_clip": 0.01183543,
      "auxiliary_loss_mlp": 0.01030639,
      "balance_loss_clip": 1.05531883,
      "balance_loss_mlp": 1.02227616,
      "epoch": 0.6400529068718812,
      "flos": 23911048195200.0,
      "grad_norm": 1.7346663372125497,
      "language_loss": 0.7393415,
      "learning_rate": 1.211842525941946e-06,
      "loss": 0.76148331,
      "num_input_tokens_seen": 114568115,
      "step": 5323,
      "time_per_iteration": 3.5616652965545654
    },
    {
      "auxiliary_loss_clip": 0.01165042,
      "auxiliary_loss_mlp": 0.01030219,
      "balance_loss_clip": 1.05264354,
      "balance_loss_mlp": 1.02244687,
      "epoch": 0.6401731497625203,
      "flos": 44018724890880.0,
      "grad_norm": 3.0153587289273016,
      "language_loss": 0.79369038,
      "learning_rate": 1.2111266510638105e-06,
      "loss": 0.81564301,
      "num_input_tokens_seen": 114591040,
      "step": 5324,
      "time_per_iteration": 2.8620336055755615
    },
    {
      "auxiliary_loss_clip": 0.01154621,
      "auxiliary_loss_mlp": 0.01027321,
      "balance_loss_clip": 1.05458069,
      "balance_loss_mlp": 1.01890492,
      "epoch": 0.6402933926531594,
      "flos": 20662209838080.0,
      "grad_norm": 2.2871103539060433,
      "language_loss": 0.79877883,
      "learning_rate": 1.2104108958571346e-06,
      "loss": 0.82059824,
      "num_input_tokens_seen": 114609310,
      "step": 5325,
      "time_per_iteration": 2.71982741355896
    },
    {
      "auxiliary_loss_clip": 0.01155509,
      "auxiliary_loss_mlp": 0.01026354,
      "balance_loss_clip": 1.05316341,
      "balance_loss_mlp": 1.01843262,
      "epoch": 0.6404136355437985,
      "flos": 24863327614080.0,
      "grad_norm": 1.6819118922693461,
      "language_loss": 0.75793993,
      "learning_rate": 1.2096952604304975e-06,
      "loss": 0.77975857,
      "num_input_tokens_seen": 114629740,
      "step": 5326,
      "time_per_iteration": 2.7172040939331055
    },
    {
      "auxiliary_loss_clip": 0.01169143,
      "auxiliary_loss_mlp": 0.01026479,
      "balance_loss_clip": 1.05063105,
      "balance_loss_mlp": 1.01753807,
      "epoch": 0.6405338784344375,
      "flos": 40479548901120.0,
      "grad_norm": 2.3353379486860955,
      "language_loss": 0.70813841,
      "learning_rate": 1.2089797448924616e-06,
      "loss": 0.73009467,
      "num_input_tokens_seen": 114653615,
      "step": 5327,
      "time_per_iteration": 2.837947368621826
    },
    {
      "auxiliary_loss_clip": 0.01156077,
      "auxiliary_loss_mlp": 0.01028392,
      "balance_loss_clip": 1.05300522,
      "balance_loss_mlp": 1.01958871,
      "epoch": 0.6406541213250767,
      "flos": 20886041439360.0,
      "grad_norm": 2.0626133127672013,
      "language_loss": 0.66419661,
      "learning_rate": 1.2082643493515692e-06,
      "loss": 0.68604124,
      "num_input_tokens_seen": 114671935,
      "step": 5328,
      "time_per_iteration": 2.74971079826355
    },
    {
      "auxiliary_loss_clip": 0.01181182,
      "auxiliary_loss_mlp": 0.01027722,
      "balance_loss_clip": 1.05440784,
      "balance_loss_mlp": 1.01958013,
      "epoch": 0.6407743642157158,
      "flos": 23295970679040.0,
      "grad_norm": 1.8259224990920686,
      "language_loss": 0.82114351,
      "learning_rate": 1.207549073916346e-06,
      "loss": 0.84323251,
      "num_input_tokens_seen": 114692870,
      "step": 5329,
      "time_per_iteration": 2.65494704246521
    },
    {
      "auxiliary_loss_clip": 0.01184251,
      "auxiliary_loss_mlp": 0.01026387,
      "balance_loss_clip": 1.05719221,
      "balance_loss_mlp": 1.01904356,
      "epoch": 0.6408946071063548,
      "flos": 15012636122880.0,
      "grad_norm": 2.3198149532560595,
      "language_loss": 0.78091431,
      "learning_rate": 1.2068339186952976e-06,
      "loss": 0.80302072,
      "num_input_tokens_seen": 114710410,
      "step": 5330,
      "time_per_iteration": 2.6525723934173584
    },
    {
      "auxiliary_loss_clip": 0.01163512,
      "auxiliary_loss_mlp": 0.01029828,
      "balance_loss_clip": 1.05252981,
      "balance_loss_mlp": 1.02100098,
      "epoch": 0.6410148499969939,
      "flos": 22528595496960.0,
      "grad_norm": 2.608308234782063,
      "language_loss": 0.73586184,
      "learning_rate": 1.2061188837969136e-06,
      "loss": 0.75779521,
      "num_input_tokens_seen": 114730020,
      "step": 5331,
      "time_per_iteration": 2.739917516708374
    },
    {
      "auxiliary_loss_clip": 0.01182261,
      "auxiliary_loss_mlp": 0.01026146,
      "balance_loss_clip": 1.05242276,
      "balance_loss_mlp": 1.01761103,
      "epoch": 0.641135092887633,
      "flos": 12422004537600.0,
      "grad_norm": 2.940635885431581,
      "language_loss": 0.84102535,
      "learning_rate": 1.2054039693296631e-06,
      "loss": 0.86310947,
      "num_input_tokens_seen": 114748015,
      "step": 5332,
      "time_per_iteration": 3.466233015060425
    },
    {
      "auxiliary_loss_clip": 0.01179856,
      "auxiliary_loss_mlp": 0.01034088,
      "balance_loss_clip": 1.05477655,
      "balance_loss_mlp": 1.02577949,
      "epoch": 0.6412553357782721,
      "flos": 22127329687680.0,
      "grad_norm": 2.139289052725836,
      "language_loss": 0.8167448,
      "learning_rate": 1.2046891754019992e-06,
      "loss": 0.83888423,
      "num_input_tokens_seen": 114768625,
      "step": 5333,
      "time_per_iteration": 2.7162973880767822
    },
    {
      "auxiliary_loss_clip": 0.01180792,
      "auxiliary_loss_mlp": 0.01028598,
      "balance_loss_clip": 1.05556107,
      "balance_loss_mlp": 1.02003896,
      "epoch": 0.6413755786689112,
      "flos": 15888605097600.0,
      "grad_norm": 2.951037542058698,
      "language_loss": 0.82731354,
      "learning_rate": 1.2039745021223548e-06,
      "loss": 0.84940743,
      "num_input_tokens_seen": 114786045,
      "step": 5334,
      "time_per_iteration": 2.5498058795928955
    },
    {
      "auxiliary_loss_clip": 0.01054933,
      "auxiliary_loss_mlp": 0.01003377,
      "balance_loss_clip": 1.0086658,
      "balance_loss_mlp": 1.00242925,
      "epoch": 0.6414958215595503,
      "flos": 68039159955840.0,
      "grad_norm": 0.7884439637613325,
      "language_loss": 0.56932902,
      "learning_rate": 1.2032599495991456e-06,
      "loss": 0.58991218,
      "num_input_tokens_seen": 114850785,
      "step": 5335,
      "time_per_iteration": 3.353733539581299
    },
    {
      "auxiliary_loss_clip": 0.01173281,
      "auxiliary_loss_mlp": 0.0103531,
      "balance_loss_clip": 1.05380321,
      "balance_loss_mlp": 1.02649462,
      "epoch": 0.6416160644501894,
      "flos": 44091300320640.0,
      "grad_norm": 1.9257201571687161,
      "language_loss": 0.69908643,
      "learning_rate": 1.2025455179407685e-06,
      "loss": 0.72117233,
      "num_input_tokens_seen": 114871945,
      "step": 5336,
      "time_per_iteration": 2.879931926727295
    },
    {
      "auxiliary_loss_clip": 0.01170144,
      "auxiliary_loss_mlp": 0.0102888,
      "balance_loss_clip": 1.05371547,
      "balance_loss_mlp": 1.02050567,
      "epoch": 0.6417363073408284,
      "flos": 20959837931520.0,
      "grad_norm": 2.0904299602175773,
      "language_loss": 0.74099344,
      "learning_rate": 1.2018312072556022e-06,
      "loss": 0.76298368,
      "num_input_tokens_seen": 114890445,
      "step": 5337,
      "time_per_iteration": 2.6591005325317383
    },
    {
      "auxiliary_loss_clip": 0.01154556,
      "auxiliary_loss_mlp": 0.01026512,
      "balance_loss_clip": 1.0506382,
      "balance_loss_mlp": 1.01798213,
      "epoch": 0.6418565502314676,
      "flos": 22455122227200.0,
      "grad_norm": 1.9544672198872193,
      "language_loss": 0.749524,
      "learning_rate": 1.2011170176520077e-06,
      "loss": 0.77133471,
      "num_input_tokens_seen": 114911360,
      "step": 5338,
      "time_per_iteration": 2.7603189945220947
    },
    {
      "auxiliary_loss_clip": 0.01177841,
      "auxiliary_loss_mlp": 0.01023502,
      "balance_loss_clip": 1.05300546,
      "balance_loss_mlp": 1.01536584,
      "epoch": 0.6419767931221066,
      "flos": 25045502417280.0,
      "grad_norm": 1.6660426032493882,
      "language_loss": 0.81440783,
      "learning_rate": 1.2004029492383256e-06,
      "loss": 0.83642125,
      "num_input_tokens_seen": 114932700,
      "step": 5339,
      "time_per_iteration": 2.636061429977417
    },
    {
      "auxiliary_loss_clip": 0.01153277,
      "auxiliary_loss_mlp": 0.01031332,
      "balance_loss_clip": 1.05229783,
      "balance_loss_mlp": 1.02305889,
      "epoch": 0.6420970360127457,
      "flos": 19463691709440.0,
      "grad_norm": 1.878830006035123,
      "language_loss": 0.73891699,
      "learning_rate": 1.1996890021228814e-06,
      "loss": 0.76076305,
      "num_input_tokens_seen": 114949475,
      "step": 5340,
      "time_per_iteration": 2.6946747303009033
    },
    {
      "auxiliary_loss_clip": 0.01102215,
      "auxiliary_loss_mlp": 0.01027912,
      "balance_loss_clip": 1.04033768,
      "balance_loss_mlp": 1.01931441,
      "epoch": 0.6422172789033849,
      "flos": 40406147458560.0,
      "grad_norm": 1.523125565536143,
      "language_loss": 0.70335221,
      "learning_rate": 1.1989751764139785e-06,
      "loss": 0.72465348,
      "num_input_tokens_seen": 114973125,
      "step": 5341,
      "time_per_iteration": 2.900400161743164
    },
    {
      "auxiliary_loss_clip": 0.01181895,
      "auxiliary_loss_mlp": 0.01027587,
      "balance_loss_clip": 1.05247581,
      "balance_loss_mlp": 1.01879537,
      "epoch": 0.6423375217940239,
      "flos": 27672870637440.0,
      "grad_norm": 1.6522765779228221,
      "language_loss": 0.83423376,
      "learning_rate": 1.1982614722199044e-06,
      "loss": 0.85632861,
      "num_input_tokens_seen": 114994300,
      "step": 5342,
      "time_per_iteration": 2.6824162006378174
    },
    {
      "auxiliary_loss_clip": 0.01165025,
      "auxiliary_loss_mlp": 0.01023985,
      "balance_loss_clip": 1.05297613,
      "balance_loss_mlp": 1.01594758,
      "epoch": 0.642457764684663,
      "flos": 18369242259840.0,
      "grad_norm": 2.062972227426197,
      "language_loss": 0.78025419,
      "learning_rate": 1.1975478896489276e-06,
      "loss": 0.80214429,
      "num_input_tokens_seen": 115012135,
      "step": 5343,
      "time_per_iteration": 2.671083450317383
    },
    {
      "auxiliary_loss_clip": 0.01176831,
      "auxiliary_loss_mlp": 0.01032763,
      "balance_loss_clip": 1.05167544,
      "balance_loss_mlp": 1.02488959,
      "epoch": 0.6425780075753021,
      "flos": 19750509809280.0,
      "grad_norm": 2.1075736778208,
      "language_loss": 0.7623952,
      "learning_rate": 1.1968344288092981e-06,
      "loss": 0.78449106,
      "num_input_tokens_seen": 115028715,
      "step": 5344,
      "time_per_iteration": 2.6051716804504395
    },
    {
      "auxiliary_loss_clip": 0.0116619,
      "auxiliary_loss_mlp": 0.0102754,
      "balance_loss_clip": 1.05424607,
      "balance_loss_mlp": 1.01912379,
      "epoch": 0.6426982504659412,
      "flos": 20558536208640.0,
      "grad_norm": 2.943008091176807,
      "language_loss": 0.6468696,
      "learning_rate": 1.1961210898092468e-06,
      "loss": 0.66880685,
      "num_input_tokens_seen": 115047665,
      "step": 5345,
      "time_per_iteration": 2.642268657684326
    },
    {
      "auxiliary_loss_clip": 0.01147676,
      "auxiliary_loss_mlp": 0.01035252,
      "balance_loss_clip": 1.05106139,
      "balance_loss_mlp": 1.02668715,
      "epoch": 0.6428184933565803,
      "flos": 17851984456320.0,
      "grad_norm": 2.258529487097124,
      "language_loss": 0.79299629,
      "learning_rate": 1.1954078727569874e-06,
      "loss": 0.81482553,
      "num_input_tokens_seen": 115064965,
      "step": 5346,
      "time_per_iteration": 3.586683750152588
    },
    {
      "auxiliary_loss_clip": 0.01182155,
      "auxiliary_loss_mlp": 0.01027862,
      "balance_loss_clip": 1.05510187,
      "balance_loss_mlp": 1.01937389,
      "epoch": 0.6429387362472194,
      "flos": 22456953820800.0,
      "grad_norm": 1.6523320494529252,
      "language_loss": 0.7766912,
      "learning_rate": 1.1946947777607141e-06,
      "loss": 0.79879135,
      "num_input_tokens_seen": 115086100,
      "step": 5347,
      "time_per_iteration": 3.602169990539551
    },
    {
      "auxiliary_loss_clip": 0.01168683,
      "auxiliary_loss_mlp": 0.01026847,
      "balance_loss_clip": 1.05445313,
      "balance_loss_mlp": 1.01868725,
      "epoch": 0.6430589791378585,
      "flos": 24752579005440.0,
      "grad_norm": 2.0275771670609424,
      "language_loss": 0.80221546,
      "learning_rate": 1.1939818049286024e-06,
      "loss": 0.82417083,
      "num_input_tokens_seen": 115104260,
      "step": 5348,
      "time_per_iteration": 2.659499168395996
    },
    {
      "auxiliary_loss_clip": 0.01180851,
      "auxiliary_loss_mlp": 0.01033192,
      "balance_loss_clip": 1.05444622,
      "balance_loss_mlp": 1.02490664,
      "epoch": 0.6431792220284975,
      "flos": 24901249397760.0,
      "grad_norm": 9.559797372533401,
      "language_loss": 0.75551939,
      "learning_rate": 1.1932689543688101e-06,
      "loss": 0.77765983,
      "num_input_tokens_seen": 115125365,
      "step": 5349,
      "time_per_iteration": 3.598194122314453
    },
    {
      "auxiliary_loss_clip": 0.01151453,
      "auxiliary_loss_mlp": 0.01032384,
      "balance_loss_clip": 1.04866576,
      "balance_loss_mlp": 1.02359247,
      "epoch": 0.6432994649191367,
      "flos": 21032305620480.0,
      "grad_norm": 2.067114958172618,
      "language_loss": 0.72853786,
      "learning_rate": 1.1925562261894756e-06,
      "loss": 0.75037622,
      "num_input_tokens_seen": 115144445,
      "step": 5350,
      "time_per_iteration": 2.6865687370300293
    },
    {
      "auxiliary_loss_clip": 0.0117908,
      "auxiliary_loss_mlp": 0.01026265,
      "balance_loss_clip": 1.05472422,
      "balance_loss_mlp": 1.01825392,
      "epoch": 0.6434197078097758,
      "flos": 30884433655680.0,
      "grad_norm": 1.8928802677767835,
      "language_loss": 0.77763557,
      "learning_rate": 1.1918436204987207e-06,
      "loss": 0.79968899,
      "num_input_tokens_seen": 115166305,
      "step": 5351,
      "time_per_iteration": 2.6705563068389893
    },
    {
      "auxiliary_loss_clip": 0.01146854,
      "auxiliary_loss_mlp": 0.01025198,
      "balance_loss_clip": 1.04920113,
      "balance_loss_mlp": 1.01700795,
      "epoch": 0.6435399507004148,
      "flos": 15012492468480.0,
      "grad_norm": 3.771610418804161,
      "language_loss": 0.82218444,
      "learning_rate": 1.191131137404645e-06,
      "loss": 0.84390497,
      "num_input_tokens_seen": 115183045,
      "step": 5352,
      "time_per_iteration": 2.662318706512451
    },
    {
      "auxiliary_loss_clip": 0.01183495,
      "auxiliary_loss_mlp": 0.01033179,
      "balance_loss_clip": 1.05587125,
      "balance_loss_mlp": 1.02457213,
      "epoch": 0.643660193591054,
      "flos": 19901981462400.0,
      "grad_norm": 2.3305825774745017,
      "language_loss": 0.77373135,
      "learning_rate": 1.190418777015333e-06,
      "loss": 0.79589808,
      "num_input_tokens_seen": 115201955,
      "step": 5353,
      "time_per_iteration": 2.576148271560669
    },
    {
      "auxiliary_loss_clip": 0.01154851,
      "auxiliary_loss_mlp": 0.01027029,
      "balance_loss_clip": 1.0477711,
      "balance_loss_mlp": 1.01873231,
      "epoch": 0.643780436481693,
      "flos": 24133622820480.0,
      "grad_norm": 1.526022371918666,
      "language_loss": 0.73885131,
      "learning_rate": 1.1897065394388487e-06,
      "loss": 0.76067013,
      "num_input_tokens_seen": 115222395,
      "step": 5354,
      "time_per_iteration": 2.676348924636841
    },
    {
      "auxiliary_loss_clip": 0.01184899,
      "auxiliary_loss_mlp": 0.0103966,
      "balance_loss_clip": 1.05813789,
      "balance_loss_mlp": 1.03163707,
      "epoch": 0.6439006793723321,
      "flos": 23148808657920.0,
      "grad_norm": 1.902753718349394,
      "language_loss": 0.76875645,
      "learning_rate": 1.1889944247832385e-06,
      "loss": 0.79100204,
      "num_input_tokens_seen": 115242635,
      "step": 5355,
      "time_per_iteration": 2.5790176391601562
    },
    {
      "auxiliary_loss_clip": 0.01178291,
      "auxiliary_loss_mlp": 0.01030134,
      "balance_loss_clip": 1.05253601,
      "balance_loss_mlp": 1.02139568,
      "epoch": 0.6440209222629713,
      "flos": 23617909301760.0,
      "grad_norm": 2.6619226624567793,
      "language_loss": 0.71193004,
      "learning_rate": 1.1882824331565283e-06,
      "loss": 0.73401421,
      "num_input_tokens_seen": 115262095,
      "step": 5356,
      "time_per_iteration": 2.667201280593872
    },
    {
      "auxiliary_loss_clip": 0.01165795,
      "auxiliary_loss_mlp": 0.01025161,
      "balance_loss_clip": 1.05451179,
      "balance_loss_mlp": 1.01756191,
      "epoch": 0.6441411651536103,
      "flos": 16544872535040.0,
      "grad_norm": 2.5100053988391084,
      "language_loss": 0.89191294,
      "learning_rate": 1.1875705646667287e-06,
      "loss": 0.91382253,
      "num_input_tokens_seen": 115279985,
      "step": 5357,
      "time_per_iteration": 2.61251163482666
    },
    {
      "auxiliary_loss_clip": 0.01178799,
      "auxiliary_loss_mlp": 0.01027961,
      "balance_loss_clip": 1.05241883,
      "balance_loss_mlp": 1.01981878,
      "epoch": 0.6442614080442494,
      "flos": 25410965345280.0,
      "grad_norm": 2.727076804677742,
      "language_loss": 0.75627494,
      "learning_rate": 1.1868588194218282e-06,
      "loss": 0.77834255,
      "num_input_tokens_seen": 115300365,
      "step": 5358,
      "time_per_iteration": 3.588226318359375
    },
    {
      "auxiliary_loss_clip": 0.01180043,
      "auxiliary_loss_mlp": 0.01033412,
      "balance_loss_clip": 1.05251634,
      "balance_loss_mlp": 1.02540112,
      "epoch": 0.6443816509348885,
      "flos": 28294017552000.0,
      "grad_norm": 1.664000120947257,
      "language_loss": 0.74438,
      "learning_rate": 1.1861471975297979e-06,
      "loss": 0.76651454,
      "num_input_tokens_seen": 115322060,
      "step": 5359,
      "time_per_iteration": 2.749664306640625
    },
    {
      "auxiliary_loss_clip": 0.01181254,
      "auxiliary_loss_mlp": 0.01033924,
      "balance_loss_clip": 1.05329406,
      "balance_loss_mlp": 1.02527547,
      "epoch": 0.6445018938255276,
      "flos": 36690075964800.0,
      "grad_norm": 2.0252654293591714,
      "language_loss": 0.71235704,
      "learning_rate": 1.185435699098591e-06,
      "loss": 0.73450887,
      "num_input_tokens_seen": 115348255,
      "step": 5360,
      "time_per_iteration": 2.727743625640869
    },
    {
      "auxiliary_loss_clip": 0.01182259,
      "auxiliary_loss_mlp": 0.01031939,
      "balance_loss_clip": 1.05436778,
      "balance_loss_mlp": 1.02405298,
      "epoch": 0.6446221367161666,
      "flos": 14501411804160.0,
      "grad_norm": 2.741086900301432,
      "language_loss": 0.78468704,
      "learning_rate": 1.1847243242361403e-06,
      "loss": 0.80682898,
      "num_input_tokens_seen": 115366845,
      "step": 5361,
      "time_per_iteration": 2.6831722259521484
    },
    {
      "auxiliary_loss_clip": 0.01179381,
      "auxiliary_loss_mlp": 0.01023882,
      "balance_loss_clip": 1.05247211,
      "balance_loss_mlp": 1.01568651,
      "epoch": 0.6447423796068057,
      "flos": 24609367480320.0,
      "grad_norm": 1.5695285541433912,
      "language_loss": 0.78211939,
      "learning_rate": 1.1840130730503624e-06,
      "loss": 0.80415213,
      "num_input_tokens_seen": 115388125,
      "step": 5362,
      "time_per_iteration": 2.6591527462005615
    },
    {
      "auxiliary_loss_clip": 0.01151047,
      "auxiliary_loss_mlp": 0.01025456,
      "balance_loss_clip": 1.04649854,
      "balance_loss_mlp": 1.0173974,
      "epoch": 0.6448626224974449,
      "flos": 25047298097280.0,
      "grad_norm": 1.7515464230202766,
      "language_loss": 0.75262147,
      "learning_rate": 1.1833019456491518e-06,
      "loss": 0.77438653,
      "num_input_tokens_seen": 115409655,
      "step": 5363,
      "time_per_iteration": 2.8042638301849365
    },
    {
      "auxiliary_loss_clip": 0.01165874,
      "auxiliary_loss_mlp": 0.01029305,
      "balance_loss_clip": 1.05347419,
      "balance_loss_mlp": 1.0215745,
      "epoch": 0.6449828653880839,
      "flos": 22530355263360.0,
      "grad_norm": 2.1393427209540365,
      "language_loss": 0.79515326,
      "learning_rate": 1.1825909421403871e-06,
      "loss": 0.81710505,
      "num_input_tokens_seen": 115428750,
      "step": 5364,
      "time_per_iteration": 2.6356210708618164
    },
    {
      "auxiliary_loss_clip": 0.01148112,
      "auxiliary_loss_mlp": 0.01028498,
      "balance_loss_clip": 1.04910398,
      "balance_loss_mlp": 1.02054429,
      "epoch": 0.645103108278723,
      "flos": 25695736369920.0,
      "grad_norm": 5.244261544461515,
      "language_loss": 0.76343012,
      "learning_rate": 1.181880062631926e-06,
      "loss": 0.7851963,
      "num_input_tokens_seen": 115448085,
      "step": 5365,
      "time_per_iteration": 2.6895272731781006
    },
    {
      "auxiliary_loss_clip": 0.01182821,
      "auxiliary_loss_mlp": 0.01028737,
      "balance_loss_clip": 1.05514979,
      "balance_loss_mlp": 1.01999331,
      "epoch": 0.6452233511693621,
      "flos": 27450331925760.0,
      "grad_norm": 2.3122185619680278,
      "language_loss": 0.85031104,
      "learning_rate": 1.1811693072316093e-06,
      "loss": 0.87242663,
      "num_input_tokens_seen": 115465765,
      "step": 5366,
      "time_per_iteration": 2.5708093643188477
    },
    {
      "auxiliary_loss_clip": 0.01179732,
      "auxiliary_loss_mlp": 0.0102568,
      "balance_loss_clip": 1.05315006,
      "balance_loss_mlp": 1.01784253,
      "epoch": 0.6453435940600012,
      "flos": 19208618254080.0,
      "grad_norm": 2.215716873663332,
      "language_loss": 0.84082556,
      "learning_rate": 1.1804586760472574e-06,
      "loss": 0.86287963,
      "num_input_tokens_seen": 115482230,
      "step": 5367,
      "time_per_iteration": 2.63185715675354
    },
    {
      "auxiliary_loss_clip": 0.01160979,
      "auxiliary_loss_mlp": 0.01034918,
      "balance_loss_clip": 1.04778516,
      "balance_loss_mlp": 1.0267942,
      "epoch": 0.6454638369506402,
      "flos": 25737680476800.0,
      "grad_norm": 2.555283200983936,
      "language_loss": 0.80472887,
      "learning_rate": 1.1797481691866736e-06,
      "loss": 0.82668781,
      "num_input_tokens_seen": 115499455,
      "step": 5368,
      "time_per_iteration": 2.6721980571746826
    },
    {
      "auxiliary_loss_clip": 0.01135391,
      "auxiliary_loss_mlp": 0.01028573,
      "balance_loss_clip": 1.04977369,
      "balance_loss_mlp": 1.02072358,
      "epoch": 0.6455840798412794,
      "flos": 20989176364800.0,
      "grad_norm": 2.488975337876172,
      "language_loss": 0.8342061,
      "learning_rate": 1.1790377867576393e-06,
      "loss": 0.85584581,
      "num_input_tokens_seen": 115517205,
      "step": 5369,
      "time_per_iteration": 2.742781400680542
    },
    {
      "auxiliary_loss_clip": 0.01163511,
      "auxiliary_loss_mlp": 0.0103331,
      "balance_loss_clip": 1.04995608,
      "balance_loss_mlp": 1.02494156,
      "epoch": 0.6457043227319185,
      "flos": 26067556005120.0,
      "grad_norm": 1.8878601497848226,
      "language_loss": 0.76712298,
      "learning_rate": 1.1783275288679203e-06,
      "loss": 0.78909123,
      "num_input_tokens_seen": 115534370,
      "step": 5370,
      "time_per_iteration": 2.6792750358581543
    },
    {
      "auxiliary_loss_clip": 0.01057374,
      "auxiliary_loss_mlp": 0.01002423,
      "balance_loss_clip": 1.00846314,
      "balance_loss_mlp": 1.00138569,
      "epoch": 0.6458245656225575,
      "flos": 60370831088640.0,
      "grad_norm": 0.8519227777915507,
      "language_loss": 0.57046014,
      "learning_rate": 1.177617395625262e-06,
      "loss": 0.59105808,
      "num_input_tokens_seen": 115592345,
      "step": 5371,
      "time_per_iteration": 3.182633876800537
    },
    {
      "auxiliary_loss_clip": 0.01151154,
      "auxiliary_loss_mlp": 0.010239,
      "balance_loss_clip": 1.04862356,
      "balance_loss_mlp": 1.01586556,
      "epoch": 0.6459448085131967,
      "flos": 23076771932160.0,
      "grad_norm": 1.9736442050403054,
      "language_loss": 0.75548154,
      "learning_rate": 1.1769073871373908e-06,
      "loss": 0.77723211,
      "num_input_tokens_seen": 115612550,
      "step": 5372,
      "time_per_iteration": 2.6781556606292725
    },
    {
      "auxiliary_loss_clip": 0.01165574,
      "auxiliary_loss_mlp": 0.01031077,
      "balance_loss_clip": 1.05090678,
      "balance_loss_mlp": 1.02226102,
      "epoch": 0.6460650514038357,
      "flos": 22598190097920.0,
      "grad_norm": 1.8978299505480016,
      "language_loss": 0.84103286,
      "learning_rate": 1.176197503512015e-06,
      "loss": 0.86299932,
      "num_input_tokens_seen": 115632265,
      "step": 5373,
      "time_per_iteration": 4.583801984786987
    },
    {
      "auxiliary_loss_clip": 0.01167945,
      "auxiliary_loss_mlp": 0.01026882,
      "balance_loss_clip": 1.05649495,
      "balance_loss_mlp": 1.01944971,
      "epoch": 0.6461852942944748,
      "flos": 20266726118400.0,
      "grad_norm": 2.4334067607106946,
      "language_loss": 0.82842588,
      "learning_rate": 1.1754877448568223e-06,
      "loss": 0.8503741,
      "num_input_tokens_seen": 115651720,
      "step": 5374,
      "time_per_iteration": 2.6252553462982178
    },
    {
      "auxiliary_loss_clip": 0.0117807,
      "auxiliary_loss_mlp": 0.01029094,
      "balance_loss_clip": 1.0534389,
      "balance_loss_mlp": 1.02171469,
      "epoch": 0.646305537185114,
      "flos": 23367109564800.0,
      "grad_norm": 3.5581368947148637,
      "language_loss": 0.90076756,
      "learning_rate": 1.1747781112794837e-06,
      "loss": 0.92283928,
      "num_input_tokens_seen": 115668215,
      "step": 5375,
      "time_per_iteration": 3.533332109451294
    },
    {
      "auxiliary_loss_clip": 0.01158742,
      "auxiliary_loss_mlp": 0.01028729,
      "balance_loss_clip": 1.05224991,
      "balance_loss_mlp": 1.02105808,
      "epoch": 0.646425780075753,
      "flos": 24277480790400.0,
      "grad_norm": 1.6352844582404313,
      "language_loss": 0.83034164,
      "learning_rate": 1.1740686028876487e-06,
      "loss": 0.85221636,
      "num_input_tokens_seen": 115687080,
      "step": 5376,
      "time_per_iteration": 2.687899351119995
    },
    {
      "auxiliary_loss_clip": 0.01145978,
      "auxiliary_loss_mlp": 0.01024268,
      "balance_loss_clip": 1.05023098,
      "balance_loss_mlp": 1.01638544,
      "epoch": 0.6465460229663921,
      "flos": 20813968800000.0,
      "grad_norm": 4.697328641827072,
      "language_loss": 0.7501775,
      "learning_rate": 1.1733592197889507e-06,
      "loss": 0.77188003,
      "num_input_tokens_seen": 115703990,
      "step": 5377,
      "time_per_iteration": 2.6653285026550293
    },
    {
      "auxiliary_loss_clip": 0.01166943,
      "auxiliary_loss_mlp": 0.01033776,
      "balance_loss_clip": 1.05314565,
      "balance_loss_mlp": 1.02622354,
      "epoch": 0.6466662658570312,
      "flos": 22853299466880.0,
      "grad_norm": 2.0432207424533444,
      "language_loss": 0.72755545,
      "learning_rate": 1.1726499620910014e-06,
      "loss": 0.74956262,
      "num_input_tokens_seen": 115724270,
      "step": 5378,
      "time_per_iteration": 2.6423237323760986
    },
    {
      "auxiliary_loss_clip": 0.01178376,
      "auxiliary_loss_mlp": 0.01031737,
      "balance_loss_clip": 1.05251837,
      "balance_loss_mlp": 1.02333856,
      "epoch": 0.6467865087476703,
      "flos": 15304553953920.0,
      "grad_norm": 2.883638101354063,
      "language_loss": 0.77721661,
      "learning_rate": 1.1719408299013955e-06,
      "loss": 0.79931772,
      "num_input_tokens_seen": 115742995,
      "step": 5379,
      "time_per_iteration": 2.6160542964935303
    },
    {
      "auxiliary_loss_clip": 0.01177339,
      "auxiliary_loss_mlp": 0.0102775,
      "balance_loss_clip": 1.05376887,
      "balance_loss_mlp": 1.02032876,
      "epoch": 0.6469067516383094,
      "flos": 19573650218880.0,
      "grad_norm": 2.286438685220183,
      "language_loss": 0.76365,
      "learning_rate": 1.1712318233277067e-06,
      "loss": 0.78570092,
      "num_input_tokens_seen": 115762015,
      "step": 5380,
      "time_per_iteration": 2.5708305835723877
    },
    {
      "auxiliary_loss_clip": 0.0106628,
      "auxiliary_loss_mlp": 0.01001804,
      "balance_loss_clip": 1.00891232,
      "balance_loss_mlp": 1.00083852,
      "epoch": 0.6470269945289485,
      "flos": 65098002522240.0,
      "grad_norm": 0.7508173117740923,
      "language_loss": 0.57831419,
      "learning_rate": 1.1705229424774916e-06,
      "loss": 0.59899509,
      "num_input_tokens_seen": 115816285,
      "step": 5381,
      "time_per_iteration": 3.0711944103240967
    },
    {
      "auxiliary_loss_clip": 0.01167567,
      "auxiliary_loss_mlp": 0.01028922,
      "balance_loss_clip": 1.05502713,
      "balance_loss_mlp": 1.02105463,
      "epoch": 0.6471472374195876,
      "flos": 30696943639680.0,
      "grad_norm": 2.2636192654591656,
      "language_loss": 0.64222026,
      "learning_rate": 1.1698141874582867e-06,
      "loss": 0.66418517,
      "num_input_tokens_seen": 115837330,
      "step": 5382,
      "time_per_iteration": 2.719473123550415
    },
    {
      "auxiliary_loss_clip": 0.01177303,
      "auxiliary_loss_mlp": 0.01034071,
      "balance_loss_clip": 1.05359483,
      "balance_loss_mlp": 1.0257802,
      "epoch": 0.6472674803102266,
      "flos": 20521835487360.0,
      "grad_norm": 1.8913408342190638,
      "language_loss": 0.72470564,
      "learning_rate": 1.169105558377609e-06,
      "loss": 0.74681938,
      "num_input_tokens_seen": 115857420,
      "step": 5383,
      "time_per_iteration": 2.685114860534668
    },
    {
      "auxiliary_loss_clip": 0.01168561,
      "auxiliary_loss_mlp": 0.00762112,
      "balance_loss_clip": 1.05383182,
      "balance_loss_mlp": 1.00096893,
      "epoch": 0.6473877232008658,
      "flos": 24715447320960.0,
      "grad_norm": 1.9317955669272056,
      "language_loss": 0.78920895,
      "learning_rate": 1.1683970553429587e-06,
      "loss": 0.80851567,
      "num_input_tokens_seen": 115878875,
      "step": 5384,
      "time_per_iteration": 3.6412835121154785
    },
    {
      "auxiliary_loss_clip": 0.01167281,
      "auxiliary_loss_mlp": 0.01031932,
      "balance_loss_clip": 1.05438972,
      "balance_loss_mlp": 1.02387929,
      "epoch": 0.6475079660915048,
      "flos": 15885552441600.0,
      "grad_norm": 1.9723046556707355,
      "language_loss": 0.82232076,
      "learning_rate": 1.1676886784618128e-06,
      "loss": 0.84431291,
      "num_input_tokens_seen": 115895540,
      "step": 5385,
      "time_per_iteration": 2.6473379135131836
    },
    {
      "auxiliary_loss_clip": 0.01182644,
      "auxiliary_loss_mlp": 0.0103105,
      "balance_loss_clip": 1.05442166,
      "balance_loss_mlp": 1.02181125,
      "epoch": 0.6476282089821439,
      "flos": 17381590922880.0,
      "grad_norm": 2.0120221152609092,
      "language_loss": 0.84122789,
      "learning_rate": 1.1669804278416332e-06,
      "loss": 0.86336482,
      "num_input_tokens_seen": 115910265,
      "step": 5386,
      "time_per_iteration": 2.6014764308929443
    },
    {
      "auxiliary_loss_clip": 0.01157963,
      "auxiliary_loss_mlp": 0.01026773,
      "balance_loss_clip": 1.0493716,
      "balance_loss_mlp": 1.01764178,
      "epoch": 0.6477484518727831,
      "flos": 20194078861440.0,
      "grad_norm": 2.074966574270886,
      "language_loss": 0.71707964,
      "learning_rate": 1.1662723035898602e-06,
      "loss": 0.73892701,
      "num_input_tokens_seen": 115930025,
      "step": 5387,
      "time_per_iteration": 2.627110242843628
    },
    {
      "auxiliary_loss_clip": 0.01181547,
      "auxiliary_loss_mlp": 0.01026767,
      "balance_loss_clip": 1.05544114,
      "balance_loss_mlp": 1.01834738,
      "epoch": 0.6478686947634221,
      "flos": 25410426641280.0,
      "grad_norm": 1.8683238075382014,
      "language_loss": 0.82045752,
      "learning_rate": 1.165564305813915e-06,
      "loss": 0.84254062,
      "num_input_tokens_seen": 115949025,
      "step": 5388,
      "time_per_iteration": 2.6811020374298096
    },
    {
      "auxiliary_loss_clip": 0.01178287,
      "auxiliary_loss_mlp": 0.00762008,
      "balance_loss_clip": 1.05477262,
      "balance_loss_mlp": 1.00097048,
      "epoch": 0.6479889376540612,
      "flos": 20083581648000.0,
      "grad_norm": 2.0912111860278975,
      "language_loss": 0.81378156,
      "learning_rate": 1.1648564346212019e-06,
      "loss": 0.83318448,
      "num_input_tokens_seen": 115968145,
      "step": 5389,
      "time_per_iteration": 2.62353515625
    },
    {
      "auxiliary_loss_clip": 0.01166541,
      "auxiliary_loss_mlp": 0.01028571,
      "balance_loss_clip": 1.05363619,
      "balance_loss_mlp": 1.02093554,
      "epoch": 0.6481091805447003,
      "flos": 26758082039040.0,
      "grad_norm": 2.6109842298948145,
      "language_loss": 0.76464593,
      "learning_rate": 1.164148690119104e-06,
      "loss": 0.78659701,
      "num_input_tokens_seen": 115989425,
      "step": 5390,
      "time_per_iteration": 2.7218334674835205
    },
    {
      "auxiliary_loss_clip": 0.01177531,
      "auxiliary_loss_mlp": 0.01030195,
      "balance_loss_clip": 1.05367637,
      "balance_loss_mlp": 1.02238715,
      "epoch": 0.6482294234353394,
      "flos": 23952094462080.0,
      "grad_norm": 1.8704718191837808,
      "language_loss": 0.73744494,
      "learning_rate": 1.163441072414985e-06,
      "loss": 0.7595222,
      "num_input_tokens_seen": 116009630,
      "step": 5391,
      "time_per_iteration": 2.639918327331543
    },
    {
      "auxiliary_loss_clip": 0.01179924,
      "auxiliary_loss_mlp": 0.01029939,
      "balance_loss_clip": 1.05565906,
      "balance_loss_mlp": 1.02190995,
      "epoch": 0.6483496663259785,
      "flos": 26209833776640.0,
      "grad_norm": 2.0291885139237658,
      "language_loss": 0.70026124,
      "learning_rate": 1.16273358161619e-06,
      "loss": 0.7223599,
      "num_input_tokens_seen": 116029965,
      "step": 5392,
      "time_per_iteration": 2.648672580718994
    },
    {
      "auxiliary_loss_clip": 0.01183844,
      "auxiliary_loss_mlp": 0.01038297,
      "balance_loss_clip": 1.05619526,
      "balance_loss_mlp": 1.02970839,
      "epoch": 0.6484699092166175,
      "flos": 20922239370240.0,
      "grad_norm": 1.858697081887715,
      "language_loss": 0.8361218,
      "learning_rate": 1.1620262178300446e-06,
      "loss": 0.85834324,
      "num_input_tokens_seen": 116048580,
      "step": 5393,
      "time_per_iteration": 2.6411640644073486
    },
    {
      "auxiliary_loss_clip": 0.01176999,
      "auxiliary_loss_mlp": 0.01025036,
      "balance_loss_clip": 1.05379009,
      "balance_loss_mlp": 1.01718628,
      "epoch": 0.6485901521072567,
      "flos": 33072865678080.0,
      "grad_norm": 2.993283578443975,
      "language_loss": 0.75969857,
      "learning_rate": 1.1613189811638563e-06,
      "loss": 0.78171891,
      "num_input_tokens_seen": 116070305,
      "step": 5394,
      "time_per_iteration": 2.761744499206543
    },
    {
      "auxiliary_loss_clip": 0.01153724,
      "auxiliary_loss_mlp": 0.01035392,
      "balance_loss_clip": 1.05481255,
      "balance_loss_mlp": 1.02727437,
      "epoch": 0.6487103949978957,
      "flos": 22274060745600.0,
      "grad_norm": 1.8389433352359221,
      "language_loss": 0.78292167,
      "learning_rate": 1.1606118717249117e-06,
      "loss": 0.80481291,
      "num_input_tokens_seen": 116090405,
      "step": 5395,
      "time_per_iteration": 2.812216281890869
    },
    {
      "auxiliary_loss_clip": 0.01181458,
      "auxiliary_loss_mlp": 0.01027393,
      "balance_loss_clip": 1.05220246,
      "balance_loss_mlp": 1.01929259,
      "epoch": 0.6488306378885348,
      "flos": 22930400010240.0,
      "grad_norm": 1.8680077358999778,
      "language_loss": 0.67888212,
      "learning_rate": 1.1599048896204787e-06,
      "loss": 0.70097065,
      "num_input_tokens_seen": 116110285,
      "step": 5396,
      "time_per_iteration": 2.60333514213562
    },
    {
      "auxiliary_loss_clip": 0.01167388,
      "auxiliary_loss_mlp": 0.01028433,
      "balance_loss_clip": 1.05146182,
      "balance_loss_mlp": 1.01991558,
      "epoch": 0.648950880779174,
      "flos": 20376110010240.0,
      "grad_norm": 3.9786789004024987,
      "language_loss": 0.81375074,
      "learning_rate": 1.1591980349578061e-06,
      "loss": 0.83570898,
      "num_input_tokens_seen": 116128955,
      "step": 5397,
      "time_per_iteration": 2.653059959411621
    },
    {
      "auxiliary_loss_clip": 0.01066293,
      "auxiliary_loss_mlp": 0.01001076,
      "balance_loss_clip": 1.00918472,
      "balance_loss_mlp": 1.00015223,
      "epoch": 0.649071123669813,
      "flos": 59930889310080.0,
      "grad_norm": 0.7381375534679672,
      "language_loss": 0.54255378,
      "learning_rate": 1.158491307844123e-06,
      "loss": 0.56322742,
      "num_input_tokens_seen": 116188875,
      "step": 5398,
      "time_per_iteration": 3.1348235607147217
    },
    {
      "auxiliary_loss_clip": 0.01162464,
      "auxiliary_loss_mlp": 0.01026463,
      "balance_loss_clip": 1.05328584,
      "balance_loss_mlp": 1.01878595,
      "epoch": 0.6491913665604521,
      "flos": 20446566537600.0,
      "grad_norm": 1.6947280434882577,
      "language_loss": 0.84247017,
      "learning_rate": 1.1577847083866387e-06,
      "loss": 0.8643595,
      "num_input_tokens_seen": 116207910,
      "step": 5399,
      "time_per_iteration": 3.6454758644104004
    },
    {
      "auxiliary_loss_clip": 0.01163678,
      "auxiliary_loss_mlp": 0.01023548,
      "balance_loss_clip": 1.05240941,
      "balance_loss_mlp": 1.01496482,
      "epoch": 0.6493116094510912,
      "flos": 16946820702720.0,
      "grad_norm": 8.986796026595329,
      "language_loss": 0.72393459,
      "learning_rate": 1.1570782366925453e-06,
      "loss": 0.74580681,
      "num_input_tokens_seen": 116226425,
      "step": 5400,
      "time_per_iteration": 2.5868377685546875
    },
    {
      "auxiliary_loss_clip": 0.01163389,
      "auxiliary_loss_mlp": 0.01026313,
      "balance_loss_clip": 1.05292261,
      "balance_loss_mlp": 1.01871324,
      "epoch": 0.6494318523417303,
      "flos": 18802935072000.0,
      "grad_norm": 2.0281506209455666,
      "language_loss": 0.75806093,
      "learning_rate": 1.1563718928690132e-06,
      "loss": 0.77995789,
      "num_input_tokens_seen": 116243860,
      "step": 5401,
      "time_per_iteration": 3.522160291671753
    },
    {
      "auxiliary_loss_clip": 0.01176275,
      "auxiliary_loss_mlp": 0.01042732,
      "balance_loss_clip": 1.05105615,
      "balance_loss_mlp": 1.03420293,
      "epoch": 0.6495520952323693,
      "flos": 18982847318400.0,
      "grad_norm": 2.3232531832411674,
      "language_loss": 0.71605754,
      "learning_rate": 1.1556656770231942e-06,
      "loss": 0.73824757,
      "num_input_tokens_seen": 116260055,
      "step": 5402,
      "time_per_iteration": 2.6047163009643555
    },
    {
      "auxiliary_loss_clip": 0.01163948,
      "auxiliary_loss_mlp": 0.01028625,
      "balance_loss_clip": 1.05317163,
      "balance_loss_mlp": 1.02104974,
      "epoch": 0.6496723381230085,
      "flos": 22745388032640.0,
      "grad_norm": 2.180839868482946,
      "language_loss": 0.76575243,
      "learning_rate": 1.1549595892622207e-06,
      "loss": 0.78767818,
      "num_input_tokens_seen": 116278825,
      "step": 5403,
      "time_per_iteration": 2.6146328449249268
    },
    {
      "auxiliary_loss_clip": 0.0106723,
      "auxiliary_loss_mlp": 0.01000542,
      "balance_loss_clip": 1.0098803,
      "balance_loss_mlp": 0.99960577,
      "epoch": 0.6497925810136476,
      "flos": 62145283887360.0,
      "grad_norm": 0.8256075253829431,
      "language_loss": 0.5896073,
      "learning_rate": 1.1542536296932047e-06,
      "loss": 0.61028504,
      "num_input_tokens_seen": 116342360,
      "step": 5404,
      "time_per_iteration": 3.169621467590332
    },
    {
      "auxiliary_loss_clip": 0.011594,
      "auxiliary_loss_mlp": 0.01029686,
      "balance_loss_clip": 1.05160344,
      "balance_loss_mlp": 1.02181172,
      "epoch": 0.6499128239042866,
      "flos": 20156731695360.0,
      "grad_norm": 2.1527579631718288,
      "language_loss": 0.70241988,
      "learning_rate": 1.1535477984232414e-06,
      "loss": 0.72431076,
      "num_input_tokens_seen": 116362235,
      "step": 5405,
      "time_per_iteration": 2.7213594913482666
    },
    {
      "auxiliary_loss_clip": 0.01175694,
      "auxiliary_loss_mlp": 0.0102618,
      "balance_loss_clip": 1.05163932,
      "balance_loss_mlp": 1.0181272,
      "epoch": 0.6500330667949258,
      "flos": 24462420940800.0,
      "grad_norm": 2.5534983834284533,
      "language_loss": 0.77158153,
      "learning_rate": 1.152842095559404e-06,
      "loss": 0.79360026,
      "num_input_tokens_seen": 116382895,
      "step": 5406,
      "time_per_iteration": 2.69836163520813
    },
    {
      "auxiliary_loss_clip": 0.01160163,
      "auxiliary_loss_mlp": 0.01023214,
      "balance_loss_clip": 1.04709339,
      "balance_loss_mlp": 1.0156827,
      "epoch": 0.6501533096855648,
      "flos": 25477399549440.0,
      "grad_norm": 2.0005896797268443,
      "language_loss": 0.76654303,
      "learning_rate": 1.1521365212087474e-06,
      "loss": 0.78837681,
      "num_input_tokens_seen": 116402880,
      "step": 5407,
      "time_per_iteration": 2.6642277240753174
    },
    {
      "auxiliary_loss_clip": 0.01181773,
      "auxiliary_loss_mlp": 0.01027871,
      "balance_loss_clip": 1.05411839,
      "balance_loss_mlp": 1.01963401,
      "epoch": 0.6502735525762039,
      "flos": 44819245347840.0,
      "grad_norm": 1.9702729244638173,
      "language_loss": 0.707744,
      "learning_rate": 1.1514310754783062e-06,
      "loss": 0.7298404,
      "num_input_tokens_seen": 116425830,
      "step": 5408,
      "time_per_iteration": 2.7773020267486572
    },
    {
      "auxiliary_loss_clip": 0.01162941,
      "auxiliary_loss_mlp": 0.01029292,
      "balance_loss_clip": 1.05223513,
      "balance_loss_mlp": 1.02080989,
      "epoch": 0.6503937954668431,
      "flos": 28658546726400.0,
      "grad_norm": 2.0057101977981153,
      "language_loss": 0.73080146,
      "learning_rate": 1.1507257584750964e-06,
      "loss": 0.75272381,
      "num_input_tokens_seen": 116446010,
      "step": 5409,
      "time_per_iteration": 2.7192678451538086
    },
    {
      "auxiliary_loss_clip": 0.01142862,
      "auxiliary_loss_mlp": 0.01027631,
      "balance_loss_clip": 1.04906917,
      "balance_loss_mlp": 1.01932812,
      "epoch": 0.6505140383574821,
      "flos": 20922562592640.0,
      "grad_norm": 1.9689207602858707,
      "language_loss": 0.77683032,
      "learning_rate": 1.150020570306113e-06,
      "loss": 0.79853523,
      "num_input_tokens_seen": 116465150,
      "step": 5410,
      "time_per_iteration": 3.585505962371826
    },
    {
      "auxiliary_loss_clip": 0.01170072,
      "auxiliary_loss_mlp": 0.01028959,
      "balance_loss_clip": 1.0521462,
      "balance_loss_mlp": 1.02066469,
      "epoch": 0.6506342812481212,
      "flos": 20595236929920.0,
      "grad_norm": 2.1522210712669043,
      "language_loss": 0.75240093,
      "learning_rate": 1.1493155110783338e-06,
      "loss": 0.77439123,
      "num_input_tokens_seen": 116483675,
      "step": 5411,
      "time_per_iteration": 2.661309242248535
    },
    {
      "auxiliary_loss_clip": 0.01179506,
      "auxiliary_loss_mlp": 0.01025762,
      "balance_loss_clip": 1.05304444,
      "balance_loss_mlp": 1.01812661,
      "epoch": 0.6507545241387603,
      "flos": 30226478279040.0,
      "grad_norm": 2.7454546192665115,
      "language_loss": 0.70599663,
      "learning_rate": 1.1486105808987155e-06,
      "loss": 0.72804928,
      "num_input_tokens_seen": 116505165,
      "step": 5412,
      "time_per_iteration": 2.7010881900787354
    },
    {
      "auxiliary_loss_clip": 0.01183284,
      "auxiliary_loss_mlp": 0.01037852,
      "balance_loss_clip": 1.05702019,
      "balance_loss_mlp": 1.02935791,
      "epoch": 0.6508747670293994,
      "flos": 17128241320320.0,
      "grad_norm": 1.7837443180050396,
      "language_loss": 0.81474584,
      "learning_rate": 1.1479057798741947e-06,
      "loss": 0.83695722,
      "num_input_tokens_seen": 116523220,
      "step": 5413,
      "time_per_iteration": 2.601304292678833
    },
    {
      "auxiliary_loss_clip": 0.01047019,
      "auxiliary_loss_mlp": 0.01000798,
      "balance_loss_clip": 1.01286983,
      "balance_loss_mlp": 0.99985582,
      "epoch": 0.6509950099200384,
      "flos": 68559826573440.0,
      "grad_norm": 0.7882645805888161,
      "language_loss": 0.53255177,
      "learning_rate": 1.14720110811169e-06,
      "loss": 0.55303001,
      "num_input_tokens_seen": 116580450,
      "step": 5414,
      "time_per_iteration": 3.309295415878296
    },
    {
      "auxiliary_loss_clip": 0.01169085,
      "auxiliary_loss_mlp": 0.01034928,
      "balance_loss_clip": 1.05214643,
      "balance_loss_mlp": 1.02649426,
      "epoch": 0.6511152528106776,
      "flos": 22347462188160.0,
      "grad_norm": 1.7115320108103578,
      "language_loss": 0.76911616,
      "learning_rate": 1.146496565718098e-06,
      "loss": 0.79115629,
      "num_input_tokens_seen": 116601020,
      "step": 5415,
      "time_per_iteration": 2.593580961227417
    },
    {
      "auxiliary_loss_clip": 0.0111257,
      "auxiliary_loss_mlp": 0.01022555,
      "balance_loss_clip": 1.04492509,
      "balance_loss_mlp": 1.01454401,
      "epoch": 0.6512354957013167,
      "flos": 20522158709760.0,
      "grad_norm": 2.325660210785599,
      "language_loss": 0.75980008,
      "learning_rate": 1.1457921528002996e-06,
      "loss": 0.78115141,
      "num_input_tokens_seen": 116619455,
      "step": 5416,
      "time_per_iteration": 2.7564492225646973
    },
    {
      "auxiliary_loss_clip": 0.01168814,
      "auxiliary_loss_mlp": 0.0102824,
      "balance_loss_clip": 1.05107856,
      "balance_loss_mlp": 1.02009845,
      "epoch": 0.6513557385919557,
      "flos": 32337342881280.0,
      "grad_norm": 6.856320198045292,
      "language_loss": 0.72379112,
      "learning_rate": 1.1450878694651522e-06,
      "loss": 0.74576169,
      "num_input_tokens_seen": 116640020,
      "step": 5417,
      "time_per_iteration": 2.718010663986206
    },
    {
      "auxiliary_loss_clip": 0.01177033,
      "auxiliary_loss_mlp": 0.01027218,
      "balance_loss_clip": 1.05204201,
      "balance_loss_mlp": 1.01892662,
      "epoch": 0.6514759814825949,
      "flos": 12093206417280.0,
      "grad_norm": 4.556085744513724,
      "language_loss": 0.63230622,
      "learning_rate": 1.1443837158194954e-06,
      "loss": 0.65434867,
      "num_input_tokens_seen": 116655165,
      "step": 5418,
      "time_per_iteration": 2.594867706298828
    },
    {
      "auxiliary_loss_clip": 0.01181029,
      "auxiliary_loss_mlp": 0.01028993,
      "balance_loss_clip": 1.05511165,
      "balance_loss_mlp": 1.02133358,
      "epoch": 0.651596224373234,
      "flos": 22526907557760.0,
      "grad_norm": 1.758911562284557,
      "language_loss": 0.74762499,
      "learning_rate": 1.1436796919701484e-06,
      "loss": 0.7697252,
      "num_input_tokens_seen": 116673880,
      "step": 5419,
      "time_per_iteration": 2.616083860397339
    },
    {
      "auxiliary_loss_clip": 0.0117903,
      "auxiliary_loss_mlp": 0.01025878,
      "balance_loss_clip": 1.05385494,
      "balance_loss_mlp": 1.01787341,
      "epoch": 0.651716467263873,
      "flos": 27818955250560.0,
      "grad_norm": 1.8027762756843317,
      "language_loss": 0.62186694,
      "learning_rate": 1.1429757980239115e-06,
      "loss": 0.64391601,
      "num_input_tokens_seen": 116694305,
      "step": 5420,
      "time_per_iteration": 2.6770150661468506
    },
    {
      "auxiliary_loss_clip": 0.0118414,
      "auxiliary_loss_mlp": 0.01030253,
      "balance_loss_clip": 1.05600834,
      "balance_loss_mlp": 1.0216341,
      "epoch": 0.6518367101545122,
      "flos": 24316300414080.0,
      "grad_norm": 2.7352869799983304,
      "language_loss": 0.81930834,
      "learning_rate": 1.1422720340875636e-06,
      "loss": 0.84145224,
      "num_input_tokens_seen": 116713055,
      "step": 5421,
      "time_per_iteration": 2.689364194869995
    },
    {
      "auxiliary_loss_clip": 0.01154361,
      "auxiliary_loss_mlp": 0.01027725,
      "balance_loss_clip": 1.05378437,
      "balance_loss_mlp": 1.01933861,
      "epoch": 0.6519569530451512,
      "flos": 20011939971840.0,
      "grad_norm": 2.10046908902418,
      "language_loss": 0.79445946,
      "learning_rate": 1.1415684002678671e-06,
      "loss": 0.81628036,
      "num_input_tokens_seen": 116731815,
      "step": 5422,
      "time_per_iteration": 2.680788993835449
    },
    {
      "auxiliary_loss_clip": 0.01151322,
      "auxiliary_loss_mlp": 0.0102859,
      "balance_loss_clip": 1.05186975,
      "balance_loss_mlp": 1.01948833,
      "epoch": 0.6520771959357903,
      "flos": 21576064682880.0,
      "grad_norm": 6.036163505622509,
      "language_loss": 0.78094077,
      "learning_rate": 1.1408648966715617e-06,
      "loss": 0.80273992,
      "num_input_tokens_seen": 116749335,
      "step": 5423,
      "time_per_iteration": 2.7019195556640625
    },
    {
      "auxiliary_loss_clip": 0.01170047,
      "auxiliary_loss_mlp": 0.01028729,
      "balance_loss_clip": 1.05084264,
      "balance_loss_mlp": 1.02005613,
      "epoch": 0.6521974388264293,
      "flos": 22711021695360.0,
      "grad_norm": 2.1137612864436344,
      "language_loss": 0.72728932,
      "learning_rate": 1.1401615234053683e-06,
      "loss": 0.74927706,
      "num_input_tokens_seen": 116768155,
      "step": 5424,
      "time_per_iteration": 2.6468610763549805
    },
    {
      "auxiliary_loss_clip": 0.01148187,
      "auxiliary_loss_mlp": 0.01032863,
      "balance_loss_clip": 1.05140662,
      "balance_loss_mlp": 1.02462554,
      "epoch": 0.6523176817170685,
      "flos": 23002939526400.0,
      "grad_norm": 4.5330672522021285,
      "language_loss": 0.76414895,
      "learning_rate": 1.1394582805759885e-06,
      "loss": 0.78595948,
      "num_input_tokens_seen": 116787435,
      "step": 5425,
      "time_per_iteration": 3.675351142883301
    },
    {
      "auxiliary_loss_clip": 0.01180984,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 1.05444765,
      "balance_loss_mlp": 1.02032745,
      "epoch": 0.6524379246077076,
      "flos": 21688249835520.0,
      "grad_norm": 1.9302928000033854,
      "language_loss": 0.76074511,
      "learning_rate": 1.1387551682901022e-06,
      "loss": 0.78284246,
      "num_input_tokens_seen": 116808040,
      "step": 5426,
      "time_per_iteration": 2.5604870319366455
    },
    {
      "auxiliary_loss_clip": 0.0114917,
      "auxiliary_loss_mlp": 0.01030999,
      "balance_loss_clip": 1.05026245,
      "balance_loss_mlp": 1.02313757,
      "epoch": 0.6525581674983466,
      "flos": 19390936711680.0,
      "grad_norm": 3.198692253001996,
      "language_loss": 0.70607221,
      "learning_rate": 1.138052186654373e-06,
      "loss": 0.72787386,
      "num_input_tokens_seen": 116825510,
      "step": 5427,
      "time_per_iteration": 3.535250425338745
    },
    {
      "auxiliary_loss_clip": 0.01169587,
      "auxiliary_loss_mlp": 0.01025563,
      "balance_loss_clip": 1.0554285,
      "balance_loss_mlp": 1.01712918,
      "epoch": 0.6526784103889858,
      "flos": 17165444832000.0,
      "grad_norm": 2.1825913455101786,
      "language_loss": 0.88096356,
      "learning_rate": 1.1373493357754417e-06,
      "loss": 0.90291512,
      "num_input_tokens_seen": 116844415,
      "step": 5428,
      "time_per_iteration": 2.671363115310669
    },
    {
      "auxiliary_loss_clip": 0.01175449,
      "auxiliary_loss_mlp": 0.01022688,
      "balance_loss_clip": 1.05122447,
      "balance_loss_mlp": 1.0148201,
      "epoch": 0.6527986532796248,
      "flos": 18989168112000.0,
      "grad_norm": 2.393115651882019,
      "language_loss": 0.7787714,
      "learning_rate": 1.1366466157599303e-06,
      "loss": 0.80075276,
      "num_input_tokens_seen": 116863690,
      "step": 5429,
      "time_per_iteration": 2.540811061859131
    },
    {
      "auxiliary_loss_clip": 0.0116476,
      "auxiliary_loss_mlp": 0.01030633,
      "balance_loss_clip": 1.05243516,
      "balance_loss_mlp": 1.02215433,
      "epoch": 0.6529188961702639,
      "flos": 14238581011200.0,
      "grad_norm": 2.5221890485055973,
      "language_loss": 0.7622841,
      "learning_rate": 1.1359440267144412e-06,
      "loss": 0.78423798,
      "num_input_tokens_seen": 116881145,
      "step": 5430,
      "time_per_iteration": 2.597353458404541
    },
    {
      "auxiliary_loss_clip": 0.01177866,
      "auxiliary_loss_mlp": 0.0102712,
      "balance_loss_clip": 1.05430198,
      "balance_loss_mlp": 1.01963913,
      "epoch": 0.653039139060903,
      "flos": 36682929158400.0,
      "grad_norm": 1.887879268239119,
      "language_loss": 0.74343652,
      "learning_rate": 1.1352415687455556e-06,
      "loss": 0.76548636,
      "num_input_tokens_seen": 116902405,
      "step": 5431,
      "time_per_iteration": 2.692561388015747
    },
    {
      "auxiliary_loss_clip": 0.0115447,
      "auxiliary_loss_mlp": 0.01035014,
      "balance_loss_clip": 1.05221701,
      "balance_loss_mlp": 1.02698267,
      "epoch": 0.6531593819515421,
      "flos": 25376275785600.0,
      "grad_norm": 2.9244971262798574,
      "language_loss": 0.64217758,
      "learning_rate": 1.1345392419598362e-06,
      "loss": 0.66407251,
      "num_input_tokens_seen": 116921285,
      "step": 5432,
      "time_per_iteration": 2.753622531890869
    },
    {
      "auxiliary_loss_clip": 0.01179514,
      "auxiliary_loss_mlp": 0.0103056,
      "balance_loss_clip": 1.05408669,
      "balance_loss_mlp": 1.02248669,
      "epoch": 0.6532796248421812,
      "flos": 21178533888000.0,
      "grad_norm": 1.6955323469537529,
      "language_loss": 0.71848035,
      "learning_rate": 1.1338370464638263e-06,
      "loss": 0.74058115,
      "num_input_tokens_seen": 116940685,
      "step": 5433,
      "time_per_iteration": 2.623959541320801
    },
    {
      "auxiliary_loss_clip": 0.01140385,
      "auxiliary_loss_mlp": 0.01028729,
      "balance_loss_clip": 1.04727912,
      "balance_loss_mlp": 1.0210433,
      "epoch": 0.6533998677328203,
      "flos": 17675950878720.0,
      "grad_norm": 4.034521140382307,
      "language_loss": 0.64245033,
      "learning_rate": 1.1331349823640474e-06,
      "loss": 0.66414154,
      "num_input_tokens_seen": 116958115,
      "step": 5434,
      "time_per_iteration": 2.7691423892974854
    },
    {
      "auxiliary_loss_clip": 0.01169227,
      "auxiliary_loss_mlp": 0.01030212,
      "balance_loss_clip": 1.05453837,
      "balance_loss_mlp": 1.02284729,
      "epoch": 0.6535201106234594,
      "flos": 28400384701440.0,
      "grad_norm": 2.596944726610589,
      "language_loss": 0.78673363,
      "learning_rate": 1.132433049767003e-06,
      "loss": 0.80872798,
      "num_input_tokens_seen": 116976030,
      "step": 5435,
      "time_per_iteration": 2.7713263034820557
    },
    {
      "auxiliary_loss_clip": 0.01176041,
      "auxiliary_loss_mlp": 0.01023672,
      "balance_loss_clip": 1.05386877,
      "balance_loss_mlp": 1.01608419,
      "epoch": 0.6536403535140984,
      "flos": 23586667447680.0,
      "grad_norm": 18.197020516902917,
      "language_loss": 0.81384879,
      "learning_rate": 1.1317312487791748e-06,
      "loss": 0.83584589,
      "num_input_tokens_seen": 116997680,
      "step": 5436,
      "time_per_iteration": 2.663691520690918
    },
    {
      "auxiliary_loss_clip": 0.01150916,
      "auxiliary_loss_mlp": 0.0103479,
      "balance_loss_clip": 1.05159986,
      "balance_loss_mlp": 1.02662981,
      "epoch": 0.6537605964047376,
      "flos": 21579476474880.0,
      "grad_norm": 3.4214702994417516,
      "language_loss": 0.73333263,
      "learning_rate": 1.1310295795070253e-06,
      "loss": 0.75518966,
      "num_input_tokens_seen": 117017620,
      "step": 5437,
      "time_per_iteration": 3.6102824211120605
    },
    {
      "auxiliary_loss_clip": 0.01167229,
      "auxiliary_loss_mlp": 0.01026358,
      "balance_loss_clip": 1.05088782,
      "balance_loss_mlp": 1.01849663,
      "epoch": 0.6538808392953767,
      "flos": 26833997433600.0,
      "grad_norm": 2.0053848343300738,
      "language_loss": 0.80803668,
      "learning_rate": 1.1303280420569982e-06,
      "loss": 0.82997262,
      "num_input_tokens_seen": 117039505,
      "step": 5438,
      "time_per_iteration": 2.66208815574646
    },
    {
      "auxiliary_loss_clip": 0.01148084,
      "auxiliary_loss_mlp": 0.01028404,
      "balance_loss_clip": 1.05373514,
      "balance_loss_mlp": 1.02027369,
      "epoch": 0.6540010821860157,
      "flos": 30738241301760.0,
      "grad_norm": 1.6522640390254133,
      "language_loss": 0.77613401,
      "learning_rate": 1.1296266365355158e-06,
      "loss": 0.79789889,
      "num_input_tokens_seen": 117062890,
      "step": 5439,
      "time_per_iteration": 2.8085451126098633
    },
    {
      "auxiliary_loss_clip": 0.01165114,
      "auxiliary_loss_mlp": 0.01034274,
      "balance_loss_clip": 1.05219173,
      "balance_loss_mlp": 1.02561975,
      "epoch": 0.6541213250766549,
      "flos": 26907147480960.0,
      "grad_norm": 2.22752153421406,
      "language_loss": 0.74038929,
      "learning_rate": 1.1289253630489806e-06,
      "loss": 0.76238322,
      "num_input_tokens_seen": 117083940,
      "step": 5440,
      "time_per_iteration": 2.7359063625335693
    },
    {
      "auxiliary_loss_clip": 0.01164918,
      "auxiliary_loss_mlp": 0.01032554,
      "balance_loss_clip": 1.05148244,
      "balance_loss_mlp": 1.0234822,
      "epoch": 0.6542415679672939,
      "flos": 19172384409600.0,
      "grad_norm": 2.819372341604244,
      "language_loss": 0.72452772,
      "learning_rate": 1.1282242217037753e-06,
      "loss": 0.74650252,
      "num_input_tokens_seen": 117101440,
      "step": 5441,
      "time_per_iteration": 2.644598960876465
    },
    {
      "auxiliary_loss_clip": 0.01154484,
      "auxiliary_loss_mlp": 0.00763391,
      "balance_loss_clip": 1.04872274,
      "balance_loss_mlp": 1.00088465,
      "epoch": 0.654361810857933,
      "flos": 48173517100800.0,
      "grad_norm": 3.1569722113362393,
      "language_loss": 0.61925137,
      "learning_rate": 1.127523212606262e-06,
      "loss": 0.63843012,
      "num_input_tokens_seen": 117124265,
      "step": 5442,
      "time_per_iteration": 2.897103786468506
    },
    {
      "auxiliary_loss_clip": 0.01176538,
      "auxiliary_loss_mlp": 0.0102434,
      "balance_loss_clip": 1.05287266,
      "balance_loss_mlp": 1.01679993,
      "epoch": 0.6544820537485722,
      "flos": 26943165843840.0,
      "grad_norm": 2.24144131483694,
      "language_loss": 0.73429227,
      "learning_rate": 1.1268223358627835e-06,
      "loss": 0.75630105,
      "num_input_tokens_seen": 117146755,
      "step": 5443,
      "time_per_iteration": 2.6650822162628174
    },
    {
      "auxiliary_loss_clip": 0.01163827,
      "auxiliary_loss_mlp": 0.01029389,
      "balance_loss_clip": 1.05115533,
      "balance_loss_mlp": 1.0210743,
      "epoch": 0.6546022966392112,
      "flos": 20886328748160.0,
      "grad_norm": 2.9887182393631706,
      "language_loss": 0.72244674,
      "learning_rate": 1.126121591579663e-06,
      "loss": 0.74437886,
      "num_input_tokens_seen": 117165960,
      "step": 5444,
      "time_per_iteration": 2.6542859077453613
    },
    {
      "auxiliary_loss_clip": 0.01177617,
      "auxiliary_loss_mlp": 0.01032946,
      "balance_loss_clip": 1.05429101,
      "balance_loss_mlp": 1.02501535,
      "epoch": 0.6547225395298503,
      "flos": 24936693143040.0,
      "grad_norm": 1.5938717400903755,
      "language_loss": 0.69668734,
      "learning_rate": 1.1254209798632018e-06,
      "loss": 0.71879292,
      "num_input_tokens_seen": 117186980,
      "step": 5445,
      "time_per_iteration": 2.665088176727295
    },
    {
      "auxiliary_loss_clip": 0.01176841,
      "auxiliary_loss_mlp": 0.01027577,
      "balance_loss_clip": 1.05306149,
      "balance_loss_mlp": 1.02026379,
      "epoch": 0.6548427824204894,
      "flos": 22565942663040.0,
      "grad_norm": 3.004142199090674,
      "language_loss": 0.84693444,
      "learning_rate": 1.124720500819683e-06,
      "loss": 0.86897862,
      "num_input_tokens_seen": 117205135,
      "step": 5446,
      "time_per_iteration": 2.6107468605041504
    },
    {
      "auxiliary_loss_clip": 0.01169106,
      "auxiliary_loss_mlp": 0.01030259,
      "balance_loss_clip": 1.05358577,
      "balance_loss_mlp": 1.02172947,
      "epoch": 0.6549630253111285,
      "flos": 18442500048000.0,
      "grad_norm": 1.9790545445549053,
      "language_loss": 0.82792044,
      "learning_rate": 1.1240201545553682e-06,
      "loss": 0.84991407,
      "num_input_tokens_seen": 117222935,
      "step": 5447,
      "time_per_iteration": 2.6333744525909424
    },
    {
      "auxiliary_loss_clip": 0.01164509,
      "auxiliary_loss_mlp": 0.01028979,
      "balance_loss_clip": 1.05310833,
      "balance_loss_mlp": 1.02092648,
      "epoch": 0.6550832682017675,
      "flos": 25187313312000.0,
      "grad_norm": 1.822963732209719,
      "language_loss": 0.73214865,
      "learning_rate": 1.1233199411764987e-06,
      "loss": 0.75408351,
      "num_input_tokens_seen": 117242370,
      "step": 5448,
      "time_per_iteration": 2.720904588699341
    },
    {
      "auxiliary_loss_clip": 0.0114412,
      "auxiliary_loss_mlp": 0.01023572,
      "balance_loss_clip": 1.048334,
      "balance_loss_mlp": 1.01577234,
      "epoch": 0.6552035110924067,
      "flos": 22748153379840.0,
      "grad_norm": 1.8069248234902664,
      "language_loss": 0.69067574,
      "learning_rate": 1.1226198607892978e-06,
      "loss": 0.71235263,
      "num_input_tokens_seen": 117262930,
      "step": 5449,
      "time_per_iteration": 2.6794371604919434
    },
    {
      "auxiliary_loss_clip": 0.01178154,
      "auxiliary_loss_mlp": 0.01030388,
      "balance_loss_clip": 1.05273569,
      "balance_loss_mlp": 1.02257347,
      "epoch": 0.6553237539830458,
      "flos": 21799178012160.0,
      "grad_norm": 1.8491991707170694,
      "language_loss": 0.8012132,
      "learning_rate": 1.1219199134999664e-06,
      "loss": 0.82329857,
      "num_input_tokens_seen": 117281430,
      "step": 5450,
      "time_per_iteration": 2.6759033203125
    },
    {
      "auxiliary_loss_clip": 0.01155549,
      "auxiliary_loss_mlp": 0.01032243,
      "balance_loss_clip": 1.0511024,
      "balance_loss_mlp": 1.02382684,
      "epoch": 0.6554439968736848,
      "flos": 20887226588160.0,
      "grad_norm": 2.411647479137624,
      "language_loss": 0.79038298,
      "learning_rate": 1.1212200994146863e-06,
      "loss": 0.81226093,
      "num_input_tokens_seen": 117299185,
      "step": 5451,
      "time_per_iteration": 3.6192233562469482
    },
    {
      "auxiliary_loss_clip": 0.0112419,
      "auxiliary_loss_mlp": 0.01027906,
      "balance_loss_clip": 1.04242158,
      "balance_loss_mlp": 1.02032983,
      "epoch": 0.655564239764324,
      "flos": 16139045698560.0,
      "grad_norm": 2.32033001367912,
      "language_loss": 0.75751066,
      "learning_rate": 1.120520418639618e-06,
      "loss": 0.77903169,
      "num_input_tokens_seen": 117317720,
      "step": 5452,
      "time_per_iteration": 3.8063547611236572
    },
    {
      "auxiliary_loss_clip": 0.01162987,
      "auxiliary_loss_mlp": 0.01025417,
      "balance_loss_clip": 1.05202794,
      "balance_loss_mlp": 1.01695871,
      "epoch": 0.655684482654963,
      "flos": 29570354496000.0,
      "grad_norm": 4.315176934502241,
      "language_loss": 0.83446634,
      "learning_rate": 1.119820871280903e-06,
      "loss": 0.8563503,
      "num_input_tokens_seen": 117338795,
      "step": 5453,
      "time_per_iteration": 3.6170034408569336
    },
    {
      "auxiliary_loss_clip": 0.01124705,
      "auxiliary_loss_mlp": 0.01029006,
      "balance_loss_clip": 1.04435825,
      "balance_loss_mlp": 1.02085781,
      "epoch": 0.6558047255456021,
      "flos": 29789409588480.0,
      "grad_norm": 2.78590290564298,
      "language_loss": 0.73389614,
      "learning_rate": 1.1191214574446614e-06,
      "loss": 0.7554332,
      "num_input_tokens_seen": 117359040,
      "step": 5454,
      "time_per_iteration": 2.7289063930511475
    },
    {
      "auxiliary_loss_clip": 0.01178163,
      "auxiliary_loss_mlp": 0.00762336,
      "balance_loss_clip": 1.05236423,
      "balance_loss_mlp": 1.00095868,
      "epoch": 0.6559249684362413,
      "flos": 29059166090880.0,
      "grad_norm": 1.5441950741399668,
      "language_loss": 0.80284244,
      "learning_rate": 1.118422177236995e-06,
      "loss": 0.82224751,
      "num_input_tokens_seen": 117380865,
      "step": 5455,
      "time_per_iteration": 2.650707960128784
    },
    {
      "auxiliary_loss_clip": 0.01147579,
      "auxiliary_loss_mlp": 0.01028578,
      "balance_loss_clip": 1.05463266,
      "balance_loss_mlp": 1.02051377,
      "epoch": 0.6560452113268803,
      "flos": 20225464369920.0,
      "grad_norm": 2.032534937370925,
      "language_loss": 0.86141032,
      "learning_rate": 1.1177230307639835e-06,
      "loss": 0.88317192,
      "num_input_tokens_seen": 117398405,
      "step": 5456,
      "time_per_iteration": 2.6746463775634766
    },
    {
      "auxiliary_loss_clip": 0.01163707,
      "auxiliary_loss_mlp": 0.01029629,
      "balance_loss_clip": 1.05185318,
      "balance_loss_mlp": 1.02160037,
      "epoch": 0.6561654542175194,
      "flos": 25045538330880.0,
      "grad_norm": 2.344940877604274,
      "language_loss": 0.78907436,
      "learning_rate": 1.1170240181316865e-06,
      "loss": 0.81100774,
      "num_input_tokens_seen": 117419850,
      "step": 5457,
      "time_per_iteration": 2.6410253047943115
    },
    {
      "auxiliary_loss_clip": 0.01169152,
      "auxiliary_loss_mlp": 0.01024112,
      "balance_loss_clip": 1.0510087,
      "balance_loss_mlp": 1.01592267,
      "epoch": 0.6562856971081584,
      "flos": 22856711258880.0,
      "grad_norm": 2.720944913162446,
      "language_loss": 0.79872477,
      "learning_rate": 1.1163251394461442e-06,
      "loss": 0.82065737,
      "num_input_tokens_seen": 117438330,
      "step": 5458,
      "time_per_iteration": 2.646044969558716
    },
    {
      "auxiliary_loss_clip": 0.01180469,
      "auxiliary_loss_mlp": 0.01025414,
      "balance_loss_clip": 1.05449986,
      "balance_loss_mlp": 1.01761746,
      "epoch": 0.6564059399987976,
      "flos": 18872565586560.0,
      "grad_norm": 2.1599744485904853,
      "language_loss": 0.82439208,
      "learning_rate": 1.1156263948133746e-06,
      "loss": 0.84645092,
      "num_input_tokens_seen": 117454985,
      "step": 5459,
      "time_per_iteration": 2.5932183265686035
    },
    {
      "auxiliary_loss_clip": 0.01180272,
      "auxiliary_loss_mlp": 0.01028348,
      "balance_loss_clip": 1.05385327,
      "balance_loss_mlp": 1.01986623,
      "epoch": 0.6565261828894366,
      "flos": 25484187219840.0,
      "grad_norm": 6.070373847065587,
      "language_loss": 0.77828193,
      "learning_rate": 1.1149277843393787e-06,
      "loss": 0.80036813,
      "num_input_tokens_seen": 117476145,
      "step": 5460,
      "time_per_iteration": 2.6256911754608154
    },
    {
      "auxiliary_loss_clip": 0.01177088,
      "auxiliary_loss_mlp": 0.01034796,
      "balance_loss_clip": 1.05267096,
      "balance_loss_mlp": 1.02648687,
      "epoch": 0.6566464257800757,
      "flos": 19683500987520.0,
      "grad_norm": 2.609616445216724,
      "language_loss": 0.63585913,
      "learning_rate": 1.1142293081301342e-06,
      "loss": 0.65797794,
      "num_input_tokens_seen": 117494025,
      "step": 5461,
      "time_per_iteration": 2.6604537963867188
    },
    {
      "auxiliary_loss_clip": 0.01160583,
      "auxiliary_loss_mlp": 0.007616,
      "balance_loss_clip": 1.0520798,
      "balance_loss_mlp": 1.00087357,
      "epoch": 0.6567666686707149,
      "flos": 23514127931520.0,
      "grad_norm": 1.6847830415275373,
      "language_loss": 0.68223464,
      "learning_rate": 1.1135309662915995e-06,
      "loss": 0.70145643,
      "num_input_tokens_seen": 117514190,
      "step": 5462,
      "time_per_iteration": 3.6606881618499756
    },
    {
      "auxiliary_loss_clip": 0.01154891,
      "auxiliary_loss_mlp": 0.00761972,
      "balance_loss_clip": 1.04704714,
      "balance_loss_mlp": 1.00085568,
      "epoch": 0.6568869115613539,
      "flos": 32781342896640.0,
      "grad_norm": 4.115722512739945,
      "language_loss": 0.60342169,
      "learning_rate": 1.112832758929712e-06,
      "loss": 0.6225903,
      "num_input_tokens_seen": 117536800,
      "step": 5463,
      "time_per_iteration": 2.712498903274536
    },
    {
      "auxiliary_loss_clip": 0.01146081,
      "auxiliary_loss_mlp": 0.01025724,
      "balance_loss_clip": 1.04925394,
      "balance_loss_mlp": 1.01767719,
      "epoch": 0.657007154451993,
      "flos": 18442428220800.0,
      "grad_norm": 1.889797556241134,
      "language_loss": 0.7501899,
      "learning_rate": 1.11213468615039e-06,
      "loss": 0.77190793,
      "num_input_tokens_seen": 117556230,
      "step": 5464,
      "time_per_iteration": 2.707443952560425
    },
    {
      "auxiliary_loss_clip": 0.01161556,
      "auxiliary_loss_mlp": 0.01031468,
      "balance_loss_clip": 1.04978681,
      "balance_loss_mlp": 1.02458715,
      "epoch": 0.6571273973426321,
      "flos": 25156717902720.0,
      "grad_norm": 2.014518714527247,
      "language_loss": 0.75473833,
      "learning_rate": 1.1114367480595292e-06,
      "loss": 0.77666855,
      "num_input_tokens_seen": 117577310,
      "step": 5465,
      "time_per_iteration": 2.6371991634368896
    },
    {
      "auxiliary_loss_clip": 0.01167013,
      "auxiliary_loss_mlp": 0.01030874,
      "balance_loss_clip": 1.05000854,
      "balance_loss_mlp": 1.02195132,
      "epoch": 0.6572476402332712,
      "flos": 17529830352000.0,
      "grad_norm": 1.9151095388539001,
      "language_loss": 0.81411523,
      "learning_rate": 1.1107389447630086e-06,
      "loss": 0.83609414,
      "num_input_tokens_seen": 117596010,
      "step": 5466,
      "time_per_iteration": 2.6520888805389404
    },
    {
      "auxiliary_loss_clip": 0.01151002,
      "auxiliary_loss_mlp": 0.01032605,
      "balance_loss_clip": 1.05063248,
      "balance_loss_mlp": 1.02485681,
      "epoch": 0.6573678831239103,
      "flos": 17014260487680.0,
      "grad_norm": 3.7449366745638737,
      "language_loss": 0.78637385,
      "learning_rate": 1.1100412763666818e-06,
      "loss": 0.8082099,
      "num_input_tokens_seen": 117611270,
      "step": 5467,
      "time_per_iteration": 2.6766107082366943
    },
    {
      "auxiliary_loss_clip": 0.0117791,
      "auxiliary_loss_mlp": 0.01028368,
      "balance_loss_clip": 1.05273104,
      "balance_loss_mlp": 1.02060175,
      "epoch": 0.6574881260145494,
      "flos": 23910078528000.0,
      "grad_norm": 1.5352478877744449,
      "language_loss": 0.80138725,
      "learning_rate": 1.1093437429763865e-06,
      "loss": 0.82345009,
      "num_input_tokens_seen": 117631535,
      "step": 5468,
      "time_per_iteration": 2.617736339569092
    },
    {
      "auxiliary_loss_clip": 0.01159635,
      "auxiliary_loss_mlp": 0.01025756,
      "balance_loss_clip": 1.05123079,
      "balance_loss_mlp": 1.01858878,
      "epoch": 0.6576083689051885,
      "flos": 11218458504960.0,
      "grad_norm": 5.57720620246915,
      "language_loss": 0.7363497,
      "learning_rate": 1.1086463446979361e-06,
      "loss": 0.75820357,
      "num_input_tokens_seen": 117649885,
      "step": 5469,
      "time_per_iteration": 2.6693851947784424
    },
    {
      "auxiliary_loss_clip": 0.01170699,
      "auxiliary_loss_mlp": 0.01031429,
      "balance_loss_clip": 1.05357718,
      "balance_loss_mlp": 1.02303648,
      "epoch": 0.6577286117958275,
      "flos": 22455553190400.0,
      "grad_norm": 1.9394164107361855,
      "language_loss": 0.77461338,
      "learning_rate": 1.1079490816371277e-06,
      "loss": 0.79663467,
      "num_input_tokens_seen": 117669650,
      "step": 5470,
      "time_per_iteration": 2.64789080619812
    },
    {
      "auxiliary_loss_clip": 0.01134666,
      "auxiliary_loss_mlp": 0.01032506,
      "balance_loss_clip": 1.04947293,
      "balance_loss_mlp": 1.02442956,
      "epoch": 0.6578488546864667,
      "flos": 21872184405120.0,
      "grad_norm": 2.8504313473712646,
      "language_loss": 0.74829888,
      "learning_rate": 1.1072519538997352e-06,
      "loss": 0.76997066,
      "num_input_tokens_seen": 117688790,
      "step": 5471,
      "time_per_iteration": 2.750152587890625
    },
    {
      "auxiliary_loss_clip": 0.01146086,
      "auxiliary_loss_mlp": 0.01028073,
      "balance_loss_clip": 1.04952097,
      "balance_loss_mlp": 1.02008057,
      "epoch": 0.6579690975771058,
      "flos": 23543753673600.0,
      "grad_norm": 2.0875715567110418,
      "language_loss": 0.82864165,
      "learning_rate": 1.1065549615915095e-06,
      "loss": 0.85038322,
      "num_input_tokens_seen": 117708620,
      "step": 5472,
      "time_per_iteration": 2.686032295227051
    },
    {
      "auxiliary_loss_clip": 0.01154499,
      "auxiliary_loss_mlp": 0.01032239,
      "balance_loss_clip": 1.04950249,
      "balance_loss_mlp": 1.0244875,
      "epoch": 0.6580893404677448,
      "flos": 32743995730560.0,
      "grad_norm": 3.093788832414114,
      "language_loss": 0.78587693,
      "learning_rate": 1.105858104818187e-06,
      "loss": 0.80774426,
      "num_input_tokens_seen": 117729775,
      "step": 5473,
      "time_per_iteration": 2.8462493419647217
    },
    {
      "auxiliary_loss_clip": 0.0114682,
      "auxiliary_loss_mlp": 0.0102915,
      "balance_loss_clip": 1.04908276,
      "balance_loss_mlp": 1.02093029,
      "epoch": 0.658209583358384,
      "flos": 15888138220800.0,
      "grad_norm": 10.399348560184313,
      "language_loss": 0.75241804,
      "learning_rate": 1.105161383685478e-06,
      "loss": 0.77417767,
      "num_input_tokens_seen": 117746160,
      "step": 5474,
      "time_per_iteration": 2.666067123413086
    },
    {
      "auxiliary_loss_clip": 0.01069076,
      "auxiliary_loss_mlp": 0.01003977,
      "balance_loss_clip": 1.01150787,
      "balance_loss_mlp": 1.00310636,
      "epoch": 0.658329826249023,
      "flos": 62695902447360.0,
      "grad_norm": 0.7273817716407487,
      "language_loss": 0.56196773,
      "learning_rate": 1.1044647982990771e-06,
      "loss": 0.58269829,
      "num_input_tokens_seen": 117808045,
      "step": 5475,
      "time_per_iteration": 3.138049364089966
    },
    {
      "auxiliary_loss_clip": 0.01164261,
      "auxiliary_loss_mlp": 0.01030345,
      "balance_loss_clip": 1.05125141,
      "balance_loss_mlp": 1.02201843,
      "epoch": 0.6584500691396621,
      "flos": 31722624501120.0,
      "grad_norm": 2.5667022896236924,
      "language_loss": 0.64778376,
      "learning_rate": 1.1037683487646536e-06,
      "loss": 0.66972977,
      "num_input_tokens_seen": 117828330,
      "step": 5476,
      "time_per_iteration": 2.7907543182373047
    },
    {
      "auxiliary_loss_clip": 0.01177981,
      "auxiliary_loss_mlp": 0.01021639,
      "balance_loss_clip": 1.05414963,
      "balance_loss_mlp": 1.01372385,
      "epoch": 0.6585703120303013,
      "flos": 18406086635520.0,
      "grad_norm": 1.786975545611144,
      "language_loss": 0.77434397,
      "learning_rate": 1.1030720351878583e-06,
      "loss": 0.79634023,
      "num_input_tokens_seen": 117846450,
      "step": 5477,
      "time_per_iteration": 3.5832557678222656
    },
    {
      "auxiliary_loss_clip": 0.01060775,
      "auxiliary_loss_mlp": 0.01000509,
      "balance_loss_clip": 1.01133919,
      "balance_loss_mlp": 0.9996506,
      "epoch": 0.6586905549209403,
      "flos": 58309880434560.0,
      "grad_norm": 0.8249599484934346,
      "language_loss": 0.57530338,
      "learning_rate": 1.102375857674323e-06,
      "loss": 0.59591627,
      "num_input_tokens_seen": 117908365,
      "step": 5478,
      "time_per_iteration": 4.130535125732422
    },
    {
      "auxiliary_loss_clip": 0.01154556,
      "auxiliary_loss_mlp": 0.01027091,
      "balance_loss_clip": 1.05012679,
      "balance_loss_mlp": 1.01950669,
      "epoch": 0.6588107978115794,
      "flos": 22782627457920.0,
      "grad_norm": 1.9268740114028398,
      "language_loss": 0.90246934,
      "learning_rate": 1.1016798163296561e-06,
      "loss": 0.92428577,
      "num_input_tokens_seen": 117927565,
      "step": 5479,
      "time_per_iteration": 3.759093761444092
    },
    {
      "auxiliary_loss_clip": 0.01149195,
      "auxiliary_loss_mlp": 0.01025088,
      "balance_loss_clip": 1.04991651,
      "balance_loss_mlp": 1.01765537,
      "epoch": 0.6589310407022185,
      "flos": 20667525050880.0,
      "grad_norm": 1.8360875669902554,
      "language_loss": 0.66207081,
      "learning_rate": 1.1009839112594471e-06,
      "loss": 0.68381363,
      "num_input_tokens_seen": 117945590,
      "step": 5480,
      "time_per_iteration": 2.673145294189453
    },
    {
      "auxiliary_loss_clip": 0.01178905,
      "auxiliary_loss_mlp": 0.01027174,
      "balance_loss_clip": 1.05293381,
      "balance_loss_mlp": 1.0192554,
      "epoch": 0.6590512835928576,
      "flos": 25630595055360.0,
      "grad_norm": 2.2420228377049525,
      "language_loss": 0.71958947,
      "learning_rate": 1.1002881425692638e-06,
      "loss": 0.74165022,
      "num_input_tokens_seen": 117966020,
      "step": 5481,
      "time_per_iteration": 2.6809208393096924
    },
    {
      "auxiliary_loss_clip": 0.01162586,
      "auxiliary_loss_mlp": 0.01029509,
      "balance_loss_clip": 1.04849088,
      "balance_loss_mlp": 1.02140903,
      "epoch": 0.6591715264834966,
      "flos": 23726108044800.0,
      "grad_norm": 1.742555661960993,
      "language_loss": 0.75440156,
      "learning_rate": 1.0995925103646532e-06,
      "loss": 0.7763226,
      "num_input_tokens_seen": 117984620,
      "step": 5482,
      "time_per_iteration": 2.659689426422119
    },
    {
      "auxiliary_loss_clip": 0.01178948,
      "auxiliary_loss_mlp": 0.01026094,
      "balance_loss_clip": 1.05451512,
      "balance_loss_mlp": 1.01850951,
      "epoch": 0.6592917693741358,
      "flos": 35773850822400.0,
      "grad_norm": 1.5485920862419011,
      "language_loss": 0.67013896,
      "learning_rate": 1.0988970147511437e-06,
      "loss": 0.69218946,
      "num_input_tokens_seen": 118006500,
      "step": 5483,
      "time_per_iteration": 2.7183680534362793
    },
    {
      "auxiliary_loss_clip": 0.01177588,
      "auxiliary_loss_mlp": 0.01031597,
      "balance_loss_clip": 1.05410695,
      "balance_loss_mlp": 1.02395535,
      "epoch": 0.6594120122647749,
      "flos": 21396834794880.0,
      "grad_norm": 2.9579099177944284,
      "language_loss": 0.80642164,
      "learning_rate": 1.0982016558342405e-06,
      "loss": 0.8285135,
      "num_input_tokens_seen": 118025470,
      "step": 5484,
      "time_per_iteration": 2.569959878921509
    },
    {
      "auxiliary_loss_clip": 0.01175996,
      "auxiliary_loss_mlp": 0.01028649,
      "balance_loss_clip": 1.05116951,
      "balance_loss_mlp": 1.02096629,
      "epoch": 0.6595322551554139,
      "flos": 19351829779200.0,
      "grad_norm": 2.6526722449168103,
      "language_loss": 0.71400619,
      "learning_rate": 1.0975064337194291e-06,
      "loss": 0.73605263,
      "num_input_tokens_seen": 118043515,
      "step": 5485,
      "time_per_iteration": 2.6304209232330322
    },
    {
      "auxiliary_loss_clip": 0.01164094,
      "auxiliary_loss_mlp": 0.01028707,
      "balance_loss_clip": 1.05084395,
      "balance_loss_mlp": 1.02079737,
      "epoch": 0.6596524980460531,
      "flos": 16837113588480.0,
      "grad_norm": 1.560539972808896,
      "language_loss": 0.70408082,
      "learning_rate": 1.0968113485121743e-06,
      "loss": 0.72600889,
      "num_input_tokens_seen": 118063105,
      "step": 5486,
      "time_per_iteration": 2.724025011062622
    },
    {
      "auxiliary_loss_clip": 0.01165483,
      "auxiliary_loss_mlp": 0.01023581,
      "balance_loss_clip": 1.05045116,
      "balance_loss_mlp": 1.01533175,
      "epoch": 0.6597727409366921,
      "flos": 21798567480960.0,
      "grad_norm": 1.986174347420425,
      "language_loss": 0.80103111,
      "learning_rate": 1.0961164003179185e-06,
      "loss": 0.82292175,
      "num_input_tokens_seen": 118081615,
      "step": 5487,
      "time_per_iteration": 2.612027645111084
    },
    {
      "auxiliary_loss_clip": 0.01168839,
      "auxiliary_loss_mlp": 0.01026285,
      "balance_loss_clip": 1.05334353,
      "balance_loss_mlp": 1.01834548,
      "epoch": 0.6598929838273312,
      "flos": 23730704985600.0,
      "grad_norm": 1.939741392928909,
      "language_loss": 0.84269392,
      "learning_rate": 1.0954215892420884e-06,
      "loss": 0.86464512,
      "num_input_tokens_seen": 118102315,
      "step": 5488,
      "time_per_iteration": 3.636425256729126
    },
    {
      "auxiliary_loss_clip": 0.01180607,
      "auxiliary_loss_mlp": 0.01039201,
      "balance_loss_clip": 1.05333042,
      "balance_loss_mlp": 1.03030252,
      "epoch": 0.6600132267179702,
      "flos": 19974520978560.0,
      "grad_norm": 1.9964507278616266,
      "language_loss": 0.70586061,
      "learning_rate": 1.094726915390082e-06,
      "loss": 0.7280587,
      "num_input_tokens_seen": 118120650,
      "step": 5489,
      "time_per_iteration": 2.541405200958252
    },
    {
      "auxiliary_loss_clip": 0.01150147,
      "auxiliary_loss_mlp": 0.01023815,
      "balance_loss_clip": 1.04811418,
      "balance_loss_mlp": 1.01617384,
      "epoch": 0.6601334696086094,
      "flos": 22342649765760.0,
      "grad_norm": 2.514346093633704,
      "language_loss": 0.69932973,
      "learning_rate": 1.0940323788672836e-06,
      "loss": 0.7210694,
      "num_input_tokens_seen": 118139825,
      "step": 5490,
      "time_per_iteration": 2.6687097549438477
    },
    {
      "auxiliary_loss_clip": 0.01171094,
      "auxiliary_loss_mlp": 0.01026492,
      "balance_loss_clip": 1.05031323,
      "balance_loss_mlp": 1.01923203,
      "epoch": 0.6602537124992485,
      "flos": 25703098657920.0,
      "grad_norm": 1.8250239011022225,
      "language_loss": 0.73994172,
      "learning_rate": 1.0933379797790522e-06,
      "loss": 0.76191759,
      "num_input_tokens_seen": 118159240,
      "step": 5491,
      "time_per_iteration": 2.6445720195770264
    },
    {
      "auxiliary_loss_clip": 0.0117831,
      "auxiliary_loss_mlp": 0.010272,
      "balance_loss_clip": 1.05418015,
      "balance_loss_mlp": 1.01862884,
      "epoch": 0.6603739553898875,
      "flos": 25848572739840.0,
      "grad_norm": 2.8562956627683613,
      "language_loss": 0.71452153,
      "learning_rate": 1.0926437182307293e-06,
      "loss": 0.73657662,
      "num_input_tokens_seen": 118178050,
      "step": 5492,
      "time_per_iteration": 2.6890084743499756
    },
    {
      "auxiliary_loss_clip": 0.01176178,
      "auxiliary_loss_mlp": 0.01026616,
      "balance_loss_clip": 1.05100703,
      "balance_loss_mlp": 1.01894522,
      "epoch": 0.6604941982805267,
      "flos": 24570296461440.0,
      "grad_norm": 1.888628541055224,
      "language_loss": 0.7806744,
      "learning_rate": 1.0919495943276338e-06,
      "loss": 0.80270231,
      "num_input_tokens_seen": 118199070,
      "step": 5493,
      "time_per_iteration": 2.6164801120758057
    },
    {
      "auxiliary_loss_clip": 0.01170437,
      "auxiliary_loss_mlp": 0.01032266,
      "balance_loss_clip": 1.05249381,
      "balance_loss_mlp": 1.02468467,
      "epoch": 0.6606144411711657,
      "flos": 13261775581440.0,
      "grad_norm": 5.323980572907349,
      "language_loss": 0.76650846,
      "learning_rate": 1.0912556081750611e-06,
      "loss": 0.78853554,
      "num_input_tokens_seen": 118217000,
      "step": 5494,
      "time_per_iteration": 2.6460869312286377
    },
    {
      "auxiliary_loss_clip": 0.01167922,
      "auxiliary_loss_mlp": 0.01026829,
      "balance_loss_clip": 1.05124593,
      "balance_loss_mlp": 1.01947093,
      "epoch": 0.6607346840618048,
      "flos": 25155281358720.0,
      "grad_norm": 1.8725231190634681,
      "language_loss": 0.76930213,
      "learning_rate": 1.0905617598782909e-06,
      "loss": 0.79124963,
      "num_input_tokens_seen": 118237205,
      "step": 5495,
      "time_per_iteration": 2.7120447158813477
    },
    {
      "auxiliary_loss_clip": 0.0116285,
      "auxiliary_loss_mlp": 0.010307,
      "balance_loss_clip": 1.05373406,
      "balance_loss_mlp": 1.02299285,
      "epoch": 0.660854926952444,
      "flos": 17638029095040.0,
      "grad_norm": 4.047349632137754,
      "language_loss": 0.81364578,
      "learning_rate": 1.0898680495425775e-06,
      "loss": 0.8355813,
      "num_input_tokens_seen": 118255495,
      "step": 5496,
      "time_per_iteration": 2.6155953407287598
    },
    {
      "auxiliary_loss_clip": 0.01177668,
      "auxiliary_loss_mlp": 0.01030697,
      "balance_loss_clip": 1.05334854,
      "balance_loss_mlp": 1.0223341,
      "epoch": 0.660975169843083,
      "flos": 16836000266880.0,
      "grad_norm": 1.6444494931417464,
      "language_loss": 0.80618107,
      "learning_rate": 1.0891744772731594e-06,
      "loss": 0.82826471,
      "num_input_tokens_seen": 118273310,
      "step": 5497,
      "time_per_iteration": 2.602064371109009
    },
    {
      "auxiliary_loss_clip": 0.01147114,
      "auxiliary_loss_mlp": 0.0102691,
      "balance_loss_clip": 1.04645514,
      "balance_loss_mlp": 1.01915812,
      "epoch": 0.6610954127337221,
      "flos": 26870410846080.0,
      "grad_norm": 3.321460941357235,
      "language_loss": 0.66116905,
      "learning_rate": 1.088481043175248e-06,
      "loss": 0.68290925,
      "num_input_tokens_seen": 118293880,
      "step": 5498,
      "time_per_iteration": 2.701842784881592
    },
    {
      "auxiliary_loss_clip": 0.01165256,
      "auxiliary_loss_mlp": 0.01027562,
      "balance_loss_clip": 1.0528841,
      "balance_loss_mlp": 1.02006316,
      "epoch": 0.6612156556243612,
      "flos": 26465697331200.0,
      "grad_norm": 2.0279285284794883,
      "language_loss": 0.75823438,
      "learning_rate": 1.0877877473540368e-06,
      "loss": 0.78016257,
      "num_input_tokens_seen": 118314465,
      "step": 5499,
      "time_per_iteration": 2.7067596912384033
    },
    {
      "auxiliary_loss_clip": 0.01159749,
      "auxiliary_loss_mlp": 0.01026671,
      "balance_loss_clip": 1.04989028,
      "balance_loss_mlp": 1.01835048,
      "epoch": 0.6613358985150003,
      "flos": 19791915212160.0,
      "grad_norm": 1.872403398714239,
      "language_loss": 0.7303555,
      "learning_rate": 1.0870945899147002e-06,
      "loss": 0.75221968,
      "num_input_tokens_seen": 118331110,
      "step": 5500,
      "time_per_iteration": 2.610161066055298
    },
    {
      "auxiliary_loss_clip": 0.01175513,
      "auxiliary_loss_mlp": 0.01025979,
      "balance_loss_clip": 1.05209947,
      "balance_loss_mlp": 1.01764023,
      "epoch": 0.6614561414056394,
      "flos": 26831627136000.0,
      "grad_norm": 1.9771858013071668,
      "language_loss": 0.76733255,
      "learning_rate": 1.0864015709623879e-06,
      "loss": 0.78934747,
      "num_input_tokens_seen": 118351980,
      "step": 5501,
      "time_per_iteration": 2.6433184146881104
    },
    {
      "auxiliary_loss_clip": 0.01177778,
      "auxiliary_loss_mlp": 0.0102467,
      "balance_loss_clip": 1.05301046,
      "balance_loss_mlp": 1.01693928,
      "epoch": 0.6615763842962785,
      "flos": 22894597128960.0,
      "grad_norm": 4.401063765600463,
      "language_loss": 0.80450779,
      "learning_rate": 1.0857086906022313e-06,
      "loss": 0.82653224,
      "num_input_tokens_seen": 118370315,
      "step": 5502,
      "time_per_iteration": 2.6172738075256348
    },
    {
      "auxiliary_loss_clip": 0.01147516,
      "auxiliary_loss_mlp": 0.01029487,
      "balance_loss_clip": 1.04855704,
      "balance_loss_mlp": 1.02105582,
      "epoch": 0.6616966271869176,
      "flos": 24790321221120.0,
      "grad_norm": 2.029734485669914,
      "language_loss": 0.73339868,
      "learning_rate": 1.0850159489393388e-06,
      "loss": 0.75516868,
      "num_input_tokens_seen": 118389575,
      "step": 5503,
      "time_per_iteration": 3.727156639099121
    },
    {
      "auxiliary_loss_clip": 0.01175576,
      "auxiliary_loss_mlp": 0.01021672,
      "balance_loss_clip": 1.05057967,
      "balance_loss_mlp": 1.01398849,
      "epoch": 0.6618168700775566,
      "flos": 17202109639680.0,
      "grad_norm": 3.1239876963814477,
      "language_loss": 0.82327569,
      "learning_rate": 1.0843233460787992e-06,
      "loss": 0.8452481,
      "num_input_tokens_seen": 118406790,
      "step": 5504,
      "time_per_iteration": 3.5447263717651367
    },
    {
      "auxiliary_loss_clip": 0.0117571,
      "auxiliary_loss_mlp": 0.01028263,
      "balance_loss_clip": 1.05287147,
      "balance_loss_mlp": 1.02051449,
      "epoch": 0.6619371129681958,
      "flos": 25447091448960.0,
      "grad_norm": 1.9616485253010403,
      "language_loss": 0.77905607,
      "learning_rate": 1.0836308821256805e-06,
      "loss": 0.80109572,
      "num_input_tokens_seen": 118427590,
      "step": 5505,
      "time_per_iteration": 3.5515871047973633
    },
    {
      "auxiliary_loss_clip": 0.01176128,
      "auxiliary_loss_mlp": 0.01026504,
      "balance_loss_clip": 1.05195594,
      "balance_loss_mlp": 1.01867771,
      "epoch": 0.6620573558588349,
      "flos": 18040444139520.0,
      "grad_norm": 2.31476366000882,
      "language_loss": 0.77950418,
      "learning_rate": 1.0829385571850282e-06,
      "loss": 0.80153048,
      "num_input_tokens_seen": 118444570,
      "step": 5506,
      "time_per_iteration": 2.6499295234680176
    },
    {
      "auxiliary_loss_clip": 0.01182502,
      "auxiliary_loss_mlp": 0.01030499,
      "balance_loss_clip": 1.05363894,
      "balance_loss_mlp": 1.02136755,
      "epoch": 0.6621775987494739,
      "flos": 17785586165760.0,
      "grad_norm": 3.5465968267941435,
      "language_loss": 0.84102339,
      "learning_rate": 1.0822463713618679e-06,
      "loss": 0.86315334,
      "num_input_tokens_seen": 118461425,
      "step": 5507,
      "time_per_iteration": 2.642101764678955
    },
    {
      "auxiliary_loss_clip": 0.0117663,
      "auxiliary_loss_mlp": 0.01033923,
      "balance_loss_clip": 1.0521487,
      "balance_loss_mlp": 1.02620387,
      "epoch": 0.6622978416401131,
      "flos": 17492590926720.0,
      "grad_norm": 2.9687120332116703,
      "language_loss": 0.85148251,
      "learning_rate": 1.0815543247612034e-06,
      "loss": 0.87358803,
      "num_input_tokens_seen": 118478495,
      "step": 5508,
      "time_per_iteration": 2.632850170135498
    },
    {
      "auxiliary_loss_clip": 0.01166331,
      "auxiliary_loss_mlp": 0.0102891,
      "balance_loss_clip": 1.05051494,
      "balance_loss_mlp": 1.02082729,
      "epoch": 0.6624180845307521,
      "flos": 21648352803840.0,
      "grad_norm": 1.6744467700535675,
      "language_loss": 0.82885122,
      "learning_rate": 1.0808624174880168e-06,
      "loss": 0.85080367,
      "num_input_tokens_seen": 118499145,
      "step": 5509,
      "time_per_iteration": 2.6142513751983643
    },
    {
      "auxiliary_loss_clip": 0.01111366,
      "auxiliary_loss_mlp": 0.01029347,
      "balance_loss_clip": 1.04643321,
      "balance_loss_mlp": 1.02194977,
      "epoch": 0.6625383274213912,
      "flos": 23805902108160.0,
      "grad_norm": 1.6859293708032124,
      "language_loss": 0.8002485,
      "learning_rate": 1.080170649647272e-06,
      "loss": 0.82165563,
      "num_input_tokens_seen": 118518950,
      "step": 5510,
      "time_per_iteration": 2.77362322807312
    },
    {
      "auxiliary_loss_clip": 0.01141542,
      "auxiliary_loss_mlp": 0.01025464,
      "balance_loss_clip": 1.04780841,
      "balance_loss_mlp": 1.01809716,
      "epoch": 0.6626585703120303,
      "flos": 33262941473280.0,
      "grad_norm": 1.7416329679061502,
      "language_loss": 0.67717195,
      "learning_rate": 1.0794790213439068e-06,
      "loss": 0.69884199,
      "num_input_tokens_seen": 118545850,
      "step": 5511,
      "time_per_iteration": 2.7785656452178955
    },
    {
      "auxiliary_loss_clip": 0.01167061,
      "auxiliary_loss_mlp": 0.01029786,
      "balance_loss_clip": 1.0538516,
      "balance_loss_mlp": 1.02136683,
      "epoch": 0.6627788132026694,
      "flos": 22085780630400.0,
      "grad_norm": 2.040210298459837,
      "language_loss": 0.78675485,
      "learning_rate": 1.078787532682843e-06,
      "loss": 0.80872333,
      "num_input_tokens_seen": 118563325,
      "step": 5512,
      "time_per_iteration": 2.6253507137298584
    },
    {
      "auxiliary_loss_clip": 0.01177294,
      "auxiliary_loss_mlp": 0.01029754,
      "balance_loss_clip": 1.05198967,
      "balance_loss_mlp": 1.02236009,
      "epoch": 0.6628990560933085,
      "flos": 36173608260480.0,
      "grad_norm": 2.14278150084487,
      "language_loss": 0.75811934,
      "learning_rate": 1.0780961837689773e-06,
      "loss": 0.78018987,
      "num_input_tokens_seen": 118582835,
      "step": 5513,
      "time_per_iteration": 2.7428581714630127
    },
    {
      "auxiliary_loss_clip": 0.01177725,
      "auxiliary_loss_mlp": 0.01026499,
      "balance_loss_clip": 1.05481386,
      "balance_loss_mlp": 1.01847911,
      "epoch": 0.6630192989839476,
      "flos": 18513567106560.0,
      "grad_norm": 1.7920104188583719,
      "language_loss": 0.70109236,
      "learning_rate": 1.0774049747071883e-06,
      "loss": 0.72313464,
      "num_input_tokens_seen": 118600715,
      "step": 5514,
      "time_per_iteration": 3.4662137031555176
    },
    {
      "auxiliary_loss_clip": 0.01169958,
      "auxiliary_loss_mlp": 0.01028805,
      "balance_loss_clip": 1.05540276,
      "balance_loss_mlp": 1.02108908,
      "epoch": 0.6631395418745867,
      "flos": 35809510049280.0,
      "grad_norm": 1.6364796213819814,
      "language_loss": 0.68283355,
      "learning_rate": 1.076713905602332e-06,
      "loss": 0.70482117,
      "num_input_tokens_seen": 118621290,
      "step": 5515,
      "time_per_iteration": 2.81115460395813
    },
    {
      "auxiliary_loss_clip": 0.0118183,
      "auxiliary_loss_mlp": 0.01033901,
      "balance_loss_clip": 1.05730152,
      "balance_loss_mlp": 1.02519321,
      "epoch": 0.6632597847652257,
      "flos": 20047742853120.0,
      "grad_norm": 1.679951506160503,
      "language_loss": 0.81403208,
      "learning_rate": 1.07602297655924e-06,
      "loss": 0.83618939,
      "num_input_tokens_seen": 118639610,
      "step": 5516,
      "time_per_iteration": 2.575620174407959
    },
    {
      "auxiliary_loss_clip": 0.01166912,
      "auxiliary_loss_mlp": 0.0102764,
      "balance_loss_clip": 1.05255294,
      "balance_loss_mlp": 1.0192951,
      "epoch": 0.6633800276558649,
      "flos": 21214480423680.0,
      "grad_norm": 1.8242404067385303,
      "language_loss": 0.81068218,
      "learning_rate": 1.0753321876827292e-06,
      "loss": 0.83262771,
      "num_input_tokens_seen": 118658895,
      "step": 5517,
      "time_per_iteration": 2.684159994125366
    },
    {
      "auxiliary_loss_clip": 0.01165086,
      "auxiliary_loss_mlp": 0.0102801,
      "balance_loss_clip": 1.04945922,
      "balance_loss_mlp": 1.02016628,
      "epoch": 0.663500270546504,
      "flos": 23987753688960.0,
      "grad_norm": 3.077753081158075,
      "language_loss": 0.74245262,
      "learning_rate": 1.0746415390775893e-06,
      "loss": 0.76438355,
      "num_input_tokens_seen": 118677025,
      "step": 5518,
      "time_per_iteration": 2.659950017929077
    },
    {
      "auxiliary_loss_clip": 0.01162083,
      "auxiliary_loss_mlp": 0.01028222,
      "balance_loss_clip": 1.05299044,
      "balance_loss_mlp": 1.02090287,
      "epoch": 0.663620513437143,
      "flos": 17932389050880.0,
      "grad_norm": 2.4353956604809714,
      "language_loss": 0.7677424,
      "learning_rate": 1.0739510308485939e-06,
      "loss": 0.78964543,
      "num_input_tokens_seen": 118694240,
      "step": 5519,
      "time_per_iteration": 2.5985639095306396
    },
    {
      "auxiliary_loss_clip": 0.01052053,
      "auxiliary_loss_mlp": 0.00752093,
      "balance_loss_clip": 1.01135492,
      "balance_loss_mlp": 1.00013816,
      "epoch": 0.6637407563277821,
      "flos": 57840241086720.0,
      "grad_norm": 0.8081147160168962,
      "language_loss": 0.6247673,
      "learning_rate": 1.07326066310049e-06,
      "loss": 0.64280874,
      "num_input_tokens_seen": 118758365,
      "step": 5520,
      "time_per_iteration": 3.2582905292510986
    },
    {
      "auxiliary_loss_clip": 0.01171808,
      "auxiliary_loss_mlp": 0.01032882,
      "balance_loss_clip": 1.05167317,
      "balance_loss_mlp": 1.02418566,
      "epoch": 0.6638609992184212,
      "flos": 27306007079040.0,
      "grad_norm": 1.8145016449114415,
      "language_loss": 0.79884565,
      "learning_rate": 1.0725704359380059e-06,
      "loss": 0.82089257,
      "num_input_tokens_seen": 118778220,
      "step": 5521,
      "time_per_iteration": 2.6303446292877197
    },
    {
      "auxiliary_loss_clip": 0.01178974,
      "auxiliary_loss_mlp": 0.01028577,
      "balance_loss_clip": 1.05377173,
      "balance_loss_mlp": 1.02057838,
      "epoch": 0.6639812421090603,
      "flos": 18624854419200.0,
      "grad_norm": 2.0928538813076494,
      "language_loss": 0.72392923,
      "learning_rate": 1.0718803494658497e-06,
      "loss": 0.74600476,
      "num_input_tokens_seen": 118797110,
      "step": 5522,
      "time_per_iteration": 2.59891676902771
    },
    {
      "auxiliary_loss_clip": 0.01153667,
      "auxiliary_loss_mlp": 0.01034245,
      "balance_loss_clip": 1.04779649,
      "balance_loss_mlp": 1.02649641,
      "epoch": 0.6641014849996993,
      "flos": 15924479806080.0,
      "grad_norm": 2.4620987369058716,
      "language_loss": 0.83662474,
      "learning_rate": 1.071190403788707e-06,
      "loss": 0.85850394,
      "num_input_tokens_seen": 118812415,
      "step": 5523,
      "time_per_iteration": 2.5770459175109863
    },
    {
      "auxiliary_loss_clip": 0.01148199,
      "auxiliary_loss_mlp": 0.01025985,
      "balance_loss_clip": 1.05077434,
      "balance_loss_mlp": 1.01787925,
      "epoch": 0.6642217278903385,
      "flos": 26505486622080.0,
      "grad_norm": 2.3058742896052626,
      "language_loss": 0.76141077,
      "learning_rate": 1.0705005990112415e-06,
      "loss": 0.78315264,
      "num_input_tokens_seen": 118832195,
      "step": 5524,
      "time_per_iteration": 2.7553224563598633
    },
    {
      "auxiliary_loss_clip": 0.01177589,
      "auxiliary_loss_mlp": 0.00762517,
      "balance_loss_clip": 1.05328083,
      "balance_loss_mlp": 1.00087452,
      "epoch": 0.6643419707809776,
      "flos": 15377308951680.0,
      "grad_norm": 2.570249675272785,
      "language_loss": 0.74874824,
      "learning_rate": 1.0698109352380957e-06,
      "loss": 0.76814932,
      "num_input_tokens_seen": 118849795,
      "step": 5525,
      "time_per_iteration": 2.556891441345215
    },
    {
      "auxiliary_loss_clip": 0.01145245,
      "auxiliary_loss_mlp": 0.01035107,
      "balance_loss_clip": 1.05009985,
      "balance_loss_mlp": 1.02669716,
      "epoch": 0.6644622136716166,
      "flos": 25117610970240.0,
      "grad_norm": 2.493805320189572,
      "language_loss": 0.78200924,
      "learning_rate": 1.0691214125738909e-06,
      "loss": 0.80381274,
      "num_input_tokens_seen": 118870000,
      "step": 5526,
      "time_per_iteration": 2.740760326385498
    },
    {
      "auxiliary_loss_clip": 0.0106805,
      "auxiliary_loss_mlp": 0.01001327,
      "balance_loss_clip": 1.01087666,
      "balance_loss_mlp": 1.00040269,
      "epoch": 0.6645824565622558,
      "flos": 66201717680640.0,
      "grad_norm": 0.8943926345420933,
      "language_loss": 0.57458436,
      "learning_rate": 1.0684320311232287e-06,
      "loss": 0.59527814,
      "num_input_tokens_seen": 118932905,
      "step": 5527,
      "time_per_iteration": 3.251999855041504
    },
    {
      "auxiliary_loss_clip": 0.011672,
      "auxiliary_loss_mlp": 0.01027645,
      "balance_loss_clip": 1.05224657,
      "balance_loss_mlp": 1.01955032,
      "epoch": 0.6647026994528948,
      "flos": 25082131311360.0,
      "grad_norm": 2.0133196500457347,
      "language_loss": 0.81485784,
      "learning_rate": 1.0677427909906865e-06,
      "loss": 0.8368063,
      "num_input_tokens_seen": 118953355,
      "step": 5528,
      "time_per_iteration": 2.7098822593688965
    },
    {
      "auxiliary_loss_clip": 0.01179931,
      "auxiliary_loss_mlp": 0.01029834,
      "balance_loss_clip": 1.05359578,
      "balance_loss_mlp": 1.02132869,
      "epoch": 0.6648229423435339,
      "flos": 18222187979520.0,
      "grad_norm": 1.7865489826783016,
      "language_loss": 0.72404927,
      "learning_rate": 1.0670536922808216e-06,
      "loss": 0.74614692,
      "num_input_tokens_seen": 118973480,
      "step": 5529,
      "time_per_iteration": 3.6466662883758545
    },
    {
      "auxiliary_loss_clip": 0.01174511,
      "auxiliary_loss_mlp": 0.01030471,
      "balance_loss_clip": 1.05153608,
      "balance_loss_mlp": 1.0226928,
      "epoch": 0.6649431852341731,
      "flos": 18296882311680.0,
      "grad_norm": 2.1119442153781836,
      "language_loss": 0.72084975,
      "learning_rate": 1.06636473509817e-06,
      "loss": 0.74289966,
      "num_input_tokens_seen": 118989860,
      "step": 5530,
      "time_per_iteration": 2.584517240524292
    },
    {
      "auxiliary_loss_clip": 0.01178077,
      "auxiliary_loss_mlp": 0.01031856,
      "balance_loss_clip": 1.05215085,
      "balance_loss_mlp": 1.02380383,
      "epoch": 0.6650634281248121,
      "flos": 17019575700480.0,
      "grad_norm": 2.0790572490442765,
      "language_loss": 0.80775976,
      "learning_rate": 1.0656759195472447e-06,
      "loss": 0.82985908,
      "num_input_tokens_seen": 119007150,
      "step": 5531,
      "time_per_iteration": 4.478582382202148
    },
    {
      "auxiliary_loss_clip": 0.0106811,
      "auxiliary_loss_mlp": 0.0100131,
      "balance_loss_clip": 1.01098144,
      "balance_loss_mlp": 1.00043988,
      "epoch": 0.6651836710154512,
      "flos": 69294810666240.0,
      "grad_norm": 0.7676200263110865,
      "language_loss": 0.59677958,
      "learning_rate": 1.0649872457325414e-06,
      "loss": 0.61747378,
      "num_input_tokens_seen": 119068435,
      "step": 5532,
      "time_per_iteration": 3.2248551845550537
    },
    {
      "auxiliary_loss_clip": 0.01059042,
      "auxiliary_loss_mlp": 0.0100175,
      "balance_loss_clip": 1.01081896,
      "balance_loss_mlp": 1.00083196,
      "epoch": 0.6653039139060903,
      "flos": 66883444882560.0,
      "grad_norm": 0.8467184009697789,
      "language_loss": 0.55051535,
      "learning_rate": 1.0642987137585278e-06,
      "loss": 0.5711233,
      "num_input_tokens_seen": 119127960,
      "step": 5533,
      "time_per_iteration": 3.167285680770874
    },
    {
      "auxiliary_loss_clip": 0.01158809,
      "auxiliary_loss_mlp": 0.01032136,
      "balance_loss_clip": 1.05139446,
      "balance_loss_mlp": 1.02348733,
      "epoch": 0.6654241567967294,
      "flos": 21470056669440.0,
      "grad_norm": 3.2276764688854547,
      "language_loss": 0.83051181,
      "learning_rate": 1.0636103237296561e-06,
      "loss": 0.85242128,
      "num_input_tokens_seen": 119146885,
      "step": 5534,
      "time_per_iteration": 2.7205991744995117
    },
    {
      "auxiliary_loss_clip": 0.01177371,
      "auxiliary_loss_mlp": 0.01024929,
      "balance_loss_clip": 1.05567741,
      "balance_loss_mlp": 1.01791382,
      "epoch": 0.6655443996873684,
      "flos": 25119514391040.0,
      "grad_norm": 15.749758791093978,
      "language_loss": 0.84078699,
      "learning_rate": 1.062922075750353e-06,
      "loss": 0.86281002,
      "num_input_tokens_seen": 119166900,
      "step": 5535,
      "time_per_iteration": 2.640183448791504
    },
    {
      "auxiliary_loss_clip": 0.01157787,
      "auxiliary_loss_mlp": 0.01028705,
      "balance_loss_clip": 1.05056477,
      "balance_loss_mlp": 1.02120113,
      "epoch": 0.6656646425780076,
      "flos": 17457326749440.0,
      "grad_norm": 2.507307473862099,
      "language_loss": 0.72145277,
      "learning_rate": 1.0622339699250267e-06,
      "loss": 0.74331772,
      "num_input_tokens_seen": 119184820,
      "step": 5536,
      "time_per_iteration": 2.5619447231292725
    },
    {
      "auxiliary_loss_clip": 0.01175663,
      "auxiliary_loss_mlp": 0.01032215,
      "balance_loss_clip": 1.05161738,
      "balance_loss_mlp": 1.02417445,
      "epoch": 0.6657848854686467,
      "flos": 23434190213760.0,
      "grad_norm": 1.8086205635711257,
      "language_loss": 0.79433024,
      "learning_rate": 1.0615460063580624e-06,
      "loss": 0.81640911,
      "num_input_tokens_seen": 119203295,
      "step": 5537,
      "time_per_iteration": 2.6241252422332764
    },
    {
      "auxiliary_loss_clip": 0.01159819,
      "auxiliary_loss_mlp": 0.0103071,
      "balance_loss_clip": 1.04963183,
      "balance_loss_mlp": 1.02271712,
      "epoch": 0.6659051283592857,
      "flos": 11509909459200.0,
      "grad_norm": 1.942031826332197,
      "language_loss": 0.73300058,
      "learning_rate": 1.060858185153821e-06,
      "loss": 0.75490582,
      "num_input_tokens_seen": 119221395,
      "step": 5538,
      "time_per_iteration": 2.6123523712158203
    },
    {
      "auxiliary_loss_clip": 0.01171842,
      "auxiliary_loss_mlp": 0.01030387,
      "balance_loss_clip": 1.0525918,
      "balance_loss_mlp": 1.02150607,
      "epoch": 0.6660253712499249,
      "flos": 20594554571520.0,
      "grad_norm": 2.3874802805611695,
      "language_loss": 0.76447636,
      "learning_rate": 1.0601705064166474e-06,
      "loss": 0.78649861,
      "num_input_tokens_seen": 119239790,
      "step": 5539,
      "time_per_iteration": 2.686508893966675
    },
    {
      "auxiliary_loss_clip": 0.01160477,
      "auxiliary_loss_mlp": 0.01025872,
      "balance_loss_clip": 1.04815578,
      "balance_loss_mlp": 1.01885986,
      "epoch": 0.666145614140564,
      "flos": 21251504367360.0,
      "grad_norm": 8.18725422089488,
      "language_loss": 0.73672283,
      "learning_rate": 1.0594829702508596e-06,
      "loss": 0.75858629,
      "num_input_tokens_seen": 119257505,
      "step": 5540,
      "time_per_iteration": 2.6901960372924805
    },
    {
      "auxiliary_loss_clip": 0.01177249,
      "auxiliary_loss_mlp": 0.00761974,
      "balance_loss_clip": 1.05402374,
      "balance_loss_mlp": 1.00085115,
      "epoch": 0.666265857031203,
      "flos": 33726188200320.0,
      "grad_norm": 1.5764218535804537,
      "language_loss": 0.54942161,
      "learning_rate": 1.0587955767607592e-06,
      "loss": 0.5688138,
      "num_input_tokens_seen": 119279365,
      "step": 5541,
      "time_per_iteration": 3.6619279384613037
    },
    {
      "auxiliary_loss_clip": 0.01154024,
      "auxiliary_loss_mlp": 0.01027016,
      "balance_loss_clip": 1.04920614,
      "balance_loss_mlp": 1.01911807,
      "epoch": 0.6663860999218422,
      "flos": 17456644391040.0,
      "grad_norm": 2.4643525618438367,
      "language_loss": 0.76704222,
      "learning_rate": 1.0581083260506206e-06,
      "loss": 0.78885269,
      "num_input_tokens_seen": 119296150,
      "step": 5542,
      "time_per_iteration": 2.5943946838378906
    },
    {
      "auxiliary_loss_clip": 0.01139707,
      "auxiliary_loss_mlp": 0.01026483,
      "balance_loss_clip": 1.04652274,
      "balance_loss_mlp": 1.0190711,
      "epoch": 0.6665063428124812,
      "flos": 17676740977920.0,
      "grad_norm": 2.07991098989037,
      "language_loss": 0.76829159,
      "learning_rate": 1.0574212182246993e-06,
      "loss": 0.78995347,
      "num_input_tokens_seen": 119314845,
      "step": 5543,
      "time_per_iteration": 2.7270116806030273
    },
    {
      "auxiliary_loss_clip": 0.01178738,
      "auxiliary_loss_mlp": 0.01029227,
      "balance_loss_clip": 1.05179501,
      "balance_loss_mlp": 1.02081084,
      "epoch": 0.6666265857031203,
      "flos": 27673265687040.0,
      "grad_norm": 3.0623709722005485,
      "language_loss": 0.75908393,
      "learning_rate": 1.0567342533872303e-06,
      "loss": 0.78116357,
      "num_input_tokens_seen": 119334875,
      "step": 5544,
      "time_per_iteration": 2.6849968433380127
    },
    {
      "auxiliary_loss_clip": 0.01165201,
      "auxiliary_loss_mlp": 0.01028081,
      "balance_loss_clip": 1.05194163,
      "balance_loss_mlp": 1.02000749,
      "epoch": 0.6667468285937594,
      "flos": 25046831220480.0,
      "grad_norm": 1.8465665750569247,
      "language_loss": 0.80889785,
      "learning_rate": 1.0560474316424255e-06,
      "loss": 0.83083075,
      "num_input_tokens_seen": 119354635,
      "step": 5545,
      "time_per_iteration": 2.6834182739257812
    },
    {
      "auxiliary_loss_clip": 0.01159942,
      "auxiliary_loss_mlp": 0.01025799,
      "balance_loss_clip": 1.04783428,
      "balance_loss_mlp": 1.01673269,
      "epoch": 0.6668670714843985,
      "flos": 22780472641920.0,
      "grad_norm": 2.4443261129237044,
      "language_loss": 0.73888886,
      "learning_rate": 1.0553607530944746e-06,
      "loss": 0.7607463,
      "num_input_tokens_seen": 119372690,
      "step": 5546,
      "time_per_iteration": 2.6815130710601807
    },
    {
      "auxiliary_loss_clip": 0.01149868,
      "auxiliary_loss_mlp": 0.01026495,
      "balance_loss_clip": 1.04925907,
      "balance_loss_mlp": 1.01807904,
      "epoch": 0.6669873143750376,
      "flos": 22163886754560.0,
      "grad_norm": 2.1191369169518968,
      "language_loss": 0.89762479,
      "learning_rate": 1.0546742178475463e-06,
      "loss": 0.91938841,
      "num_input_tokens_seen": 119391685,
      "step": 5547,
      "time_per_iteration": 2.6293253898620605
    },
    {
      "auxiliary_loss_clip": 0.01146375,
      "auxiliary_loss_mlp": 0.01030446,
      "balance_loss_clip": 1.05076778,
      "balance_loss_mlp": 1.02330852,
      "epoch": 0.6671075572656767,
      "flos": 20514832335360.0,
      "grad_norm": 2.080070561671904,
      "language_loss": 0.86074555,
      "learning_rate": 1.0539878260057868e-06,
      "loss": 0.88251376,
      "num_input_tokens_seen": 119410725,
      "step": 5548,
      "time_per_iteration": 2.713894844055176
    },
    {
      "auxiliary_loss_clip": 0.01167238,
      "auxiliary_loss_mlp": 0.01028891,
      "balance_loss_clip": 1.05323851,
      "balance_loss_mlp": 1.02032018,
      "epoch": 0.6672278001563158,
      "flos": 17931203902080.0,
      "grad_norm": 2.652218790790962,
      "language_loss": 0.68751287,
      "learning_rate": 1.0533015776733226e-06,
      "loss": 0.70947421,
      "num_input_tokens_seen": 119426875,
      "step": 5549,
      "time_per_iteration": 2.605836868286133
    },
    {
      "auxiliary_loss_clip": 0.01139623,
      "auxiliary_loss_mlp": 0.01035137,
      "balance_loss_clip": 1.04784119,
      "balance_loss_mlp": 1.02626801,
      "epoch": 0.6673480430469548,
      "flos": 22342146975360.0,
      "grad_norm": 2.2531912669680945,
      "language_loss": 0.79021835,
      "learning_rate": 1.0526154729542566e-06,
      "loss": 0.81196594,
      "num_input_tokens_seen": 119446935,
      "step": 5550,
      "time_per_iteration": 2.72446870803833
    },
    {
      "auxiliary_loss_clip": 0.01156374,
      "auxiliary_loss_mlp": 0.01028022,
      "balance_loss_clip": 1.04981804,
      "balance_loss_mlp": 1.0196836,
      "epoch": 0.6674682859375939,
      "flos": 20703830722560.0,
      "grad_norm": 2.480160286199085,
      "language_loss": 0.8045516,
      "learning_rate": 1.0519295119526699e-06,
      "loss": 0.82639551,
      "num_input_tokens_seen": 119463240,
      "step": 5551,
      "time_per_iteration": 2.729884147644043
    },
    {
      "auxiliary_loss_clip": 0.01179951,
      "auxiliary_loss_mlp": 0.01026732,
      "balance_loss_clip": 1.0542841,
      "balance_loss_mlp": 1.01890552,
      "epoch": 0.667588528828233,
      "flos": 26206673379840.0,
      "grad_norm": 2.513372506840752,
      "language_loss": 0.83311349,
      "learning_rate": 1.0512436947726227e-06,
      "loss": 0.85518026,
      "num_input_tokens_seen": 119484655,
      "step": 5552,
      "time_per_iteration": 2.6451141834259033
    },
    {
      "auxiliary_loss_clip": 0.01169435,
      "auxiliary_loss_mlp": 0.01026929,
      "balance_loss_clip": 1.05116904,
      "balance_loss_mlp": 1.01856089,
      "epoch": 0.6677087717188721,
      "flos": 23071025756160.0,
      "grad_norm": 2.1658386855869187,
      "language_loss": 0.65832967,
      "learning_rate": 1.0505580215181517e-06,
      "loss": 0.68029332,
      "num_input_tokens_seen": 119502895,
      "step": 5553,
      "time_per_iteration": 2.6819145679473877
    },
    {
      "auxiliary_loss_clip": 0.01067713,
      "auxiliary_loss_mlp": 0.01002807,
      "balance_loss_clip": 1.01059556,
      "balance_loss_mlp": 1.00188911,
      "epoch": 0.6678290146095112,
      "flos": 70941315219840.0,
      "grad_norm": 0.7789146056245186,
      "language_loss": 0.56564802,
      "learning_rate": 1.0498724922932753e-06,
      "loss": 0.58635324,
      "num_input_tokens_seen": 119561010,
      "step": 5554,
      "time_per_iteration": 3.1912145614624023
    },
    {
      "auxiliary_loss_clip": 0.01172718,
      "auxiliary_loss_mlp": 0.00762524,
      "balance_loss_clip": 1.05352449,
      "balance_loss_mlp": 1.00087154,
      "epoch": 0.6679492575001503,
      "flos": 18661088263680.0,
      "grad_norm": 2.3812370438259514,
      "language_loss": 0.87023306,
      "learning_rate": 1.0491871072019851e-06,
      "loss": 0.88958549,
      "num_input_tokens_seen": 119578900,
      "step": 5555,
      "time_per_iteration": 3.5341174602508545
    },
    {
      "auxiliary_loss_clip": 0.01178784,
      "auxiliary_loss_mlp": 0.01027644,
      "balance_loss_clip": 1.05297971,
      "balance_loss_mlp": 1.02014315,
      "epoch": 0.6680695003907894,
      "flos": 29711985822720.0,
      "grad_norm": 1.764935782706634,
      "language_loss": 0.64129758,
      "learning_rate": 1.0485018663482555e-06,
      "loss": 0.66336191,
      "num_input_tokens_seen": 119598920,
      "step": 5556,
      "time_per_iteration": 2.6826424598693848
    },
    {
      "auxiliary_loss_clip": 0.01179811,
      "auxiliary_loss_mlp": 0.01029673,
      "balance_loss_clip": 1.05418086,
      "balance_loss_mlp": 1.02159095,
      "epoch": 0.6681897432814284,
      "flos": 28218964083840.0,
      "grad_norm": 2.536006761893195,
      "language_loss": 0.71022463,
      "learning_rate": 1.0478167698360354e-06,
      "loss": 0.73231947,
      "num_input_tokens_seen": 119618220,
      "step": 5557,
      "time_per_iteration": 4.596595764160156
    },
    {
      "auxiliary_loss_clip": 0.01154318,
      "auxiliary_loss_mlp": 0.01025632,
      "balance_loss_clip": 1.04804659,
      "balance_loss_mlp": 1.01769614,
      "epoch": 0.6683099861720676,
      "flos": 25046543911680.0,
      "grad_norm": 2.9819564284174067,
      "language_loss": 0.70239854,
      "learning_rate": 1.0471318177692556e-06,
      "loss": 0.72419798,
      "num_input_tokens_seen": 119638520,
      "step": 5558,
      "time_per_iteration": 2.7211313247680664
    },
    {
      "auxiliary_loss_clip": 0.01165612,
      "auxiliary_loss_mlp": 0.0102808,
      "balance_loss_clip": 1.04910851,
      "balance_loss_mlp": 1.01994431,
      "epoch": 0.6684302290627067,
      "flos": 22996977868800.0,
      "grad_norm": 2.4133083042208945,
      "language_loss": 0.75533116,
      "learning_rate": 1.046447010251821e-06,
      "loss": 0.77726805,
      "num_input_tokens_seen": 119655850,
      "step": 5559,
      "time_per_iteration": 2.593857526779175
    },
    {
      "auxiliary_loss_clip": 0.0115242,
      "auxiliary_loss_mlp": 0.01030074,
      "balance_loss_clip": 1.05064678,
      "balance_loss_mlp": 1.02230191,
      "epoch": 0.6685504719533457,
      "flos": 26573824247040.0,
      "grad_norm": 1.9081477126389135,
      "language_loss": 0.7573998,
      "learning_rate": 1.0457623473876157e-06,
      "loss": 0.77922475,
      "num_input_tokens_seen": 119675355,
      "step": 5560,
      "time_per_iteration": 2.722200632095337
    },
    {
      "auxiliary_loss_clip": 0.01149036,
      "auxiliary_loss_mlp": 0.01027322,
      "balance_loss_clip": 1.05033135,
      "balance_loss_mlp": 1.02015424,
      "epoch": 0.6686707148439849,
      "flos": 28986087870720.0,
      "grad_norm": 1.8649703926760575,
      "language_loss": 0.71355546,
      "learning_rate": 1.0450778292805046e-06,
      "loss": 0.73531902,
      "num_input_tokens_seen": 119695340,
      "step": 5561,
      "time_per_iteration": 2.8134875297546387
    },
    {
      "auxiliary_loss_clip": 0.01179393,
      "auxiliary_loss_mlp": 0.01029495,
      "balance_loss_clip": 1.05285072,
      "balance_loss_mlp": 1.02169824,
      "epoch": 0.6687909577346239,
      "flos": 23623152687360.0,
      "grad_norm": 1.6461087353255557,
      "language_loss": 0.78890914,
      "learning_rate": 1.0443934560343267e-06,
      "loss": 0.81099796,
      "num_input_tokens_seen": 119716750,
      "step": 5562,
      "time_per_iteration": 2.7462031841278076
    },
    {
      "auxiliary_loss_clip": 0.01177139,
      "auxiliary_loss_mlp": 0.0102309,
      "balance_loss_clip": 1.05292511,
      "balance_loss_mlp": 1.01551127,
      "epoch": 0.668911200625263,
      "flos": 23148593176320.0,
      "grad_norm": 1.99972357211984,
      "language_loss": 0.77825046,
      "learning_rate": 1.0437092277529034e-06,
      "loss": 0.8002528,
      "num_input_tokens_seen": 119736005,
      "step": 5563,
      "time_per_iteration": 2.6584670543670654
    },
    {
      "auxiliary_loss_clip": 0.01176083,
      "auxiliary_loss_mlp": 0.01026654,
      "balance_loss_clip": 1.05300915,
      "balance_loss_mlp": 1.01954937,
      "epoch": 0.6690314435159022,
      "flos": 18551919853440.0,
      "grad_norm": 3.200891794236133,
      "language_loss": 0.74436033,
      "learning_rate": 1.0430251445400292e-06,
      "loss": 0.76638764,
      "num_input_tokens_seen": 119754050,
      "step": 5564,
      "time_per_iteration": 2.669586420059204
    },
    {
      "auxiliary_loss_clip": 0.01163035,
      "auxiliary_loss_mlp": 0.01027419,
      "balance_loss_clip": 1.05249417,
      "balance_loss_mlp": 1.01955128,
      "epoch": 0.6691516864065412,
      "flos": 31759540704000.0,
      "grad_norm": 2.695269900716533,
      "language_loss": 0.62983012,
      "learning_rate": 1.0423412064994787e-06,
      "loss": 0.65173471,
      "num_input_tokens_seen": 119774820,
      "step": 5565,
      "time_per_iteration": 2.7005276679992676
    },
    {
      "auxiliary_loss_clip": 0.01155394,
      "auxiliary_loss_mlp": 0.01027623,
      "balance_loss_clip": 1.04882407,
      "balance_loss_mlp": 1.02025604,
      "epoch": 0.6692719292971803,
      "flos": 34933864296960.0,
      "grad_norm": 2.005113876203503,
      "language_loss": 0.73820114,
      "learning_rate": 1.0416574137350064e-06,
      "loss": 0.76003134,
      "num_input_tokens_seen": 119795525,
      "step": 5566,
      "time_per_iteration": 3.688369035720825
    },
    {
      "auxiliary_loss_clip": 0.01150559,
      "auxiliary_loss_mlp": 0.01028356,
      "balance_loss_clip": 1.05335093,
      "balance_loss_mlp": 1.01971376,
      "epoch": 0.6693921721878194,
      "flos": 20449188230400.0,
      "grad_norm": 2.094277161815442,
      "language_loss": 0.81293768,
      "learning_rate": 1.0409737663503428e-06,
      "loss": 0.83472681,
      "num_input_tokens_seen": 119813905,
      "step": 5567,
      "time_per_iteration": 2.736104726791382
    },
    {
      "auxiliary_loss_clip": 0.01176084,
      "auxiliary_loss_mlp": 0.01025004,
      "balance_loss_clip": 1.05043089,
      "balance_loss_mlp": 1.01714778,
      "epoch": 0.6695124150784585,
      "flos": 16614538963200.0,
      "grad_norm": 1.8878528650181767,
      "language_loss": 0.83077657,
      "learning_rate": 1.040290264449196e-06,
      "loss": 0.85278738,
      "num_input_tokens_seen": 119832010,
      "step": 5568,
      "time_per_iteration": 2.5575110912323
    },
    {
      "auxiliary_loss_clip": 0.01172695,
      "auxiliary_loss_mlp": 0.01029132,
      "balance_loss_clip": 1.05177009,
      "balance_loss_mlp": 1.02165413,
      "epoch": 0.6696326579690975,
      "flos": 26652145852800.0,
      "grad_norm": 3.8300787952478736,
      "language_loss": 0.64004475,
      "learning_rate": 1.0396069081352532e-06,
      "loss": 0.662063,
      "num_input_tokens_seen": 119851165,
      "step": 5569,
      "time_per_iteration": 2.696648597717285
    },
    {
      "auxiliary_loss_clip": 0.0104497,
      "auxiliary_loss_mlp": 0.0100081,
      "balance_loss_clip": 1.00970018,
      "balance_loss_mlp": 0.99993342,
      "epoch": 0.6697529008597367,
      "flos": 66964603662720.0,
      "grad_norm": 0.773274139985125,
      "language_loss": 0.56014633,
      "learning_rate": 1.0389236975121782e-06,
      "loss": 0.58060414,
      "num_input_tokens_seen": 119906015,
      "step": 5570,
      "time_per_iteration": 3.129173517227173
    },
    {
      "auxiliary_loss_clip": 0.01177964,
      "auxiliary_loss_mlp": 0.01031705,
      "balance_loss_clip": 1.05178213,
      "balance_loss_mlp": 1.02339613,
      "epoch": 0.6698731437503758,
      "flos": 20886939279360.0,
      "grad_norm": 14.560776323642855,
      "language_loss": 0.71534717,
      "learning_rate": 1.0382406326836147e-06,
      "loss": 0.7374438,
      "num_input_tokens_seen": 119925160,
      "step": 5571,
      "time_per_iteration": 2.6366629600524902
    },
    {
      "auxiliary_loss_clip": 0.01135495,
      "auxiliary_loss_mlp": 0.01034529,
      "balance_loss_clip": 1.04913616,
      "balance_loss_mlp": 1.02620792,
      "epoch": 0.6699933866410148,
      "flos": 20409470766720.0,
      "grad_norm": 2.340426438655083,
      "language_loss": 0.75524712,
      "learning_rate": 1.0375577137531828e-06,
      "loss": 0.77694738,
      "num_input_tokens_seen": 119943720,
      "step": 5572,
      "time_per_iteration": 2.6918060779571533
    },
    {
      "auxiliary_loss_clip": 0.01111252,
      "auxiliary_loss_mlp": 0.00762511,
      "balance_loss_clip": 1.04346275,
      "balance_loss_mlp": 1.00088024,
      "epoch": 0.670113629531654,
      "flos": 29023075900800.0,
      "grad_norm": 1.7195280625268266,
      "language_loss": 0.72015333,
      "learning_rate": 1.0368749408244802e-06,
      "loss": 0.73889101,
      "num_input_tokens_seen": 119966640,
      "step": 5573,
      "time_per_iteration": 2.839231252670288
    },
    {
      "auxiliary_loss_clip": 0.01162992,
      "auxiliary_loss_mlp": 0.01026758,
      "balance_loss_clip": 1.05326653,
      "balance_loss_mlp": 1.01886916,
      "epoch": 0.670233872422293,
      "flos": 19791699730560.0,
      "grad_norm": 1.8453033962575305,
      "language_loss": 0.78500748,
      "learning_rate": 1.0361923140010836e-06,
      "loss": 0.80690497,
      "num_input_tokens_seen": 119985125,
      "step": 5574,
      "time_per_iteration": 2.6934945583343506
    },
    {
      "auxiliary_loss_clip": 0.01159098,
      "auxiliary_loss_mlp": 0.01028944,
      "balance_loss_clip": 1.04998147,
      "balance_loss_mlp": 1.02091503,
      "epoch": 0.6703541153129321,
      "flos": 24243689070720.0,
      "grad_norm": 2.4629913285549128,
      "language_loss": 0.63355219,
      "learning_rate": 1.0355098333865455e-06,
      "loss": 0.65543258,
      "num_input_tokens_seen": 120004355,
      "step": 5575,
      "time_per_iteration": 2.632002592086792
    },
    {
      "auxiliary_loss_clip": 0.0116206,
      "auxiliary_loss_mlp": 0.01031119,
      "balance_loss_clip": 1.05341971,
      "balance_loss_mlp": 1.02350771,
      "epoch": 0.6704743582035713,
      "flos": 26688523351680.0,
      "grad_norm": 1.8707541287287128,
      "language_loss": 0.69422227,
      "learning_rate": 1.0348274990844006e-06,
      "loss": 0.7161541,
      "num_input_tokens_seen": 120027115,
      "step": 5576,
      "time_per_iteration": 2.769282817840576
    },
    {
      "auxiliary_loss_clip": 0.0117774,
      "auxiliary_loss_mlp": 0.01030237,
      "balance_loss_clip": 1.05258036,
      "balance_loss_mlp": 1.02233326,
      "epoch": 0.6705946010942103,
      "flos": 23514379326720.0,
      "grad_norm": 1.7093524526973063,
      "language_loss": 0.72785532,
      "learning_rate": 1.034145311198155e-06,
      "loss": 0.74993515,
      "num_input_tokens_seen": 120047130,
      "step": 5577,
      "time_per_iteration": 2.6377854347229004
    },
    {
      "auxiliary_loss_clip": 0.01160954,
      "auxiliary_loss_mlp": 0.0103343,
      "balance_loss_clip": 1.04989195,
      "balance_loss_mlp": 1.02577662,
      "epoch": 0.6707148439848494,
      "flos": 24061011477120.0,
      "grad_norm": 1.8241151911754503,
      "language_loss": 0.64127481,
      "learning_rate": 1.0334632698312989e-06,
      "loss": 0.66321862,
      "num_input_tokens_seen": 120067925,
      "step": 5578,
      "time_per_iteration": 2.7050349712371826
    },
    {
      "auxiliary_loss_clip": 0.0116399,
      "auxiliary_loss_mlp": 0.0103221,
      "balance_loss_clip": 1.05053461,
      "balance_loss_mlp": 1.02390099,
      "epoch": 0.6708350868754885,
      "flos": 22528667324160.0,
      "grad_norm": 2.0658789740055883,
      "language_loss": 0.75349987,
      "learning_rate": 1.032781375087295e-06,
      "loss": 0.77546179,
      "num_input_tokens_seen": 120087825,
      "step": 5579,
      "time_per_iteration": 2.735915422439575
    },
    {
      "auxiliary_loss_clip": 0.01178902,
      "auxiliary_loss_mlp": 0.01035261,
      "balance_loss_clip": 1.05497468,
      "balance_loss_mlp": 1.02720261,
      "epoch": 0.6709553297661276,
      "flos": 25227749047680.0,
      "grad_norm": 1.3986370975460134,
      "language_loss": 0.67341721,
      "learning_rate": 1.0320996270695891e-06,
      "loss": 0.69555891,
      "num_input_tokens_seen": 120108895,
      "step": 5580,
      "time_per_iteration": 2.724308490753174
    },
    {
      "auxiliary_loss_clip": 0.0115138,
      "auxiliary_loss_mlp": 0.01026567,
      "balance_loss_clip": 1.04723096,
      "balance_loss_mlp": 1.01919949,
      "epoch": 0.6710755726567667,
      "flos": 20448757267200.0,
      "grad_norm": 1.7959541315522933,
      "language_loss": 0.73282421,
      "learning_rate": 1.0314180258815998e-06,
      "loss": 0.75460368,
      "num_input_tokens_seen": 120127535,
      "step": 5581,
      "time_per_iteration": 3.6621997356414795
    },
    {
      "auxiliary_loss_clip": 0.01155124,
      "auxiliary_loss_mlp": 0.01029225,
      "balance_loss_clip": 1.04870367,
      "balance_loss_mlp": 1.02199507,
      "epoch": 0.6711958155474057,
      "flos": 25995411538560.0,
      "grad_norm": 1.597263375335692,
      "language_loss": 0.74228728,
      "learning_rate": 1.0307365716267247e-06,
      "loss": 0.76413071,
      "num_input_tokens_seen": 120147980,
      "step": 5582,
      "time_per_iteration": 2.7158608436584473
    },
    {
      "auxiliary_loss_clip": 0.01167578,
      "auxiliary_loss_mlp": 0.01028031,
      "balance_loss_clip": 1.05095196,
      "balance_loss_mlp": 1.02006841,
      "epoch": 0.6713160584380449,
      "flos": 19937712516480.0,
      "grad_norm": 2.372488978840252,
      "language_loss": 0.78295183,
      "learning_rate": 1.0300552644083423e-06,
      "loss": 0.80490792,
      "num_input_tokens_seen": 120166905,
      "step": 5583,
      "time_per_iteration": 4.623666286468506
    },
    {
      "auxiliary_loss_clip": 0.0117922,
      "auxiliary_loss_mlp": 0.01029702,
      "balance_loss_clip": 1.05239177,
      "balance_loss_mlp": 1.02126193,
      "epoch": 0.6714363013286839,
      "flos": 18223373128320.0,
      "grad_norm": 2.6744907285922386,
      "language_loss": 0.73166907,
      "learning_rate": 1.0293741043298036e-06,
      "loss": 0.75375825,
      "num_input_tokens_seen": 120185255,
      "step": 5584,
      "time_per_iteration": 2.6463425159454346
    },
    {
      "auxiliary_loss_clip": 0.01183544,
      "auxiliary_loss_mlp": 0.01026669,
      "balance_loss_clip": 1.05596948,
      "balance_loss_mlp": 1.01830697,
      "epoch": 0.671556544219323,
      "flos": 25812374808960.0,
      "grad_norm": 5.0514697982287755,
      "language_loss": 0.71779978,
      "learning_rate": 1.0286930914944436e-06,
      "loss": 0.7399019,
      "num_input_tokens_seen": 120205070,
      "step": 5585,
      "time_per_iteration": 2.6340510845184326
    },
    {
      "auxiliary_loss_clip": 0.01161134,
      "auxiliary_loss_mlp": 0.01025123,
      "balance_loss_clip": 1.04816198,
      "balance_loss_mlp": 1.01733232,
      "epoch": 0.6716767871099621,
      "flos": 15850431918720.0,
      "grad_norm": 2.3444789643088915,
      "language_loss": 0.77833116,
      "learning_rate": 1.0280122260055684e-06,
      "loss": 0.80019373,
      "num_input_tokens_seen": 120220780,
      "step": 5586,
      "time_per_iteration": 2.655876636505127
    },
    {
      "auxiliary_loss_clip": 0.0117932,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.05341887,
      "balance_loss_mlp": 1.01915848,
      "epoch": 0.6717970300006012,
      "flos": 19756112330880.0,
      "grad_norm": 2.765035661166744,
      "language_loss": 0.82287008,
      "learning_rate": 1.0273315079664652e-06,
      "loss": 0.84494102,
      "num_input_tokens_seen": 120238735,
      "step": 5587,
      "time_per_iteration": 2.5517518520355225
    },
    {
      "auxiliary_loss_clip": 0.01147575,
      "auxiliary_loss_mlp": 0.01025504,
      "balance_loss_clip": 1.04834509,
      "balance_loss_mlp": 1.01759398,
      "epoch": 0.6719172728912403,
      "flos": 25485049146240.0,
      "grad_norm": 2.5194059029412275,
      "language_loss": 0.74283642,
      "learning_rate": 1.0266509374803992e-06,
      "loss": 0.76456726,
      "num_input_tokens_seen": 120259895,
      "step": 5588,
      "time_per_iteration": 2.6832351684570312
    },
    {
      "auxiliary_loss_clip": 0.01168191,
      "auxiliary_loss_mlp": 0.0102966,
      "balance_loss_clip": 1.05074942,
      "balance_loss_mlp": 1.02176273,
      "epoch": 0.6720375157818794,
      "flos": 15880344969600.0,
      "grad_norm": 2.7184589723562125,
      "language_loss": 0.84734398,
      "learning_rate": 1.0259705146506123e-06,
      "loss": 0.86932242,
      "num_input_tokens_seen": 120274790,
      "step": 5589,
      "time_per_iteration": 2.6213254928588867
    },
    {
      "auxiliary_loss_clip": 0.01142809,
      "auxiliary_loss_mlp": 0.01032105,
      "balance_loss_clip": 1.05012286,
      "balance_loss_mlp": 1.02427936,
      "epoch": 0.6721577586725185,
      "flos": 32010843231360.0,
      "grad_norm": 3.0823086020591552,
      "language_loss": 0.77580565,
      "learning_rate": 1.025290239580324e-06,
      "loss": 0.79755473,
      "num_input_tokens_seen": 120295460,
      "step": 5590,
      "time_per_iteration": 2.794644355773926
    },
    {
      "auxiliary_loss_clip": 0.01159731,
      "auxiliary_loss_mlp": 0.01032269,
      "balance_loss_clip": 1.05041003,
      "balance_loss_mlp": 1.02365589,
      "epoch": 0.6722780015631575,
      "flos": 20737873837440.0,
      "grad_norm": 1.7415887248647213,
      "language_loss": 0.75465167,
      "learning_rate": 1.0246101123727313e-06,
      "loss": 0.77657163,
      "num_input_tokens_seen": 120314440,
      "step": 5591,
      "time_per_iteration": 2.6016650199890137
    },
    {
      "auxiliary_loss_clip": 0.01160144,
      "auxiliary_loss_mlp": 0.01028034,
      "balance_loss_clip": 1.05258346,
      "balance_loss_mlp": 1.02030921,
      "epoch": 0.6723982444537967,
      "flos": 16909617191040.0,
      "grad_norm": 2.4313843293275554,
      "language_loss": 0.78683096,
      "learning_rate": 1.0239301331310085e-06,
      "loss": 0.80871272,
      "num_input_tokens_seen": 120332060,
      "step": 5592,
      "time_per_iteration": 3.609447956085205
    },
    {
      "auxiliary_loss_clip": 0.01159379,
      "auxiliary_loss_mlp": 0.0102996,
      "balance_loss_clip": 1.0522064,
      "balance_loss_mlp": 1.02246785,
      "epoch": 0.6725184873444358,
      "flos": 20667812359680.0,
      "grad_norm": 1.6536242118082807,
      "language_loss": 0.88891983,
      "learning_rate": 1.0232503019583088e-06,
      "loss": 0.91081327,
      "num_input_tokens_seen": 120351670,
      "step": 5593,
      "time_per_iteration": 2.777660369873047
    },
    {
      "auxiliary_loss_clip": 0.01153671,
      "auxiliary_loss_mlp": 0.01026822,
      "balance_loss_clip": 1.04721379,
      "balance_loss_mlp": 1.01897848,
      "epoch": 0.6726387302350748,
      "flos": 23727616416000.0,
      "grad_norm": 2.1403340459211178,
      "language_loss": 0.70008159,
      "learning_rate": 1.0225706189577619e-06,
      "loss": 0.72188652,
      "num_input_tokens_seen": 120370195,
      "step": 5594,
      "time_per_iteration": 2.7274954319000244
    },
    {
      "auxiliary_loss_clip": 0.01131277,
      "auxiliary_loss_mlp": 0.01030323,
      "balance_loss_clip": 1.04623473,
      "balance_loss_mlp": 1.02176929,
      "epoch": 0.672758973125714,
      "flos": 15188274650880.0,
      "grad_norm": 1.978045624516381,
      "language_loss": 0.74886197,
      "learning_rate": 1.021891084232475e-06,
      "loss": 0.77047801,
      "num_input_tokens_seen": 120388130,
      "step": 5595,
      "time_per_iteration": 2.8138628005981445
    },
    {
      "auxiliary_loss_clip": 0.01177501,
      "auxiliary_loss_mlp": 0.01021981,
      "balance_loss_clip": 1.05411935,
      "balance_loss_mlp": 1.0139761,
      "epoch": 0.672879216016353,
      "flos": 18077252601600.0,
      "grad_norm": 2.665687750620504,
      "language_loss": 0.79894722,
      "learning_rate": 1.0212116978855325e-06,
      "loss": 0.8209421,
      "num_input_tokens_seen": 120406145,
      "step": 5596,
      "time_per_iteration": 2.5670971870422363
    },
    {
      "auxiliary_loss_clip": 0.01175456,
      "auxiliary_loss_mlp": 0.01027103,
      "balance_loss_clip": 1.05106163,
      "balance_loss_mlp": 1.0191282,
      "epoch": 0.6729994589069921,
      "flos": 23476349802240.0,
      "grad_norm": 1.9127752446013606,
      "language_loss": 0.78991461,
      "learning_rate": 1.020532460019997e-06,
      "loss": 0.81194019,
      "num_input_tokens_seen": 120425395,
      "step": 5597,
      "time_per_iteration": 2.6372158527374268
    },
    {
      "auxiliary_loss_clip": 0.01162973,
      "auxiliary_loss_mlp": 0.01029268,
      "balance_loss_clip": 1.05252886,
      "balance_loss_mlp": 1.02106035,
      "epoch": 0.6731197017976313,
      "flos": 26322018929280.0,
      "grad_norm": 1.856446838820633,
      "language_loss": 0.71174198,
      "learning_rate": 1.0198533707389096e-06,
      "loss": 0.73366439,
      "num_input_tokens_seen": 120446270,
      "step": 5598,
      "time_per_iteration": 2.6458256244659424
    },
    {
      "auxiliary_loss_clip": 0.0117788,
      "auxiliary_loss_mlp": 0.01029698,
      "balance_loss_clip": 1.05331242,
      "balance_loss_mlp": 1.02123427,
      "epoch": 0.6732399446882703,
      "flos": 21616428591360.0,
      "grad_norm": 2.0909302119919464,
      "language_loss": 0.7327317,
      "learning_rate": 1.0191744301452853e-06,
      "loss": 0.75480747,
      "num_input_tokens_seen": 120465570,
      "step": 5599,
      "time_per_iteration": 2.6571593284606934
    },
    {
      "auxiliary_loss_clip": 0.01158049,
      "auxiliary_loss_mlp": 0.00761775,
      "balance_loss_clip": 1.04770386,
      "balance_loss_mlp": 1.00083041,
      "epoch": 0.6733601875789094,
      "flos": 25880173729920.0,
      "grad_norm": 1.6437271208334088,
      "language_loss": 0.7056942,
      "learning_rate": 1.0184956383421208e-06,
      "loss": 0.72489244,
      "num_input_tokens_seen": 120484220,
      "step": 5600,
      "time_per_iteration": 2.648007392883301
    },
    {
      "auxiliary_loss_clip": 0.01177668,
      "auxiliary_loss_mlp": 0.01026604,
      "balance_loss_clip": 1.05399704,
      "balance_loss_mlp": 1.01864707,
      "epoch": 0.6734804304695485,
      "flos": 22929573997440.0,
      "grad_norm": 2.380670443734864,
      "language_loss": 0.65507698,
      "learning_rate": 1.017816995432387e-06,
      "loss": 0.67711973,
      "num_input_tokens_seen": 120503320,
      "step": 5601,
      "time_per_iteration": 2.628929376602173
    },
    {
      "auxiliary_loss_clip": 0.01178884,
      "auxiliary_loss_mlp": 0.01028023,
      "balance_loss_clip": 1.05407846,
      "balance_loss_mlp": 1.02004242,
      "epoch": 0.6736006733601876,
      "flos": 18697968552960.0,
      "grad_norm": 2.3947845329504185,
      "language_loss": 0.73946053,
      "learning_rate": 1.0171385015190353e-06,
      "loss": 0.76152956,
      "num_input_tokens_seen": 120523180,
      "step": 5602,
      "time_per_iteration": 2.5696678161621094
    },
    {
      "auxiliary_loss_clip": 0.01175202,
      "auxiliary_loss_mlp": 0.01026062,
      "balance_loss_clip": 1.05333328,
      "balance_loss_mlp": 1.01844466,
      "epoch": 0.6737209162508266,
      "flos": 19427745173760.0,
      "grad_norm": 1.9642149188527995,
      "language_loss": 0.73121226,
      "learning_rate": 1.0164601567049908e-06,
      "loss": 0.75322485,
      "num_input_tokens_seen": 120541710,
      "step": 5603,
      "time_per_iteration": 2.6135103702545166
    },
    {
      "auxiliary_loss_clip": 0.01148227,
      "auxiliary_loss_mlp": 0.01035388,
      "balance_loss_clip": 1.04818439,
      "balance_loss_mlp": 1.02657866,
      "epoch": 0.6738411591414658,
      "flos": 20158060498560.0,
      "grad_norm": 1.824685211682053,
      "language_loss": 0.80813622,
      "learning_rate": 1.015781961093158e-06,
      "loss": 0.82997239,
      "num_input_tokens_seen": 120561030,
      "step": 5604,
      "time_per_iteration": 2.616518497467041
    },
    {
      "auxiliary_loss_clip": 0.01162518,
      "auxiliary_loss_mlp": 0.01026388,
      "balance_loss_clip": 1.0514729,
      "balance_loss_mlp": 1.0188303,
      "epoch": 0.6739614020321049,
      "flos": 21653847584640.0,
      "grad_norm": 1.6211724514379184,
      "language_loss": 0.77232802,
      "learning_rate": 1.0151039147864197e-06,
      "loss": 0.79421711,
      "num_input_tokens_seen": 120581005,
      "step": 5605,
      "time_per_iteration": 2.680450916290283
    },
    {
      "auxiliary_loss_clip": 0.01182093,
      "auxiliary_loss_mlp": 0.01032832,
      "balance_loss_clip": 1.05402184,
      "balance_loss_mlp": 1.02381968,
      "epoch": 0.6740816449227439,
      "flos": 19171702051200.0,
      "grad_norm": 2.185505976081161,
      "language_loss": 0.66406167,
      "learning_rate": 1.0144260178876336e-06,
      "loss": 0.68621087,
      "num_input_tokens_seen": 120600350,
      "step": 5606,
      "time_per_iteration": 2.6357972621917725
    },
    {
      "auxiliary_loss_clip": 0.01143398,
      "auxiliary_loss_mlp": 0.01027086,
      "balance_loss_clip": 1.04852927,
      "balance_loss_mlp": 1.01929533,
      "epoch": 0.6742018878133831,
      "flos": 21097015971840.0,
      "grad_norm": 3.003496916989272,
      "language_loss": 0.67276156,
      "learning_rate": 1.0137482704996388e-06,
      "loss": 0.69446635,
      "num_input_tokens_seen": 120614700,
      "step": 5607,
      "time_per_iteration": 3.609151601791382
    },
    {
      "auxiliary_loss_clip": 0.0113841,
      "auxiliary_loss_mlp": 0.01026993,
      "balance_loss_clip": 1.04826605,
      "balance_loss_mlp": 1.01866007,
      "epoch": 0.6743221307040221,
      "flos": 23549966726400.0,
      "grad_norm": 2.016754051594134,
      "language_loss": 0.79119253,
      "learning_rate": 1.0130706727252461e-06,
      "loss": 0.81284654,
      "num_input_tokens_seen": 120631755,
      "step": 5608,
      "time_per_iteration": 2.7364625930786133
    },
    {
      "auxiliary_loss_clip": 0.01176585,
      "auxiliary_loss_mlp": 0.00761599,
      "balance_loss_clip": 1.05225539,
      "balance_loss_mlp": 1.00079978,
      "epoch": 0.6744423735946612,
      "flos": 16249542912000.0,
      "grad_norm": 3.7807064291690557,
      "language_loss": 0.69074011,
      "learning_rate": 1.0123932246672468e-06,
      "loss": 0.71012187,
      "num_input_tokens_seen": 120645900,
      "step": 5609,
      "time_per_iteration": 4.514845371246338
    },
    {
      "auxiliary_loss_clip": 0.01047692,
      "auxiliary_loss_mlp": 0.00999098,
      "balance_loss_clip": 1.01140165,
      "balance_loss_mlp": 0.99817991,
      "epoch": 0.6745626164853004,
      "flos": 57843257829120.0,
      "grad_norm": 0.7504449766068347,
      "language_loss": 0.55732781,
      "learning_rate": 1.0117159264284114e-06,
      "loss": 0.57779574,
      "num_input_tokens_seen": 120709070,
      "step": 5610,
      "time_per_iteration": 3.1790378093719482
    },
    {
      "auxiliary_loss_clip": 0.01161934,
      "auxiliary_loss_mlp": 0.010273,
      "balance_loss_clip": 1.05213702,
      "balance_loss_mlp": 1.01963818,
      "epoch": 0.6746828593759394,
      "flos": 20485027025280.0,
      "grad_norm": 1.7580765048908655,
      "language_loss": 0.77131009,
      "learning_rate": 1.0110387781114837e-06,
      "loss": 0.79320246,
      "num_input_tokens_seen": 120727685,
      "step": 5611,
      "time_per_iteration": 2.674243450164795
    },
    {
      "auxiliary_loss_clip": 0.01175972,
      "auxiliary_loss_mlp": 0.01028853,
      "balance_loss_clip": 1.0521332,
      "balance_loss_mlp": 1.02086616,
      "epoch": 0.6748031022665785,
      "flos": 19208223204480.0,
      "grad_norm": 6.13563299274478,
      "language_loss": 0.77121556,
      "learning_rate": 1.0103617798191872e-06,
      "loss": 0.79326385,
      "num_input_tokens_seen": 120747160,
      "step": 5612,
      "time_per_iteration": 2.58469557762146
    },
    {
      "auxiliary_loss_clip": 0.01163738,
      "auxiliary_loss_mlp": 0.01027912,
      "balance_loss_clip": 1.05183744,
      "balance_loss_mlp": 1.01983547,
      "epoch": 0.6749233451572175,
      "flos": 15195026407680.0,
      "grad_norm": 2.2729277592565245,
      "language_loss": 0.83371651,
      "learning_rate": 1.0096849316542217e-06,
      "loss": 0.85563302,
      "num_input_tokens_seen": 120763710,
      "step": 5613,
      "time_per_iteration": 2.71962571144104
    },
    {
      "auxiliary_loss_clip": 0.01177956,
      "auxiliary_loss_mlp": 0.0102489,
      "balance_loss_clip": 1.05225968,
      "balance_loss_mlp": 1.01674247,
      "epoch": 0.6750435880478567,
      "flos": 26499489050880.0,
      "grad_norm": 2.1714942510619455,
      "language_loss": 0.75020194,
      "learning_rate": 1.0090082337192643e-06,
      "loss": 0.77223051,
      "num_input_tokens_seen": 120783355,
      "step": 5614,
      "time_per_iteration": 2.6431455612182617
    },
    {
      "auxiliary_loss_clip": 0.01150579,
      "auxiliary_loss_mlp": 0.01034886,
      "balance_loss_clip": 1.04667592,
      "balance_loss_mlp": 1.02763176,
      "epoch": 0.6751638309384957,
      "flos": 23404313076480.0,
      "grad_norm": 2.0800140478671802,
      "language_loss": 0.78524143,
      "learning_rate": 1.0083316861169705e-06,
      "loss": 0.80709612,
      "num_input_tokens_seen": 120802090,
      "step": 5615,
      "time_per_iteration": 2.7272160053253174
    },
    {
      "auxiliary_loss_clip": 0.01141107,
      "auxiliary_loss_mlp": 0.01027649,
      "balance_loss_clip": 1.04812884,
      "balance_loss_mlp": 1.01887548,
      "epoch": 0.6752840738291348,
      "flos": 23441408847360.0,
      "grad_norm": 2.4726027904192818,
      "language_loss": 0.71890759,
      "learning_rate": 1.0076552889499713e-06,
      "loss": 0.7405951,
      "num_input_tokens_seen": 120822855,
      "step": 5616,
      "time_per_iteration": 2.7076756954193115
    },
    {
      "auxiliary_loss_clip": 0.01178429,
      "auxiliary_loss_mlp": 0.01032557,
      "balance_loss_clip": 1.05486238,
      "balance_loss_mlp": 1.02504063,
      "epoch": 0.675404316719774,
      "flos": 30335826257280.0,
      "grad_norm": 2.102035492262149,
      "language_loss": 0.73410565,
      "learning_rate": 1.006979042320876e-06,
      "loss": 0.75621545,
      "num_input_tokens_seen": 120843070,
      "step": 5617,
      "time_per_iteration": 2.645235300064087
    },
    {
      "auxiliary_loss_clip": 0.01160087,
      "auxiliary_loss_mlp": 0.01020621,
      "balance_loss_clip": 1.05038118,
      "balance_loss_mlp": 1.01353693,
      "epoch": 0.675524559610413,
      "flos": 23622613983360.0,
      "grad_norm": 3.0739246197775607,
      "language_loss": 0.63751137,
      "learning_rate": 1.0063029463322702e-06,
      "loss": 0.65931845,
      "num_input_tokens_seen": 120863345,
      "step": 5618,
      "time_per_iteration": 2.7070741653442383
    },
    {
      "auxiliary_loss_clip": 0.01176126,
      "auxiliary_loss_mlp": 0.01030109,
      "balance_loss_clip": 1.0501802,
      "balance_loss_mlp": 1.0219965,
      "epoch": 0.6756448025010521,
      "flos": 21248631279360.0,
      "grad_norm": 3.1673269900033163,
      "language_loss": 0.75438404,
      "learning_rate": 1.0056270010867164e-06,
      "loss": 0.7764464,
      "num_input_tokens_seen": 120880915,
      "step": 5619,
      "time_per_iteration": 3.5296108722686768
    },
    {
      "auxiliary_loss_clip": 0.0117817,
      "auxiliary_loss_mlp": 0.01028563,
      "balance_loss_clip": 1.05268145,
      "balance_loss_mlp": 1.02046275,
      "epoch": 0.6757650453916912,
      "flos": 21646521210240.0,
      "grad_norm": 2.4130843587403996,
      "language_loss": 0.78954911,
      "learning_rate": 1.004951206686758e-06,
      "loss": 0.81161642,
      "num_input_tokens_seen": 120899190,
      "step": 5620,
      "time_per_iteration": 2.688469886779785
    },
    {
      "auxiliary_loss_clip": 0.01166559,
      "auxiliary_loss_mlp": 0.01030861,
      "balance_loss_clip": 1.052472,
      "balance_loss_mlp": 1.02274299,
      "epoch": 0.6758852882823303,
      "flos": 21795658479360.0,
      "grad_norm": 1.890354722412366,
      "language_loss": 0.7183249,
      "learning_rate": 1.0042755632349087e-06,
      "loss": 0.74029911,
      "num_input_tokens_seen": 120916080,
      "step": 5621,
      "time_per_iteration": 2.651280641555786
    },
    {
      "auxiliary_loss_clip": 0.0117845,
      "auxiliary_loss_mlp": 0.01027069,
      "balance_loss_clip": 1.05347049,
      "balance_loss_mlp": 1.01892662,
      "epoch": 0.6760055311729694,
      "flos": 27088783580160.0,
      "grad_norm": 2.1168636175609783,
      "language_loss": 0.62500751,
      "learning_rate": 1.0036000708336653e-06,
      "loss": 0.64706272,
      "num_input_tokens_seen": 120935210,
      "step": 5622,
      "time_per_iteration": 2.6613411903381348
    },
    {
      "auxiliary_loss_clip": 0.01177253,
      "auxiliary_loss_mlp": 0.01027995,
      "balance_loss_clip": 1.05398905,
      "balance_loss_mlp": 1.02037168,
      "epoch": 0.6761257740636085,
      "flos": 17999792922240.0,
      "grad_norm": 2.1406740476343695,
      "language_loss": 0.79881388,
      "learning_rate": 1.0029247295854984e-06,
      "loss": 0.82086641,
      "num_input_tokens_seen": 120951830,
      "step": 5623,
      "time_per_iteration": 2.581899642944336
    },
    {
      "auxiliary_loss_clip": 0.01134107,
      "auxiliary_loss_mlp": 0.01029999,
      "balance_loss_clip": 1.04757357,
      "balance_loss_mlp": 1.02251267,
      "epoch": 0.6762460169542476,
      "flos": 15121912273920.0,
      "grad_norm": 1.7881580737652,
      "language_loss": 0.72008085,
      "learning_rate": 1.0022495395928588e-06,
      "loss": 0.74172187,
      "num_input_tokens_seen": 120970310,
      "step": 5624,
      "time_per_iteration": 2.731292247772217
    },
    {
      "auxiliary_loss_clip": 0.01068405,
      "auxiliary_loss_mlp": 0.00999847,
      "balance_loss_clip": 1.01172423,
      "balance_loss_mlp": 0.99894744,
      "epoch": 0.6763662598448866,
      "flos": 67886970030720.0,
      "grad_norm": 0.7860983062888938,
      "language_loss": 0.62343585,
      "learning_rate": 1.0015745009581697e-06,
      "loss": 0.64411843,
      "num_input_tokens_seen": 121031915,
      "step": 5625,
      "time_per_iteration": 3.2522175312042236
    },
    {
      "auxiliary_loss_clip": 0.01146911,
      "auxiliary_loss_mlp": 0.01029777,
      "balance_loss_clip": 1.05010879,
      "balance_loss_mlp": 1.02222514,
      "epoch": 0.6764865027355258,
      "flos": 20631829910400.0,
      "grad_norm": 3.734726648522586,
      "language_loss": 0.66705787,
      "learning_rate": 1.0008996137838343e-06,
      "loss": 0.68882477,
      "num_input_tokens_seen": 121050890,
      "step": 5626,
      "time_per_iteration": 2.692321300506592
    },
    {
      "auxiliary_loss_clip": 0.01181816,
      "auxiliary_loss_mlp": 0.01030235,
      "balance_loss_clip": 1.05334926,
      "balance_loss_mlp": 1.02140188,
      "epoch": 0.6766067456261649,
      "flos": 21215809226880.0,
      "grad_norm": 2.4190325610797134,
      "language_loss": 0.7999391,
      "learning_rate": 1.000224878172234e-06,
      "loss": 0.82205963,
      "num_input_tokens_seen": 121070015,
      "step": 5627,
      "time_per_iteration": 2.625767946243286
    },
    {
      "auxiliary_loss_clip": 0.01175022,
      "auxiliary_loss_mlp": 0.01028108,
      "balance_loss_clip": 1.05258775,
      "balance_loss_mlp": 1.02048469,
      "epoch": 0.6767269885168039,
      "flos": 19938251220480.0,
      "grad_norm": 2.2102677862637594,
      "language_loss": 0.72656524,
      "learning_rate": 9.99550294225724e-07,
      "loss": 0.74859655,
      "num_input_tokens_seen": 121089170,
      "step": 5628,
      "time_per_iteration": 2.578357696533203
    },
    {
      "auxiliary_loss_clip": 0.01177383,
      "auxiliary_loss_mlp": 0.01025954,
      "balance_loss_clip": 1.05239582,
      "balance_loss_mlp": 1.0183959,
      "epoch": 0.6768472314074431,
      "flos": 20814076540800.0,
      "grad_norm": 2.5349619065932507,
      "language_loss": 0.72712189,
      "learning_rate": 9.988758620466402e-07,
      "loss": 0.74915528,
      "num_input_tokens_seen": 121108040,
      "step": 5629,
      "time_per_iteration": 2.65936541557312
    },
    {
      "auxiliary_loss_clip": 0.01160061,
      "auxiliary_loss_mlp": 0.01028055,
      "balance_loss_clip": 1.05191529,
      "balance_loss_mlp": 1.02048552,
      "epoch": 0.6769674742980821,
      "flos": 23186012169600.0,
      "grad_norm": 1.6454343552421704,
      "language_loss": 0.76142728,
      "learning_rate": 9.982015817372917e-07,
      "loss": 0.78330845,
      "num_input_tokens_seen": 121128480,
      "step": 5630,
      "time_per_iteration": 2.6240622997283936
    },
    {
      "auxiliary_loss_clip": 0.01140324,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 1.04935789,
      "balance_loss_mlp": 1.02043796,
      "epoch": 0.6770877171887212,
      "flos": 24242934885120.0,
      "grad_norm": 2.0960127879667723,
      "language_loss": 0.82437313,
      "learning_rate": 9.975274533999657e-07,
      "loss": 0.84606051,
      "num_input_tokens_seen": 121148010,
      "step": 5631,
      "time_per_iteration": 2.7658753395080566
    },
    {
      "auxiliary_loss_clip": 0.01178008,
      "auxiliary_loss_mlp": 0.01028147,
      "balance_loss_clip": 1.05216551,
      "balance_loss_mlp": 1.0193789,
      "epoch": 0.6772079600793603,
      "flos": 18141567903360.0,
      "grad_norm": 2.574310581368488,
      "language_loss": 0.84230256,
      "learning_rate": 9.96853477136929e-07,
      "loss": 0.86436409,
      "num_input_tokens_seen": 121162755,
      "step": 5632,
      "time_per_iteration": 2.552311897277832
    },
    {
      "auxiliary_loss_clip": 0.01164141,
      "auxiliary_loss_mlp": 0.01024506,
      "balance_loss_clip": 1.05123067,
      "balance_loss_mlp": 1.01679015,
      "epoch": 0.6773282029699994,
      "flos": 22452069571200.0,
      "grad_norm": 2.2647530540212903,
      "language_loss": 0.75411713,
      "learning_rate": 9.96179653050422e-07,
      "loss": 0.77600366,
      "num_input_tokens_seen": 121182915,
      "step": 5633,
      "time_per_iteration": 3.5545060634613037
    },
    {
      "auxiliary_loss_clip": 0.01179006,
      "auxiliary_loss_mlp": 0.01026186,
      "balance_loss_clip": 1.05240965,
      "balance_loss_mlp": 1.01807976,
      "epoch": 0.6774484458606385,
      "flos": 18693730748160.0,
      "grad_norm": 2.6403170937894354,
      "language_loss": 0.74357748,
      "learning_rate": 9.955059812426635e-07,
      "loss": 0.76562947,
      "num_input_tokens_seen": 121200445,
      "step": 5634,
      "time_per_iteration": 2.6255252361297607
    },
    {
      "auxiliary_loss_clip": 0.01178439,
      "auxiliary_loss_mlp": 0.01026422,
      "balance_loss_clip": 1.05528605,
      "balance_loss_mlp": 1.01872706,
      "epoch": 0.6775686887512776,
      "flos": 25994046821760.0,
      "grad_norm": 2.5602570445897443,
      "language_loss": 0.83191609,
      "learning_rate": 9.948324618158493e-07,
      "loss": 0.85396469,
      "num_input_tokens_seen": 121220785,
      "step": 5635,
      "time_per_iteration": 3.590602159500122
    },
    {
      "auxiliary_loss_clip": 0.0113471,
      "auxiliary_loss_mlp": 0.01026943,
      "balance_loss_clip": 1.0468204,
      "balance_loss_mlp": 1.01897955,
      "epoch": 0.6776889316419167,
      "flos": 13587987922560.0,
      "grad_norm": 2.696450732012804,
      "language_loss": 0.77676034,
      "learning_rate": 9.941590948721502e-07,
      "loss": 0.7983768,
      "num_input_tokens_seen": 121237985,
      "step": 5636,
      "time_per_iteration": 2.7254066467285156
    },
    {
      "auxiliary_loss_clip": 0.01161064,
      "auxiliary_loss_mlp": 0.01025464,
      "balance_loss_clip": 1.05110681,
      "balance_loss_mlp": 1.01799583,
      "epoch": 0.6778091745325557,
      "flos": 27601121220480.0,
      "grad_norm": 2.216224860025873,
      "language_loss": 0.76403689,
      "learning_rate": 9.934858805137188e-07,
      "loss": 0.78590214,
      "num_input_tokens_seen": 121258635,
      "step": 5637,
      "time_per_iteration": 2.6270596981048584
    },
    {
      "auxiliary_loss_clip": 0.01152478,
      "auxiliary_loss_mlp": 0.01029154,
      "balance_loss_clip": 1.04819357,
      "balance_loss_mlp": 1.02148342,
      "epoch": 0.6779294174231949,
      "flos": 18734058743040.0,
      "grad_norm": 1.7345841572297764,
      "language_loss": 0.80797315,
      "learning_rate": 9.92812818842677e-07,
      "loss": 0.8297894,
      "num_input_tokens_seen": 121277810,
      "step": 5638,
      "time_per_iteration": 2.6482205390930176
    },
    {
      "auxiliary_loss_clip": 0.01114204,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.04040897,
      "balance_loss_mlp": 1.02066803,
      "epoch": 0.678049660313834,
      "flos": 45873797765760.0,
      "grad_norm": 1.894999732967387,
      "language_loss": 0.64210641,
      "learning_rate": 9.921399099611306e-07,
      "loss": 0.66354263,
      "num_input_tokens_seen": 121298975,
      "step": 5639,
      "time_per_iteration": 2.907977342605591
    },
    {
      "auxiliary_loss_clip": 0.01160533,
      "auxiliary_loss_mlp": 0.01027309,
      "balance_loss_clip": 1.05418193,
      "balance_loss_mlp": 1.01989961,
      "epoch": 0.678169903204473,
      "flos": 19974556892160.0,
      "grad_norm": 1.8740704552551863,
      "language_loss": 0.6906414,
      "learning_rate": 9.914671539711588e-07,
      "loss": 0.71251982,
      "num_input_tokens_seen": 121318495,
      "step": 5640,
      "time_per_iteration": 2.6659929752349854
    },
    {
      "auxiliary_loss_clip": 0.01179533,
      "auxiliary_loss_mlp": 0.0102822,
      "balance_loss_clip": 1.0531671,
      "balance_loss_mlp": 1.01950562,
      "epoch": 0.6782901460951122,
      "flos": 21395613732480.0,
      "grad_norm": 2.1801899757341405,
      "language_loss": 0.78408074,
      "learning_rate": 9.90794550974817e-07,
      "loss": 0.8061583,
      "num_input_tokens_seen": 121338890,
      "step": 5641,
      "time_per_iteration": 2.5999724864959717
    },
    {
      "auxiliary_loss_clip": 0.01113998,
      "auxiliary_loss_mlp": 0.01032659,
      "balance_loss_clip": 1.04677033,
      "balance_loss_mlp": 1.02476406,
      "epoch": 0.6784103889857512,
      "flos": 21434002392960.0,
      "grad_norm": 2.735374732631969,
      "language_loss": 0.82270104,
      "learning_rate": 9.901221010741407e-07,
      "loss": 0.84416759,
      "num_input_tokens_seen": 121358210,
      "step": 5642,
      "time_per_iteration": 2.9272406101226807
    },
    {
      "auxiliary_loss_clip": 0.01165196,
      "auxiliary_loss_mlp": 0.01030893,
      "balance_loss_clip": 1.05135703,
      "balance_loss_mlp": 1.02272737,
      "epoch": 0.6785306318763903,
      "flos": 32671923091200.0,
      "grad_norm": 2.1042288164046714,
      "language_loss": 0.74588031,
      "learning_rate": 9.894498043711375e-07,
      "loss": 0.76784122,
      "num_input_tokens_seen": 121379955,
      "step": 5643,
      "time_per_iteration": 2.8204288482666016
    },
    {
      "auxiliary_loss_clip": 0.01150443,
      "auxiliary_loss_mlp": 0.01023995,
      "balance_loss_clip": 1.04993391,
      "balance_loss_mlp": 1.01687801,
      "epoch": 0.6786508747670293,
      "flos": 25632139340160.0,
      "grad_norm": 2.265577347433709,
      "language_loss": 0.69243443,
      "learning_rate": 9.887776609677962e-07,
      "loss": 0.7141788,
      "num_input_tokens_seen": 121401325,
      "step": 5644,
      "time_per_iteration": 2.721564531326294
    },
    {
      "auxiliary_loss_clip": 0.01174242,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.05060673,
      "balance_loss_mlp": 1.01906419,
      "epoch": 0.6787711176576685,
      "flos": 19171881619200.0,
      "grad_norm": 1.606342038315472,
      "language_loss": 0.72606629,
      "learning_rate": 9.88105670966079e-07,
      "loss": 0.74808526,
      "num_input_tokens_seen": 121419785,
      "step": 5645,
      "time_per_iteration": 3.6857073307037354
    },
    {
      "auxiliary_loss_clip": 0.01177233,
      "auxiliary_loss_mlp": 0.01032038,
      "balance_loss_clip": 1.05387139,
      "balance_loss_mlp": 1.0240804,
      "epoch": 0.6788913605483076,
      "flos": 13985159581440.0,
      "grad_norm": 2.9129310964112687,
      "language_loss": 0.79167807,
      "learning_rate": 9.874338344679283e-07,
      "loss": 0.81377077,
      "num_input_tokens_seen": 121435630,
      "step": 5646,
      "time_per_iteration": 2.541160821914673
    },
    {
      "auxiliary_loss_clip": 0.01175181,
      "auxiliary_loss_mlp": 0.01028799,
      "balance_loss_clip": 1.05304396,
      "balance_loss_mlp": 1.02153051,
      "epoch": 0.6790116034389466,
      "flos": 22017586659840.0,
      "grad_norm": 1.9831426010556956,
      "language_loss": 0.74186474,
      "learning_rate": 9.86762151575259e-07,
      "loss": 0.76390457,
      "num_input_tokens_seen": 121455625,
      "step": 5647,
      "time_per_iteration": 2.6771631240844727
    },
    {
      "auxiliary_loss_clip": 0.01158786,
      "auxiliary_loss_mlp": 0.01028583,
      "balance_loss_clip": 1.05153131,
      "balance_loss_mlp": 1.02147222,
      "epoch": 0.6791318463295858,
      "flos": 20922454851840.0,
      "grad_norm": 1.5320912559486495,
      "language_loss": 0.80326706,
      "learning_rate": 9.860906223899651e-07,
      "loss": 0.82514071,
      "num_input_tokens_seen": 121475020,
      "step": 5648,
      "time_per_iteration": 2.644073009490967
    },
    {
      "auxiliary_loss_clip": 0.01180679,
      "auxiliary_loss_mlp": 0.01031499,
      "balance_loss_clip": 1.05558133,
      "balance_loss_mlp": 1.02378631,
      "epoch": 0.6792520892202248,
      "flos": 28512749422080.0,
      "grad_norm": 3.1221858125857924,
      "language_loss": 0.7603004,
      "learning_rate": 9.854192470139184e-07,
      "loss": 0.78242218,
      "num_input_tokens_seen": 121496500,
      "step": 5649,
      "time_per_iteration": 2.667431592941284
    },
    {
      "auxiliary_loss_clip": 0.01178105,
      "auxiliary_loss_mlp": 0.01033506,
      "balance_loss_clip": 1.05337131,
      "balance_loss_mlp": 1.02575767,
      "epoch": 0.6793723321108639,
      "flos": 20011904058240.0,
      "grad_norm": 11.513989292269356,
      "language_loss": 0.71909654,
      "learning_rate": 9.847480255489645e-07,
      "loss": 0.74121267,
      "num_input_tokens_seen": 121515525,
      "step": 5650,
      "time_per_iteration": 2.621931552886963
    },
    {
      "auxiliary_loss_clip": 0.01175894,
      "auxiliary_loss_mlp": 0.01028938,
      "balance_loss_clip": 1.05181313,
      "balance_loss_mlp": 1.02141547,
      "epoch": 0.6794925750015031,
      "flos": 26649488246400.0,
      "grad_norm": 1.936993225818737,
      "language_loss": 0.69184119,
      "learning_rate": 9.840769580969295e-07,
      "loss": 0.71388948,
      "num_input_tokens_seen": 121535965,
      "step": 5651,
      "time_per_iteration": 2.6164426803588867
    },
    {
      "auxiliary_loss_clip": 0.01160689,
      "auxiliary_loss_mlp": 0.0102442,
      "balance_loss_clip": 1.05265808,
      "balance_loss_mlp": 1.01651037,
      "epoch": 0.6796128178921421,
      "flos": 21580374314880.0,
      "grad_norm": 1.8544478086691487,
      "language_loss": 0.80663478,
      "learning_rate": 9.834060447596114e-07,
      "loss": 0.82848585,
      "num_input_tokens_seen": 121555235,
      "step": 5652,
      "time_per_iteration": 2.6624722480773926
    },
    {
      "auxiliary_loss_clip": 0.01177683,
      "auxiliary_loss_mlp": 0.01024329,
      "balance_loss_clip": 1.05211878,
      "balance_loss_mlp": 1.016783,
      "epoch": 0.6797330607827812,
      "flos": 22492002516480.0,
      "grad_norm": 2.4814102488913994,
      "language_loss": 0.78443509,
      "learning_rate": 9.827352856387868e-07,
      "loss": 0.80645514,
      "num_input_tokens_seen": 121574945,
      "step": 5653,
      "time_per_iteration": 2.532588243484497
    },
    {
      "auxiliary_loss_clip": 0.01041578,
      "auxiliary_loss_mlp": 0.01001469,
      "balance_loss_clip": 1.01032686,
      "balance_loss_mlp": 1.00060451,
      "epoch": 0.6798533036734203,
      "flos": 66306648286080.0,
      "grad_norm": 0.7757598071415711,
      "language_loss": 0.64161962,
      "learning_rate": 9.820646808362118e-07,
      "loss": 0.66205007,
      "num_input_tokens_seen": 121641200,
      "step": 5654,
      "time_per_iteration": 3.318718671798706
    },
    {
      "auxiliary_loss_clip": 0.01175627,
      "auxiliary_loss_mlp": 0.0102802,
      "balance_loss_clip": 1.05338204,
      "balance_loss_mlp": 1.02067733,
      "epoch": 0.6799735465640594,
      "flos": 16180163792640.0,
      "grad_norm": 2.160265195927399,
      "language_loss": 0.7286498,
      "learning_rate": 9.813942304536154e-07,
      "loss": 0.75068629,
      "num_input_tokens_seen": 121659170,
      "step": 5655,
      "time_per_iteration": 2.638011932373047
    },
    {
      "auxiliary_loss_clip": 0.01145746,
      "auxiliary_loss_mlp": 0.01028208,
      "balance_loss_clip": 1.04924393,
      "balance_loss_mlp": 1.0202384,
      "epoch": 0.6800937894546984,
      "flos": 22125749489280.0,
      "grad_norm": 1.739981038499057,
      "language_loss": 0.63819778,
      "learning_rate": 9.807239345927043e-07,
      "loss": 0.65993732,
      "num_input_tokens_seen": 121679180,
      "step": 5656,
      "time_per_iteration": 2.693359613418579
    },
    {
      "auxiliary_loss_clip": 0.0114245,
      "auxiliary_loss_mlp": 0.01031045,
      "balance_loss_clip": 1.0476048,
      "balance_loss_mlp": 1.02249789,
      "epoch": 0.6802140323453376,
      "flos": 31612953300480.0,
      "grad_norm": 2.800603314064948,
      "language_loss": 0.72379267,
      "learning_rate": 9.80053793355162e-07,
      "loss": 0.74552763,
      "num_input_tokens_seen": 121697875,
      "step": 5657,
      "time_per_iteration": 2.7424657344818115
    },
    {
      "auxiliary_loss_clip": 0.01177672,
      "auxiliary_loss_mlp": 0.0103158,
      "balance_loss_clip": 1.05228782,
      "balance_loss_mlp": 1.02368259,
      "epoch": 0.6803342752359767,
      "flos": 17712938908800.0,
      "grad_norm": 2.135868490564984,
      "language_loss": 0.7552616,
      "learning_rate": 9.793838068426472e-07,
      "loss": 0.77735406,
      "num_input_tokens_seen": 121715570,
      "step": 5658,
      "time_per_iteration": 2.64363169670105
    },
    {
      "auxiliary_loss_clip": 0.01161323,
      "auxiliary_loss_mlp": 0.01032138,
      "balance_loss_clip": 1.05301213,
      "balance_loss_mlp": 1.02396965,
      "epoch": 0.6804545181266157,
      "flos": 11326800902400.0,
      "grad_norm": 2.6639454281201242,
      "language_loss": 0.61359513,
      "learning_rate": 9.78713975156799e-07,
      "loss": 0.63552982,
      "num_input_tokens_seen": 121731435,
      "step": 5659,
      "time_per_iteration": 3.5963690280914307
    },
    {
      "auxiliary_loss_clip": 0.01179873,
      "auxiliary_loss_mlp": 0.01031459,
      "balance_loss_clip": 1.05397606,
      "balance_loss_mlp": 1.02324557,
      "epoch": 0.6805747610172549,
      "flos": 29350976181120.0,
      "grad_norm": 1.8304879166703887,
      "language_loss": 0.72018135,
      "learning_rate": 9.780442983992273e-07,
      "loss": 0.74229461,
      "num_input_tokens_seen": 121749950,
      "step": 5660,
      "time_per_iteration": 2.6324996948242188
    },
    {
      "auxiliary_loss_clip": 0.01149236,
      "auxiliary_loss_mlp": 0.01029481,
      "balance_loss_clip": 1.05138671,
      "balance_loss_mlp": 1.02189887,
      "epoch": 0.680695003907894,
      "flos": 37631868612480.0,
      "grad_norm": 1.9899232048813005,
      "language_loss": 0.716851,
      "learning_rate": 9.773747766715238e-07,
      "loss": 0.73863816,
      "num_input_tokens_seen": 121770770,
      "step": 5661,
      "time_per_iteration": 3.826775550842285
    },
    {
      "auxiliary_loss_clip": 0.01175534,
      "auxiliary_loss_mlp": 0.01032532,
      "balance_loss_clip": 1.05121875,
      "balance_loss_mlp": 1.02400887,
      "epoch": 0.680815246798533,
      "flos": 22127365601280.0,
      "grad_norm": 4.220025681815003,
      "language_loss": 0.80466157,
      "learning_rate": 9.767054100752536e-07,
      "loss": 0.82674229,
      "num_input_tokens_seen": 121790720,
      "step": 5662,
      "time_per_iteration": 2.5866825580596924
    },
    {
      "auxiliary_loss_clip": 0.01148567,
      "auxiliary_loss_mlp": 0.01034085,
      "balance_loss_clip": 1.04952312,
      "balance_loss_mlp": 1.02581239,
      "epoch": 0.6809354896891722,
      "flos": 17201822330880.0,
      "grad_norm": 4.615911042735087,
      "language_loss": 0.81836843,
      "learning_rate": 9.760361987119584e-07,
      "loss": 0.84019494,
      "num_input_tokens_seen": 121808455,
      "step": 5663,
      "time_per_iteration": 2.6630194187164307
    },
    {
      "auxiliary_loss_clip": 0.01161262,
      "auxiliary_loss_mlp": 0.01027809,
      "balance_loss_clip": 1.05248547,
      "balance_loss_mlp": 1.01895213,
      "epoch": 0.6810557325798112,
      "flos": 12458166554880.0,
      "grad_norm": 2.1240915317389657,
      "language_loss": 0.67709184,
      "learning_rate": 9.753671426831592e-07,
      "loss": 0.69898254,
      "num_input_tokens_seen": 121824470,
      "step": 5664,
      "time_per_iteration": 2.6243200302124023
    },
    {
      "auxiliary_loss_clip": 0.0113341,
      "auxiliary_loss_mlp": 0.01027713,
      "balance_loss_clip": 1.05089116,
      "balance_loss_mlp": 1.02011681,
      "epoch": 0.6811759754704503,
      "flos": 22156165330560.0,
      "grad_norm": 1.9345513387613031,
      "language_loss": 0.79736161,
      "learning_rate": 9.746982420903483e-07,
      "loss": 0.81897283,
      "num_input_tokens_seen": 121842665,
      "step": 5665,
      "time_per_iteration": 2.67830491065979
    },
    {
      "auxiliary_loss_clip": 0.01174369,
      "auxiliary_loss_mlp": 0.01020543,
      "balance_loss_clip": 1.05235946,
      "balance_loss_mlp": 1.01329172,
      "epoch": 0.6812962183610894,
      "flos": 17525377065600.0,
      "grad_norm": 2.549301132563279,
      "language_loss": 0.7479046,
      "learning_rate": 9.740294970349993e-07,
      "loss": 0.76985371,
      "num_input_tokens_seen": 121859080,
      "step": 5666,
      "time_per_iteration": 2.5739076137542725
    },
    {
      "auxiliary_loss_clip": 0.01067308,
      "auxiliary_loss_mlp": 0.01003585,
      "balance_loss_clip": 1.01087189,
      "balance_loss_mlp": 1.00270927,
      "epoch": 0.6814164612517285,
      "flos": 60274480855680.0,
      "grad_norm": 0.8895486153208986,
      "language_loss": 0.608648,
      "learning_rate": 9.733609076185594e-07,
      "loss": 0.62935698,
      "num_input_tokens_seen": 121915485,
      "step": 5667,
      "time_per_iteration": 3.0719010829925537
    },
    {
      "auxiliary_loss_clip": 0.01150658,
      "auxiliary_loss_mlp": 0.01027725,
      "balance_loss_clip": 1.05023217,
      "balance_loss_mlp": 1.0195055,
      "epoch": 0.6815367041423676,
      "flos": 19317750750720.0,
      "grad_norm": 2.106791642131067,
      "language_loss": 0.84059173,
      "learning_rate": 9.72692473942455e-07,
      "loss": 0.86237556,
      "num_input_tokens_seen": 121932710,
      "step": 5668,
      "time_per_iteration": 2.706054449081421
    },
    {
      "auxiliary_loss_clip": 0.01179398,
      "auxiliary_loss_mlp": 0.01030906,
      "balance_loss_clip": 1.05166984,
      "balance_loss_mlp": 1.02193522,
      "epoch": 0.6816569470330067,
      "flos": 22161696024960.0,
      "grad_norm": 1.9401737652349365,
      "language_loss": 0.77746201,
      "learning_rate": 9.720241961080849e-07,
      "loss": 0.79956508,
      "num_input_tokens_seen": 121952025,
      "step": 5669,
      "time_per_iteration": 2.5960404872894287
    },
    {
      "auxiliary_loss_clip": 0.01174646,
      "auxiliary_loss_mlp": 0.00761734,
      "balance_loss_clip": 1.04990065,
      "balance_loss_mlp": 1.00073338,
      "epoch": 0.6817771899236458,
      "flos": 41463501137280.0,
      "grad_norm": 2.0735968391922275,
      "language_loss": 0.73458958,
      "learning_rate": 9.713560742168259e-07,
      "loss": 0.7539534,
      "num_input_tokens_seen": 121974650,
      "step": 5670,
      "time_per_iteration": 3.702220916748047
    },
    {
      "auxiliary_loss_clip": 0.01176213,
      "auxiliary_loss_mlp": 0.01030579,
      "balance_loss_clip": 1.0543983,
      "balance_loss_mlp": 1.02237117,
      "epoch": 0.6818974328142848,
      "flos": 21106138026240.0,
      "grad_norm": 2.3106220505381874,
      "language_loss": 0.71338087,
      "learning_rate": 9.706881083700333e-07,
      "loss": 0.73544878,
      "num_input_tokens_seen": 121994335,
      "step": 5671,
      "time_per_iteration": 2.652095079421997
    },
    {
      "auxiliary_loss_clip": 0.01165184,
      "auxiliary_loss_mlp": 0.0102781,
      "balance_loss_clip": 1.05199873,
      "balance_loss_mlp": 1.01935804,
      "epoch": 0.682017675704924,
      "flos": 20441897769600.0,
      "grad_norm": 1.8856592579874791,
      "language_loss": 0.83235341,
      "learning_rate": 9.700202986690357e-07,
      "loss": 0.85428333,
      "num_input_tokens_seen": 122012635,
      "step": 5672,
      "time_per_iteration": 2.643869400024414
    },
    {
      "auxiliary_loss_clip": 0.01176522,
      "auxiliary_loss_mlp": 0.01027323,
      "balance_loss_clip": 1.05241227,
      "balance_loss_mlp": 1.0192529,
      "epoch": 0.682137918595563,
      "flos": 20044438801920.0,
      "grad_norm": 2.214755946975099,
      "language_loss": 0.66650242,
      "learning_rate": 9.693526452151413e-07,
      "loss": 0.68854082,
      "num_input_tokens_seen": 122031685,
      "step": 5673,
      "time_per_iteration": 2.679497003555298
    },
    {
      "auxiliary_loss_clip": 0.01178651,
      "auxiliary_loss_mlp": 0.01027923,
      "balance_loss_clip": 1.05283344,
      "balance_loss_mlp": 1.01942027,
      "epoch": 0.6822581614862021,
      "flos": 31684559063040.0,
      "grad_norm": 1.6885843861936527,
      "language_loss": 0.75281203,
      "learning_rate": 9.686851481096305e-07,
      "loss": 0.77487779,
      "num_input_tokens_seen": 122052995,
      "step": 5674,
      "time_per_iteration": 2.639092206954956
    },
    {
      "auxiliary_loss_clip": 0.01176047,
      "auxiliary_loss_mlp": 0.01028806,
      "balance_loss_clip": 1.05194962,
      "balance_loss_mlp": 1.02066743,
      "epoch": 0.6823784043768413,
      "flos": 23477570864640.0,
      "grad_norm": 4.255567060051652,
      "language_loss": 0.71757352,
      "learning_rate": 9.68017807453762e-07,
      "loss": 0.739622,
      "num_input_tokens_seen": 122071740,
      "step": 5675,
      "time_per_iteration": 2.670535087585449
    },
    {
      "auxiliary_loss_clip": 0.01159645,
      "auxiliary_loss_mlp": 0.01023408,
      "balance_loss_clip": 1.05172849,
      "balance_loss_mlp": 1.01588559,
      "epoch": 0.6824986472674803,
      "flos": 14137134024960.0,
      "grad_norm": 1.8886829611098086,
      "language_loss": 0.72908884,
      "learning_rate": 9.673506233487721e-07,
      "loss": 0.75091934,
      "num_input_tokens_seen": 122089705,
      "step": 5676,
      "time_per_iteration": 2.629258394241333
    },
    {
      "auxiliary_loss_clip": 0.01171306,
      "auxiliary_loss_mlp": 0.01023659,
      "balance_loss_clip": 1.05072701,
      "balance_loss_mlp": 1.0165844,
      "epoch": 0.6826188901581194,
      "flos": 21504997624320.0,
      "grad_norm": 1.7885615918442486,
      "language_loss": 0.86067569,
      "learning_rate": 9.666835958958717e-07,
      "loss": 0.88262534,
      "num_input_tokens_seen": 122109025,
      "step": 5677,
      "time_per_iteration": 2.703155994415283
    },
    {
      "auxiliary_loss_clip": 0.01177449,
      "auxiliary_loss_mlp": 0.01028964,
      "balance_loss_clip": 1.05471539,
      "balance_loss_mlp": 1.02173746,
      "epoch": 0.6827391330487584,
      "flos": 20810126044800.0,
      "grad_norm": 2.1809899098491674,
      "language_loss": 0.80608213,
      "learning_rate": 9.660167251962484e-07,
      "loss": 0.82814622,
      "num_input_tokens_seen": 122127385,
      "step": 5678,
      "time_per_iteration": 2.6711580753326416
    },
    {
      "auxiliary_loss_clip": 0.01173764,
      "auxiliary_loss_mlp": 0.01029096,
      "balance_loss_clip": 1.0519917,
      "balance_loss_mlp": 1.02173519,
      "epoch": 0.6828593759393976,
      "flos": 21688788539520.0,
      "grad_norm": 1.5666189383143634,
      "language_loss": 0.78137302,
      "learning_rate": 9.653500113510654e-07,
      "loss": 0.80340165,
      "num_input_tokens_seen": 122146500,
      "step": 5679,
      "time_per_iteration": 2.6596415042877197
    },
    {
      "auxiliary_loss_clip": 0.01174731,
      "auxiliary_loss_mlp": 0.01026149,
      "balance_loss_clip": 1.05102968,
      "balance_loss_mlp": 1.01800728,
      "epoch": 0.6829796188300367,
      "flos": 25337707557120.0,
      "grad_norm": 3.161403791893772,
      "language_loss": 0.66968298,
      "learning_rate": 9.646834544614627e-07,
      "loss": 0.69169182,
      "num_input_tokens_seen": 122167000,
      "step": 5680,
      "time_per_iteration": 2.616642475128174
    },
    {
      "auxiliary_loss_clip": 0.01178191,
      "auxiliary_loss_mlp": 0.01023336,
      "balance_loss_clip": 1.05465841,
      "balance_loss_mlp": 1.01620722,
      "epoch": 0.6830998617206757,
      "flos": 20704800389760.0,
      "grad_norm": 1.8907854077194532,
      "language_loss": 0.76282728,
      "learning_rate": 9.64017054628558e-07,
      "loss": 0.78484261,
      "num_input_tokens_seen": 122185825,
      "step": 5681,
      "time_per_iteration": 2.6276190280914307
    },
    {
      "auxiliary_loss_clip": 0.01164679,
      "auxiliary_loss_mlp": 0.01024351,
      "balance_loss_clip": 1.05144548,
      "balance_loss_mlp": 1.01621473,
      "epoch": 0.6832201046113149,
      "flos": 21726638496000.0,
      "grad_norm": 1.921615752550204,
      "language_loss": 0.78937823,
      "learning_rate": 9.63350811953441e-07,
      "loss": 0.81126857,
      "num_input_tokens_seen": 122206200,
      "step": 5682,
      "time_per_iteration": 2.630903959274292
    },
    {
      "auxiliary_loss_clip": 0.01162088,
      "auxiliary_loss_mlp": 0.0102608,
      "balance_loss_clip": 1.05109179,
      "balance_loss_mlp": 1.01889181,
      "epoch": 0.6833403475019539,
      "flos": 19536554448000.0,
      "grad_norm": 1.9615204428151654,
      "language_loss": 0.70458877,
      "learning_rate": 9.626847265371826e-07,
      "loss": 0.72647047,
      "num_input_tokens_seen": 122225520,
      "step": 5683,
      "time_per_iteration": 2.6716480255126953
    },
    {
      "auxiliary_loss_clip": 0.01145688,
      "auxiliary_loss_mlp": 0.01026146,
      "balance_loss_clip": 1.04574347,
      "balance_loss_mlp": 1.01837993,
      "epoch": 0.683460590392593,
      "flos": 19352153001600.0,
      "grad_norm": 2.3061646135889413,
      "language_loss": 0.79180765,
      "learning_rate": 9.620187984808262e-07,
      "loss": 0.81352597,
      "num_input_tokens_seen": 122244320,
      "step": 5684,
      "time_per_iteration": 2.673736333847046
    },
    {
      "auxiliary_loss_clip": 0.011658,
      "auxiliary_loss_mlp": 0.01023054,
      "balance_loss_clip": 1.04974115,
      "balance_loss_mlp": 1.01581252,
      "epoch": 0.6835808332832322,
      "flos": 23288500650240.0,
      "grad_norm": 2.9947366652209206,
      "language_loss": 0.86170244,
      "learning_rate": 9.613530278853919e-07,
      "loss": 0.88359094,
      "num_input_tokens_seen": 122264295,
      "step": 5685,
      "time_per_iteration": 3.7056264877319336
    },
    {
      "auxiliary_loss_clip": 0.01176836,
      "auxiliary_loss_mlp": 0.01025648,
      "balance_loss_clip": 1.05243349,
      "balance_loss_mlp": 1.01745808,
      "epoch": 0.6837010761738712,
      "flos": 21653416621440.0,
      "grad_norm": 2.488936421079156,
      "language_loss": 0.74655312,
      "learning_rate": 9.60687414851879e-07,
      "loss": 0.76857799,
      "num_input_tokens_seen": 122285300,
      "step": 5686,
      "time_per_iteration": 2.627553701400757
    },
    {
      "auxiliary_loss_clip": 0.01179304,
      "auxiliary_loss_mlp": 0.01026836,
      "balance_loss_clip": 1.0534029,
      "balance_loss_mlp": 1.0183599,
      "epoch": 0.6838213190645103,
      "flos": 17566387418880.0,
      "grad_norm": 3.1631711634451647,
      "language_loss": 0.77070856,
      "learning_rate": 9.600219594812575e-07,
      "loss": 0.79276997,
      "num_input_tokens_seen": 122303240,
      "step": 5687,
      "time_per_iteration": 3.498765707015991
    },
    {
      "auxiliary_loss_clip": 0.01159681,
      "auxiliary_loss_mlp": 0.01025716,
      "balance_loss_clip": 1.0512228,
      "balance_loss_mlp": 1.01853108,
      "epoch": 0.6839415619551494,
      "flos": 23112538899840.0,
      "grad_norm": 1.6997145155308555,
      "language_loss": 0.7294969,
      "learning_rate": 9.593566618744786e-07,
      "loss": 0.75135082,
      "num_input_tokens_seen": 122323390,
      "step": 5688,
      "time_per_iteration": 3.567188024520874
    },
    {
      "auxiliary_loss_clip": 0.0116092,
      "auxiliary_loss_mlp": 0.01026966,
      "balance_loss_clip": 1.04829788,
      "balance_loss_mlp": 1.01908374,
      "epoch": 0.6840618048457885,
      "flos": 22127868391680.0,
      "grad_norm": 3.759982465343626,
      "language_loss": 0.74166727,
      "learning_rate": 9.58691522132466e-07,
      "loss": 0.76354611,
      "num_input_tokens_seen": 122342200,
      "step": 5689,
      "time_per_iteration": 2.6700398921966553
    },
    {
      "auxiliary_loss_clip": 0.01181011,
      "auxiliary_loss_mlp": 0.01026418,
      "balance_loss_clip": 1.05525327,
      "balance_loss_mlp": 1.01852655,
      "epoch": 0.6841820477364275,
      "flos": 22015898720640.0,
      "grad_norm": 2.3566843812827614,
      "language_loss": 0.8522836,
      "learning_rate": 9.58026540356123e-07,
      "loss": 0.87435782,
      "num_input_tokens_seen": 122360465,
      "step": 5690,
      "time_per_iteration": 2.6587066650390625
    },
    {
      "auxiliary_loss_clip": 0.01180679,
      "auxiliary_loss_mlp": 0.01030776,
      "balance_loss_clip": 1.05517244,
      "balance_loss_mlp": 1.02230632,
      "epoch": 0.6843022906270667,
      "flos": 24900531125760.0,
      "grad_norm": 1.8860485359825139,
      "language_loss": 0.87103361,
      "learning_rate": 9.573617166463246e-07,
      "loss": 0.89314818,
      "num_input_tokens_seen": 122381680,
      "step": 5691,
      "time_per_iteration": 2.6218650341033936
    },
    {
      "auxiliary_loss_clip": 0.01164792,
      "auxiliary_loss_mlp": 0.01027475,
      "balance_loss_clip": 1.04934645,
      "balance_loss_mlp": 1.01951206,
      "epoch": 0.6844225335177058,
      "flos": 19969924037760.0,
      "grad_norm": 2.170544393864064,
      "language_loss": 0.60438544,
      "learning_rate": 9.56697051103924e-07,
      "loss": 0.62630814,
      "num_input_tokens_seen": 122399120,
      "step": 5692,
      "time_per_iteration": 2.615279197692871
    },
    {
      "auxiliary_loss_clip": 0.01142491,
      "auxiliary_loss_mlp": 0.0102824,
      "balance_loss_clip": 1.05032706,
      "balance_loss_mlp": 1.02109885,
      "epoch": 0.6845427764083448,
      "flos": 25883334126720.0,
      "grad_norm": 1.9124893224314088,
      "language_loss": 0.81408954,
      "learning_rate": 9.560325438297522e-07,
      "loss": 0.83579683,
      "num_input_tokens_seen": 122417430,
      "step": 5693,
      "time_per_iteration": 2.7259128093719482
    },
    {
      "auxiliary_loss_clip": 0.01178453,
      "auxiliary_loss_mlp": 0.01031429,
      "balance_loss_clip": 1.05446947,
      "balance_loss_mlp": 1.02334046,
      "epoch": 0.684663019298984,
      "flos": 18880143356160.0,
      "grad_norm": 1.8476961949778452,
      "language_loss": 0.87037998,
      "learning_rate": 9.553681949246127e-07,
      "loss": 0.89247882,
      "num_input_tokens_seen": 122435055,
      "step": 5694,
      "time_per_iteration": 2.5456581115722656
    },
    {
      "auxiliary_loss_clip": 0.01159993,
      "auxiliary_loss_mlp": 0.01034031,
      "balance_loss_clip": 1.04965973,
      "balance_loss_mlp": 1.02541208,
      "epoch": 0.684783262189623,
      "flos": 54193725302400.0,
      "grad_norm": 1.998683562344872,
      "language_loss": 0.75714785,
      "learning_rate": 9.547040044892886e-07,
      "loss": 0.77908808,
      "num_input_tokens_seen": 122462570,
      "step": 5695,
      "time_per_iteration": 2.970255136489868
    },
    {
      "auxiliary_loss_clip": 0.01051054,
      "auxiliary_loss_mlp": 0.01003119,
      "balance_loss_clip": 1.00895333,
      "balance_loss_mlp": 1.00221908,
      "epoch": 0.6849035050802621,
      "flos": 63970264143360.0,
      "grad_norm": 0.8649854446275148,
      "language_loss": 0.60000193,
      "learning_rate": 9.540399726245354e-07,
      "loss": 0.6205436,
      "num_input_tokens_seen": 122519275,
      "step": 5696,
      "time_per_iteration": 3.0805375576019287
    },
    {
      "auxiliary_loss_clip": 0.01122209,
      "auxiliary_loss_mlp": 0.01031595,
      "balance_loss_clip": 1.04244792,
      "balance_loss_mlp": 1.02347684,
      "epoch": 0.6850237479709013,
      "flos": 25224121774080.0,
      "grad_norm": 4.940745600151119,
      "language_loss": 0.68776155,
      "learning_rate": 9.533760994310859e-07,
      "loss": 0.70929956,
      "num_input_tokens_seen": 122539675,
      "step": 5697,
      "time_per_iteration": 3.713043451309204
    },
    {
      "auxiliary_loss_clip": 0.01165214,
      "auxiliary_loss_mlp": 0.01029637,
      "balance_loss_clip": 1.05206752,
      "balance_loss_mlp": 1.02190638,
      "epoch": 0.6851439908615403,
      "flos": 19354128249600.0,
      "grad_norm": 2.0283866279743568,
      "language_loss": 0.75716442,
      "learning_rate": 9.527123850096508e-07,
      "loss": 0.77911294,
      "num_input_tokens_seen": 122558035,
      "step": 5698,
      "time_per_iteration": 2.7014214992523193
    },
    {
      "auxiliary_loss_clip": 0.0117535,
      "auxiliary_loss_mlp": 0.01024808,
      "balance_loss_clip": 1.05236077,
      "balance_loss_mlp": 1.01714861,
      "epoch": 0.6852642337521794,
      "flos": 23182133500800.0,
      "grad_norm": 3.230541883957135,
      "language_loss": 0.71992397,
      "learning_rate": 9.520488294609142e-07,
      "loss": 0.7419256,
      "num_input_tokens_seen": 122576815,
      "step": 5699,
      "time_per_iteration": 2.5910236835479736
    },
    {
      "auxiliary_loss_clip": 0.01058212,
      "auxiliary_loss_mlp": 0.00751861,
      "balance_loss_clip": 1.0105679,
      "balance_loss_mlp": 1.00006604,
      "epoch": 0.6853844766428185,
      "flos": 62647206583680.0,
      "grad_norm": 0.7429574470283916,
      "language_loss": 0.5373221,
      "learning_rate": 9.513854328855368e-07,
      "loss": 0.55542278,
      "num_input_tokens_seen": 122634690,
      "step": 5700,
      "time_per_iteration": 3.159513473510742
    },
    {
      "auxiliary_loss_clip": 0.01172131,
      "auxiliary_loss_mlp": 0.01025168,
      "balance_loss_clip": 1.05075145,
      "balance_loss_mlp": 1.01702559,
      "epoch": 0.6855047195334576,
      "flos": 23437242869760.0,
      "grad_norm": 2.1614527010543045,
      "language_loss": 0.81366646,
      "learning_rate": 9.507221953841558e-07,
      "loss": 0.83563948,
      "num_input_tokens_seen": 122652320,
      "step": 5701,
      "time_per_iteration": 2.6316866874694824
    },
    {
      "auxiliary_loss_clip": 0.01179299,
      "auxiliary_loss_mlp": 0.01025699,
      "balance_loss_clip": 1.05371952,
      "balance_loss_mlp": 1.01790237,
      "epoch": 0.6856249624240967,
      "flos": 20664831530880.0,
      "grad_norm": 3.916321347019577,
      "language_loss": 0.78308791,
      "learning_rate": 9.500591170573824e-07,
      "loss": 0.80513787,
      "num_input_tokens_seen": 122672340,
      "step": 5702,
      "time_per_iteration": 2.631728172302246
    },
    {
      "auxiliary_loss_clip": 0.01165533,
      "auxiliary_loss_mlp": 0.01030869,
      "balance_loss_clip": 1.05142641,
      "balance_loss_mlp": 1.0222621,
      "epoch": 0.6857452053147358,
      "flos": 17087302794240.0,
      "grad_norm": 3.6099051730113367,
      "language_loss": 0.74340653,
      "learning_rate": 9.493961980058078e-07,
      "loss": 0.76537055,
      "num_input_tokens_seen": 122689935,
      "step": 5703,
      "time_per_iteration": 2.6750223636627197
    },
    {
      "auxiliary_loss_clip": 0.01174348,
      "auxiliary_loss_mlp": 0.01028365,
      "balance_loss_clip": 1.05317903,
      "balance_loss_mlp": 1.02137673,
      "epoch": 0.6858654482053749,
      "flos": 30847266057600.0,
      "grad_norm": 2.7794459800959883,
      "language_loss": 0.67802227,
      "learning_rate": 9.48733438329993e-07,
      "loss": 0.7000494,
      "num_input_tokens_seen": 122710200,
      "step": 5704,
      "time_per_iteration": 2.6916162967681885
    },
    {
      "auxiliary_loss_clip": 0.01162921,
      "auxiliary_loss_mlp": 0.01025148,
      "balance_loss_clip": 1.05090761,
      "balance_loss_mlp": 1.01814437,
      "epoch": 0.6859856910960139,
      "flos": 28877314510080.0,
      "grad_norm": 1.9882805960870802,
      "language_loss": 0.74283099,
      "learning_rate": 9.480708381304807e-07,
      "loss": 0.76471162,
      "num_input_tokens_seen": 122731495,
      "step": 5705,
      "time_per_iteration": 2.6999435424804688
    },
    {
      "auxiliary_loss_clip": 0.0117894,
      "auxiliary_loss_mlp": 0.00762743,
      "balance_loss_clip": 1.05403078,
      "balance_loss_mlp": 1.0007391,
      "epoch": 0.6861059339866531,
      "flos": 19354523299200.0,
      "grad_norm": 3.4751304526623636,
      "language_loss": 0.83586639,
      "learning_rate": 9.474083975077858e-07,
      "loss": 0.85528314,
      "num_input_tokens_seen": 122748620,
      "step": 5706,
      "time_per_iteration": 2.666440963745117
    },
    {
      "auxiliary_loss_clip": 0.01150596,
      "auxiliary_loss_mlp": 0.01029124,
      "balance_loss_clip": 1.05076504,
      "balance_loss_mlp": 1.021083,
      "epoch": 0.6862261768772921,
      "flos": 22199976944640.0,
      "grad_norm": 2.145274867033842,
      "language_loss": 0.80405509,
      "learning_rate": 9.467461165623994e-07,
      "loss": 0.82585233,
      "num_input_tokens_seen": 122767670,
      "step": 5707,
      "time_per_iteration": 2.7201364040374756
    },
    {
      "auxiliary_loss_clip": 0.01177074,
      "auxiliary_loss_mlp": 0.01026761,
      "balance_loss_clip": 1.05234921,
      "balance_loss_mlp": 1.01901817,
      "epoch": 0.6863464197679312,
      "flos": 26285677344000.0,
      "grad_norm": 1.896919757592681,
      "language_loss": 0.79979116,
      "learning_rate": 9.46083995394791e-07,
      "loss": 0.82182944,
      "num_input_tokens_seen": 122785480,
      "step": 5708,
      "time_per_iteration": 2.653613328933716
    },
    {
      "auxiliary_loss_clip": 0.01156237,
      "auxiliary_loss_mlp": 0.0102225,
      "balance_loss_clip": 1.05130863,
      "balance_loss_mlp": 1.01507354,
      "epoch": 0.6864666626585703,
      "flos": 37815228564480.0,
      "grad_norm": 12.30772936062164,
      "language_loss": 0.63273525,
      "learning_rate": 9.454220341054012e-07,
      "loss": 0.65452009,
      "num_input_tokens_seen": 122810265,
      "step": 5709,
      "time_per_iteration": 2.8259685039520264
    },
    {
      "auxiliary_loss_clip": 0.01176703,
      "auxiliary_loss_mlp": 0.01029955,
      "balance_loss_clip": 1.05288649,
      "balance_loss_mlp": 1.02175915,
      "epoch": 0.6865869055492094,
      "flos": 19391152193280.0,
      "grad_norm": 1.9974026042624595,
      "language_loss": 0.80788273,
      "learning_rate": 9.447602327946512e-07,
      "loss": 0.82994932,
      "num_input_tokens_seen": 122828905,
      "step": 5710,
      "time_per_iteration": 2.568415403366089
    },
    {
      "auxiliary_loss_clip": 0.01179578,
      "auxiliary_loss_mlp": 0.01028169,
      "balance_loss_clip": 1.05336165,
      "balance_loss_mlp": 1.01980662,
      "epoch": 0.6867071484398485,
      "flos": 20375966355840.0,
      "grad_norm": 1.882881725410225,
      "language_loss": 0.76650435,
      "learning_rate": 9.440985915629338e-07,
      "loss": 0.78858179,
      "num_input_tokens_seen": 122846235,
      "step": 5711,
      "time_per_iteration": 3.5494484901428223
    },
    {
      "auxiliary_loss_clip": 0.01178597,
      "auxiliary_loss_mlp": 0.00761746,
      "balance_loss_clip": 1.05636036,
      "balance_loss_mlp": 1.00065625,
      "epoch": 0.6868273913304875,
      "flos": 15889143801600.0,
      "grad_norm": 2.078079901571578,
      "language_loss": 0.73575103,
      "learning_rate": 9.434371105106223e-07,
      "loss": 0.75515449,
      "num_input_tokens_seen": 122863835,
      "step": 5712,
      "time_per_iteration": 2.6139888763427734
    },
    {
      "auxiliary_loss_clip": 0.01177474,
      "auxiliary_loss_mlp": 0.01028884,
      "balance_loss_clip": 1.05217147,
      "balance_loss_mlp": 1.02071166,
      "epoch": 0.6869476342211267,
      "flos": 24462492768000.0,
      "grad_norm": 2.1998383719553134,
      "language_loss": 0.71112537,
      "learning_rate": 9.427757897380602e-07,
      "loss": 0.73318899,
      "num_input_tokens_seen": 122883235,
      "step": 5713,
      "time_per_iteration": 2.6170501708984375
    },
    {
      "auxiliary_loss_clip": 0.01176052,
      "auxiliary_loss_mlp": 0.01029512,
      "balance_loss_clip": 1.05094624,
      "balance_loss_mlp": 1.0211612,
      "epoch": 0.6870678771117658,
      "flos": 18442571875200.0,
      "grad_norm": 2.0638023901330738,
      "language_loss": 0.85325038,
      "learning_rate": 9.421146293455695e-07,
      "loss": 0.87530601,
      "num_input_tokens_seen": 122898975,
      "step": 5714,
      "time_per_iteration": 4.387303352355957
    },
    {
      "auxiliary_loss_clip": 0.01146458,
      "auxiliary_loss_mlp": 0.01032027,
      "balance_loss_clip": 1.05061722,
      "balance_loss_mlp": 1.02442729,
      "epoch": 0.6871881200024048,
      "flos": 22200371994240.0,
      "grad_norm": 1.930554691690815,
      "language_loss": 0.68775976,
      "learning_rate": 9.414536294334489e-07,
      "loss": 0.7095446,
      "num_input_tokens_seen": 122918995,
      "step": 5715,
      "time_per_iteration": 2.6530778408050537
    },
    {
      "auxiliary_loss_clip": 0.01160826,
      "auxiliary_loss_mlp": 0.01021786,
      "balance_loss_clip": 1.04948473,
      "balance_loss_mlp": 1.01401985,
      "epoch": 0.687308362893044,
      "flos": 22127724737280.0,
      "grad_norm": 1.843473875282834,
      "language_loss": 0.69168442,
      "learning_rate": 9.407927901019708e-07,
      "loss": 0.71351051,
      "num_input_tokens_seen": 122938125,
      "step": 5716,
      "time_per_iteration": 2.68656063079834
    },
    {
      "auxiliary_loss_clip": 0.0112133,
      "auxiliary_loss_mlp": 0.01025368,
      "balance_loss_clip": 1.04489136,
      "balance_loss_mlp": 1.01773214,
      "epoch": 0.687428605783683,
      "flos": 25040546340480.0,
      "grad_norm": 1.87747913217492,
      "language_loss": 0.76793396,
      "learning_rate": 9.401321114513854e-07,
      "loss": 0.78940094,
      "num_input_tokens_seen": 122957020,
      "step": 5717,
      "time_per_iteration": 2.8428070545196533
    },
    {
      "auxiliary_loss_clip": 0.01152117,
      "auxiliary_loss_mlp": 0.01030395,
      "balance_loss_clip": 1.05178475,
      "balance_loss_mlp": 1.02212167,
      "epoch": 0.6875488486743221,
      "flos": 23770063313280.0,
      "grad_norm": 1.73193055473067,
      "language_loss": 0.75757265,
      "learning_rate": 9.394715935819155e-07,
      "loss": 0.77939773,
      "num_input_tokens_seen": 122977410,
      "step": 5718,
      "time_per_iteration": 2.6979007720947266
    },
    {
      "auxiliary_loss_clip": 0.01136389,
      "auxiliary_loss_mlp": 0.01026618,
      "balance_loss_clip": 1.04818332,
      "balance_loss_mlp": 1.01915002,
      "epoch": 0.6876690915649613,
      "flos": 25516937445120.0,
      "grad_norm": 2.3546452956058266,
      "language_loss": 0.62164599,
      "learning_rate": 9.388112365937608e-07,
      "loss": 0.6432761,
      "num_input_tokens_seen": 122996875,
      "step": 5719,
      "time_per_iteration": 2.7687418460845947
    },
    {
      "auxiliary_loss_clip": 0.01158888,
      "auxiliary_loss_mlp": 0.0076258,
      "balance_loss_clip": 1.04884851,
      "balance_loss_mlp": 1.00058961,
      "epoch": 0.6877893344556003,
      "flos": 19427996568960.0,
      "grad_norm": 2.2051503610210927,
      "language_loss": 0.82643819,
      "learning_rate": 9.381510405870985e-07,
      "loss": 0.84565288,
      "num_input_tokens_seen": 123015890,
      "step": 5720,
      "time_per_iteration": 2.6209328174591064
    },
    {
      "auxiliary_loss_clip": 0.0115199,
      "auxiliary_loss_mlp": 0.01029365,
      "balance_loss_clip": 1.05277705,
      "balance_loss_mlp": 1.02120483,
      "epoch": 0.6879095773462394,
      "flos": 18661303745280.0,
      "grad_norm": 2.0993773894585948,
      "language_loss": 0.77969027,
      "learning_rate": 9.374910056620791e-07,
      "loss": 0.80150378,
      "num_input_tokens_seen": 123034955,
      "step": 5721,
      "time_per_iteration": 2.6615288257598877
    },
    {
      "auxiliary_loss_clip": 0.0118158,
      "auxiliary_loss_mlp": 0.01030897,
      "balance_loss_clip": 1.05482447,
      "balance_loss_mlp": 1.02276099,
      "epoch": 0.6880298202368785,
      "flos": 20883132437760.0,
      "grad_norm": 1.7054543981556074,
      "language_loss": 0.81316596,
      "learning_rate": 9.368311319188293e-07,
      "loss": 0.83529073,
      "num_input_tokens_seen": 123052770,
      "step": 5722,
      "time_per_iteration": 2.629817247390747
    },
    {
      "auxiliary_loss_clip": 0.01153956,
      "auxiliary_loss_mlp": 0.01030542,
      "balance_loss_clip": 1.04911876,
      "balance_loss_mlp": 1.02300167,
      "epoch": 0.6881500631275176,
      "flos": 30153292318080.0,
      "grad_norm": 1.8894454069682158,
      "language_loss": 0.79920983,
      "learning_rate": 9.361714194574515e-07,
      "loss": 0.82105476,
      "num_input_tokens_seen": 123075105,
      "step": 5723,
      "time_per_iteration": 3.6508054733276367
    },
    {
      "auxiliary_loss_clip": 0.01056998,
      "auxiliary_loss_mlp": 0.01002018,
      "balance_loss_clip": 1.00950265,
      "balance_loss_mlp": 1.00111806,
      "epoch": 0.6882703060181566,
      "flos": 66181537215360.0,
      "grad_norm": 0.7321686523396935,
      "language_loss": 0.58230209,
      "learning_rate": 9.355118683780228e-07,
      "loss": 0.60289228,
      "num_input_tokens_seen": 123145175,
      "step": 5724,
      "time_per_iteration": 3.3870532512664795
    },
    {
      "auxiliary_loss_clip": 0.01153018,
      "auxiliary_loss_mlp": 0.01026858,
      "balance_loss_clip": 1.04762101,
      "balance_loss_mlp": 1.01900172,
      "epoch": 0.6883905489087958,
      "flos": 18214646123520.0,
      "grad_norm": 2.2587027671968047,
      "language_loss": 0.79736441,
      "learning_rate": 9.348524787805987e-07,
      "loss": 0.8191632,
      "num_input_tokens_seen": 123160365,
      "step": 5725,
      "time_per_iteration": 2.587374210357666
    },
    {
      "auxiliary_loss_clip": 0.01148068,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.04635715,
      "balance_loss_mlp": 1.01892757,
      "epoch": 0.6885107917994349,
      "flos": 14056262553600.0,
      "grad_norm": 2.41476948635531,
      "language_loss": 0.85047466,
      "learning_rate": 9.341932507652053e-07,
      "loss": 0.87222826,
      "num_input_tokens_seen": 123174855,
      "step": 5726,
      "time_per_iteration": 2.6757616996765137
    },
    {
      "auxiliary_loss_clip": 0.01141642,
      "auxiliary_loss_mlp": 0.01029364,
      "balance_loss_clip": 1.04777575,
      "balance_loss_mlp": 1.02087677,
      "epoch": 0.6886310346900739,
      "flos": 28690722334080.0,
      "grad_norm": 1.638240259142599,
      "language_loss": 0.78862405,
      "learning_rate": 9.335341844318489e-07,
      "loss": 0.81033403,
      "num_input_tokens_seen": 123194995,
      "step": 5727,
      "time_per_iteration": 2.7144713401794434
    },
    {
      "auxiliary_loss_clip": 0.01176642,
      "auxiliary_loss_mlp": 0.01028551,
      "balance_loss_clip": 1.05262053,
      "balance_loss_mlp": 1.02049804,
      "epoch": 0.6887512775807131,
      "flos": 24535319592960.0,
      "grad_norm": 1.7150943065309052,
      "language_loss": 0.73534513,
      "learning_rate": 9.328752798805091e-07,
      "loss": 0.75739706,
      "num_input_tokens_seen": 123213465,
      "step": 5728,
      "time_per_iteration": 2.7432518005371094
    },
    {
      "auxiliary_loss_clip": 0.01164056,
      "auxiliary_loss_mlp": 0.01031142,
      "balance_loss_clip": 1.054178,
      "balance_loss_mlp": 1.02323878,
      "epoch": 0.6888715204713521,
      "flos": 22414363269120.0,
      "grad_norm": 3.1512535947337663,
      "language_loss": 0.76113367,
      "learning_rate": 9.322165372111399e-07,
      "loss": 0.78308558,
      "num_input_tokens_seen": 123231610,
      "step": 5729,
      "time_per_iteration": 2.6581993103027344
    },
    {
      "auxiliary_loss_clip": 0.01155716,
      "auxiliary_loss_mlp": 0.01031484,
      "balance_loss_clip": 1.04982424,
      "balance_loss_mlp": 1.02362776,
      "epoch": 0.6889917633619912,
      "flos": 22054323294720.0,
      "grad_norm": 2.1475289899107994,
      "language_loss": 0.76095045,
      "learning_rate": 9.315579565236747e-07,
      "loss": 0.78282243,
      "num_input_tokens_seen": 123250715,
      "step": 5730,
      "time_per_iteration": 2.7011048793792725
    },
    {
      "auxiliary_loss_clip": 0.01178617,
      "auxiliary_loss_mlp": 0.01038153,
      "balance_loss_clip": 1.05250418,
      "balance_loss_mlp": 1.02993917,
      "epoch": 0.6891120062526304,
      "flos": 23949724164480.0,
      "grad_norm": 1.7530721620301555,
      "language_loss": 0.74141884,
      "learning_rate": 9.308995379180162e-07,
      "loss": 0.76358652,
      "num_input_tokens_seen": 123270270,
      "step": 5731,
      "time_per_iteration": 2.620842218399048
    },
    {
      "auxiliary_loss_clip": 0.01057449,
      "auxiliary_loss_mlp": 0.01005311,
      "balance_loss_clip": 1.01061928,
      "balance_loss_mlp": 1.0043515,
      "epoch": 0.6892322491432694,
      "flos": 64117354337280.0,
      "grad_norm": 0.7439264542069753,
      "language_loss": 0.59478581,
      "learning_rate": 9.302412814940488e-07,
      "loss": 0.61541343,
      "num_input_tokens_seen": 123333045,
      "step": 5732,
      "time_per_iteration": 3.2953555583953857
    },
    {
      "auxiliary_loss_clip": 0.01154861,
      "auxiliary_loss_mlp": 0.01033399,
      "balance_loss_clip": 1.05132866,
      "balance_loss_mlp": 1.02571917,
      "epoch": 0.6893524920339085,
      "flos": 23002436736000.0,
      "grad_norm": 8.511712827600451,
      "language_loss": 0.7118665,
      "learning_rate": 9.295831873516276e-07,
      "loss": 0.73374909,
      "num_input_tokens_seen": 123352320,
      "step": 5733,
      "time_per_iteration": 2.7771027088165283
    },
    {
      "auxiliary_loss_clip": 0.01151279,
      "auxiliary_loss_mlp": 0.01035193,
      "balance_loss_clip": 1.04983735,
      "balance_loss_mlp": 1.02627587,
      "epoch": 0.6894727349245476,
      "flos": 21396260177280.0,
      "grad_norm": 1.9160384290295294,
      "language_loss": 0.75874627,
      "learning_rate": 9.289252555905873e-07,
      "loss": 0.78061092,
      "num_input_tokens_seen": 123372400,
      "step": 5734,
      "time_per_iteration": 2.7021965980529785
    },
    {
      "auxiliary_loss_clip": 0.01136394,
      "auxiliary_loss_mlp": 0.01034144,
      "balance_loss_clip": 1.05220199,
      "balance_loss_mlp": 1.02576399,
      "epoch": 0.6895929778151867,
      "flos": 19865316654720.0,
      "grad_norm": 1.9123536333520026,
      "language_loss": 0.75764227,
      "learning_rate": 9.282674863107334e-07,
      "loss": 0.77934766,
      "num_input_tokens_seen": 123390215,
      "step": 5735,
      "time_per_iteration": 2.7382194995880127
    },
    {
      "auxiliary_loss_clip": 0.01164929,
      "auxiliary_loss_mlp": 0.01033777,
      "balance_loss_clip": 1.0502944,
      "balance_loss_mlp": 1.0254209,
      "epoch": 0.6897132207058257,
      "flos": 18179166464640.0,
      "grad_norm": 35.863411841487775,
      "language_loss": 0.75901914,
      "learning_rate": 9.276098796118488e-07,
      "loss": 0.78100622,
      "num_input_tokens_seen": 123406700,
      "step": 5736,
      "time_per_iteration": 2.625403881072998
    },
    {
      "auxiliary_loss_clip": 0.01160586,
      "auxiliary_loss_mlp": 0.01026511,
      "balance_loss_clip": 1.05136991,
      "balance_loss_mlp": 1.01878023,
      "epoch": 0.6898334635964649,
      "flos": 32561641359360.0,
      "grad_norm": 1.961322944034209,
      "language_loss": 0.66250324,
      "learning_rate": 9.269524355936938e-07,
      "loss": 0.68437415,
      "num_input_tokens_seen": 123429880,
      "step": 5737,
      "time_per_iteration": 3.671311140060425
    },
    {
      "auxiliary_loss_clip": 0.01174693,
      "auxiliary_loss_mlp": 0.01028857,
      "balance_loss_clip": 1.05308771,
      "balance_loss_mlp": 1.02145767,
      "epoch": 0.689953706487104,
      "flos": 22819004956800.0,
      "grad_norm": 1.8162462917134057,
      "language_loss": 0.84936506,
      "learning_rate": 9.262951543560002e-07,
      "loss": 0.87140054,
      "num_input_tokens_seen": 123449105,
      "step": 5738,
      "time_per_iteration": 2.6985368728637695
    },
    {
      "auxiliary_loss_clip": 0.01179894,
      "auxiliary_loss_mlp": 0.01027518,
      "balance_loss_clip": 1.05435824,
      "balance_loss_mlp": 1.01908958,
      "epoch": 0.690073949377743,
      "flos": 18515362786560.0,
      "grad_norm": 3.7021922831506817,
      "language_loss": 0.86489439,
      "learning_rate": 9.256380359984795e-07,
      "loss": 0.88696855,
      "num_input_tokens_seen": 123466215,
      "step": 5739,
      "time_per_iteration": 2.5804476737976074
    },
    {
      "auxiliary_loss_clip": 0.01145661,
      "auxiliary_loss_mlp": 0.01030907,
      "balance_loss_clip": 1.04850245,
      "balance_loss_mlp": 1.02229464,
      "epoch": 0.6901941922683821,
      "flos": 34857194716800.0,
      "grad_norm": 1.95362473946117,
      "language_loss": 0.74840283,
      "learning_rate": 9.249810806208139e-07,
      "loss": 0.77016854,
      "num_input_tokens_seen": 123485480,
      "step": 5740,
      "time_per_iteration": 4.595240592956543
    },
    {
      "auxiliary_loss_clip": 0.01160012,
      "auxiliary_loss_mlp": 0.01029585,
      "balance_loss_clip": 1.05159783,
      "balance_loss_mlp": 1.02208662,
      "epoch": 0.6903144351590212,
      "flos": 16253672976000.0,
      "grad_norm": 2.0639911589972053,
      "language_loss": 0.80179662,
      "learning_rate": 9.243242883226627e-07,
      "loss": 0.82369256,
      "num_input_tokens_seen": 123504575,
      "step": 5741,
      "time_per_iteration": 2.617684841156006
    },
    {
      "auxiliary_loss_clip": 0.0117727,
      "auxiliary_loss_mlp": 0.01022896,
      "balance_loss_clip": 1.05049872,
      "balance_loss_mlp": 1.01446748,
      "epoch": 0.6904346780496603,
      "flos": 28035137255040.0,
      "grad_norm": 1.7619012376230592,
      "language_loss": 0.69871086,
      "learning_rate": 9.236676592036628e-07,
      "loss": 0.72071254,
      "num_input_tokens_seen": 123524250,
      "step": 5742,
      "time_per_iteration": 2.723278760910034
    },
    {
      "auxiliary_loss_clip": 0.01161468,
      "auxiliary_loss_mlp": 0.010296,
      "balance_loss_clip": 1.05180347,
      "balance_loss_mlp": 1.02140713,
      "epoch": 0.6905549209402994,
      "flos": 23624266008960.0,
      "grad_norm": 1.7328829516355309,
      "language_loss": 0.73703504,
      "learning_rate": 9.230111933634228e-07,
      "loss": 0.75894582,
      "num_input_tokens_seen": 123545845,
      "step": 5743,
      "time_per_iteration": 2.66194748878479
    },
    {
      "auxiliary_loss_clip": 0.01165499,
      "auxiliary_loss_mlp": 0.01028982,
      "balance_loss_clip": 1.05042553,
      "balance_loss_mlp": 1.02085769,
      "epoch": 0.6906751638309385,
      "flos": 23114945111040.0,
      "grad_norm": 1.519580170437584,
      "language_loss": 0.80961978,
      "learning_rate": 9.223548909015288e-07,
      "loss": 0.83156455,
      "num_input_tokens_seen": 123567535,
      "step": 5744,
      "time_per_iteration": 2.7013401985168457
    },
    {
      "auxiliary_loss_clip": 0.01175589,
      "auxiliary_loss_mlp": 0.01025159,
      "balance_loss_clip": 1.05271351,
      "balance_loss_mlp": 1.01770854,
      "epoch": 0.6907954067215776,
      "flos": 27305468375040.0,
      "grad_norm": 2.0316639498815166,
      "language_loss": 0.71804917,
      "learning_rate": 9.216987519175407e-07,
      "loss": 0.74005669,
      "num_input_tokens_seen": 123587710,
      "step": 5745,
      "time_per_iteration": 2.8809890747070312
    },
    {
      "auxiliary_loss_clip": 0.01136225,
      "auxiliary_loss_mlp": 0.01026207,
      "balance_loss_clip": 1.04762244,
      "balance_loss_mlp": 1.01817787,
      "epoch": 0.6909156496122166,
      "flos": 21689399070720.0,
      "grad_norm": 1.6552227781706517,
      "language_loss": 0.68568832,
      "learning_rate": 9.210427765109942e-07,
      "loss": 0.70731258,
      "num_input_tokens_seen": 123607385,
      "step": 5746,
      "time_per_iteration": 2.69461989402771
    },
    {
      "auxiliary_loss_clip": 0.01181464,
      "auxiliary_loss_mlp": 0.01032522,
      "balance_loss_clip": 1.0530684,
      "balance_loss_mlp": 1.02316403,
      "epoch": 0.6910358925028558,
      "flos": 22561453463040.0,
      "grad_norm": 1.9245060939059828,
      "language_loss": 0.81124187,
      "learning_rate": 9.20386964781402e-07,
      "loss": 0.83338177,
      "num_input_tokens_seen": 123625405,
      "step": 5747,
      "time_per_iteration": 2.651440143585205
    },
    {
      "auxiliary_loss_clip": 0.01163279,
      "auxiliary_loss_mlp": 0.0102613,
      "balance_loss_clip": 1.05275261,
      "balance_loss_mlp": 1.0183928,
      "epoch": 0.6911561353934949,
      "flos": 22054107813120.0,
      "grad_norm": 2.213907773199706,
      "language_loss": 0.84572816,
      "learning_rate": 9.197313168282472e-07,
      "loss": 0.86762226,
      "num_input_tokens_seen": 123642850,
      "step": 5748,
      "time_per_iteration": 2.668717384338379
    },
    {
      "auxiliary_loss_clip": 0.01163233,
      "auxiliary_loss_mlp": 0.0102515,
      "balance_loss_clip": 1.04803503,
      "balance_loss_mlp": 1.01703155,
      "epoch": 0.6912763782841339,
      "flos": 24206557386240.0,
      "grad_norm": 2.555620267882549,
      "language_loss": 0.72334957,
      "learning_rate": 9.190758327509935e-07,
      "loss": 0.74523336,
      "num_input_tokens_seen": 123661595,
      "step": 5749,
      "time_per_iteration": 3.6167619228363037
    },
    {
      "auxiliary_loss_clip": 0.01053677,
      "auxiliary_loss_mlp": 0.0100226,
      "balance_loss_clip": 1.01058054,
      "balance_loss_mlp": 1.00135362,
      "epoch": 0.6913966211747731,
      "flos": 52329641091840.0,
      "grad_norm": 0.9314024279311728,
      "language_loss": 0.64327693,
      "learning_rate": 9.184205126490767e-07,
      "loss": 0.66383636,
      "num_input_tokens_seen": 123710490,
      "step": 5750,
      "time_per_iteration": 3.0716171264648438
    },
    {
      "auxiliary_loss_clip": 0.01053508,
      "auxiliary_loss_mlp": 0.00999694,
      "balance_loss_clip": 1.01017821,
      "balance_loss_mlp": 0.99881762,
      "epoch": 0.6915168640654121,
      "flos": 66741274851840.0,
      "grad_norm": 1.1127551739669732,
      "language_loss": 0.59587705,
      "learning_rate": 9.177653566219075e-07,
      "loss": 0.61640906,
      "num_input_tokens_seen": 123765215,
      "step": 5751,
      "time_per_iteration": 3.0662553310394287
    },
    {
      "auxiliary_loss_clip": 0.01177241,
      "auxiliary_loss_mlp": 0.01028204,
      "balance_loss_clip": 1.05169868,
      "balance_loss_mlp": 1.01999092,
      "epoch": 0.6916371069560512,
      "flos": 18296523175680.0,
      "grad_norm": 3.0250749223460534,
      "language_loss": 0.76062894,
      "learning_rate": 9.171103647688744e-07,
      "loss": 0.78268337,
      "num_input_tokens_seen": 123783955,
      "step": 5752,
      "time_per_iteration": 2.6353893280029297
    },
    {
      "auxiliary_loss_clip": 0.01173671,
      "auxiliary_loss_mlp": 0.00762214,
      "balance_loss_clip": 1.05261886,
      "balance_loss_mlp": 1.00067019,
      "epoch": 0.6917573498466904,
      "flos": 19645794685440.0,
      "grad_norm": 2.048640609320984,
      "language_loss": 0.69287091,
      "learning_rate": 9.164555371893367e-07,
      "loss": 0.71222973,
      "num_input_tokens_seen": 123803885,
      "step": 5753,
      "time_per_iteration": 2.5779519081115723
    },
    {
      "auxiliary_loss_clip": 0.01175505,
      "auxiliary_loss_mlp": 0.01027425,
      "balance_loss_clip": 1.05295479,
      "balance_loss_mlp": 1.019665,
      "epoch": 0.6918775927373294,
      "flos": 14210319985920.0,
      "grad_norm": 1.962379957437876,
      "language_loss": 0.75531417,
      "learning_rate": 9.158008739826333e-07,
      "loss": 0.77734351,
      "num_input_tokens_seen": 123821485,
      "step": 5754,
      "time_per_iteration": 2.637052536010742
    },
    {
      "auxiliary_loss_clip": 0.01179998,
      "auxiliary_loss_mlp": 0.01024404,
      "balance_loss_clip": 1.05374408,
      "balance_loss_mlp": 1.01625001,
      "epoch": 0.6919978356279685,
      "flos": 23985455218560.0,
      "grad_norm": 1.9842867575610927,
      "language_loss": 0.86901605,
      "learning_rate": 9.151463752480744e-07,
      "loss": 0.89105999,
      "num_input_tokens_seen": 123840215,
      "step": 5755,
      "time_per_iteration": 2.6037864685058594
    },
    {
      "auxiliary_loss_clip": 0.01162129,
      "auxiliary_loss_mlp": 0.0102863,
      "balance_loss_clip": 1.04941702,
      "balance_loss_mlp": 1.02082181,
      "epoch": 0.6921180785186076,
      "flos": 23622937205760.0,
      "grad_norm": 1.5250548021776804,
      "language_loss": 0.80328929,
      "learning_rate": 9.144920410849493e-07,
      "loss": 0.82519692,
      "num_input_tokens_seen": 123861450,
      "step": 5756,
      "time_per_iteration": 2.6558852195739746
    },
    {
      "auxiliary_loss_clip": 0.0116293,
      "auxiliary_loss_mlp": 0.01029148,
      "balance_loss_clip": 1.05049324,
      "balance_loss_mlp": 1.02074385,
      "epoch": 0.6922383214092467,
      "flos": 21142623265920.0,
      "grad_norm": 2.459648153440322,
      "language_loss": 0.80410624,
      "learning_rate": 9.138378715925176e-07,
      "loss": 0.82602698,
      "num_input_tokens_seen": 123880545,
      "step": 5757,
      "time_per_iteration": 2.6124627590179443
    },
    {
      "auxiliary_loss_clip": 0.01175598,
      "auxiliary_loss_mlp": 0.01029021,
      "balance_loss_clip": 1.05319285,
      "balance_loss_mlp": 1.02143955,
      "epoch": 0.6923585642998857,
      "flos": 21470667200640.0,
      "grad_norm": 1.9184279536570792,
      "language_loss": 0.80995792,
      "learning_rate": 9.131838668700167e-07,
      "loss": 0.83200413,
      "num_input_tokens_seen": 123900615,
      "step": 5758,
      "time_per_iteration": 2.639786720275879
    },
    {
      "auxiliary_loss_clip": 0.01149259,
      "auxiliary_loss_mlp": 0.01033374,
      "balance_loss_clip": 1.05156159,
      "balance_loss_mlp": 1.02558661,
      "epoch": 0.6924788071905249,
      "flos": 21105204272640.0,
      "grad_norm": 1.867436612128533,
      "language_loss": 0.86393774,
      "learning_rate": 9.125300270166598e-07,
      "loss": 0.88576412,
      "num_input_tokens_seen": 123921220,
      "step": 5759,
      "time_per_iteration": 2.7080118656158447
    },
    {
      "auxiliary_loss_clip": 0.01161466,
      "auxiliary_loss_mlp": 0.01031922,
      "balance_loss_clip": 1.04923654,
      "balance_loss_mlp": 1.02346456,
      "epoch": 0.692599050081164,
      "flos": 26250018117120.0,
      "grad_norm": 1.8014222657187524,
      "language_loss": 0.85887098,
      "learning_rate": 9.118763521316324e-07,
      "loss": 0.88080484,
      "num_input_tokens_seen": 123941795,
      "step": 5760,
      "time_per_iteration": 2.7185285091400146
    },
    {
      "auxiliary_loss_clip": 0.01147236,
      "auxiliary_loss_mlp": 0.01028296,
      "balance_loss_clip": 1.05123448,
      "balance_loss_mlp": 1.02039194,
      "epoch": 0.692719292971803,
      "flos": 20885215426560.0,
      "grad_norm": 1.788308221721906,
      "language_loss": 0.76401454,
      "learning_rate": 9.112228423140987e-07,
      "loss": 0.78576988,
      "num_input_tokens_seen": 123960715,
      "step": 5761,
      "time_per_iteration": 2.6974916458129883
    },
    {
      "auxiliary_loss_clip": 0.01129445,
      "auxiliary_loss_mlp": 0.01026289,
      "balance_loss_clip": 1.04627967,
      "balance_loss_mlp": 1.0184269,
      "epoch": 0.6928395358624422,
      "flos": 25921938268800.0,
      "grad_norm": 2.4073806804508915,
      "language_loss": 0.86510807,
      "learning_rate": 9.105694976631932e-07,
      "loss": 0.88666546,
      "num_input_tokens_seen": 123978625,
      "step": 5762,
      "time_per_iteration": 2.8009936809539795
    },
    {
      "auxiliary_loss_clip": 0.01181527,
      "auxiliary_loss_mlp": 0.01034732,
      "balance_loss_clip": 1.0563612,
      "balance_loss_mlp": 1.0266552,
      "epoch": 0.6929597787530812,
      "flos": 23586559706880.0,
      "grad_norm": 2.074555562252309,
      "language_loss": 0.72761035,
      "learning_rate": 9.099163182780283e-07,
      "loss": 0.74977291,
      "num_input_tokens_seen": 123996780,
      "step": 5763,
      "time_per_iteration": 3.6294381618499756
    },
    {
      "auxiliary_loss_clip": 0.01168207,
      "auxiliary_loss_mlp": 0.01032292,
      "balance_loss_clip": 1.05252361,
      "balance_loss_mlp": 1.0239054,
      "epoch": 0.6930800216437203,
      "flos": 18255656476800.0,
      "grad_norm": 4.379991587722173,
      "language_loss": 0.48585626,
      "learning_rate": 9.092633042576916e-07,
      "loss": 0.50786126,
      "num_input_tokens_seen": 124014045,
      "step": 5764,
      "time_per_iteration": 2.594810724258423
    },
    {
      "auxiliary_loss_clip": 0.01145523,
      "auxiliary_loss_mlp": 0.010323,
      "balance_loss_clip": 1.04759479,
      "balance_loss_mlp": 1.02434278,
      "epoch": 0.6932002645343595,
      "flos": 29168621809920.0,
      "grad_norm": 2.2427898973672695,
      "language_loss": 0.565732,
      "learning_rate": 9.086104557012446e-07,
      "loss": 0.58751023,
      "num_input_tokens_seen": 124034615,
      "step": 5765,
      "time_per_iteration": 3.627570152282715
    },
    {
      "auxiliary_loss_clip": 0.01176143,
      "auxiliary_loss_mlp": 0.00762046,
      "balance_loss_clip": 1.05339885,
      "balance_loss_mlp": 1.0007112,
      "epoch": 0.6933205074249985,
      "flos": 23842746483840.0,
      "grad_norm": 2.372882745226897,
      "language_loss": 0.66001117,
      "learning_rate": 9.079577727077239e-07,
      "loss": 0.67939305,
      "num_input_tokens_seen": 124053445,
      "step": 5766,
      "time_per_iteration": 2.6626229286193848
    },
    {
      "auxiliary_loss_clip": 0.01179348,
      "auxiliary_loss_mlp": 0.01029976,
      "balance_loss_clip": 1.05334044,
      "balance_loss_mlp": 1.02164292,
      "epoch": 0.6934407503156376,
      "flos": 24166696268160.0,
      "grad_norm": 2.250482824996758,
      "language_loss": 0.72156394,
      "learning_rate": 9.073052553761404e-07,
      "loss": 0.74365717,
      "num_input_tokens_seen": 124072810,
      "step": 5767,
      "time_per_iteration": 3.5591983795166016
    },
    {
      "auxiliary_loss_clip": 0.01165877,
      "auxiliary_loss_mlp": 0.01026952,
      "balance_loss_clip": 1.05113816,
      "balance_loss_mlp": 1.01774275,
      "epoch": 0.6935609932062767,
      "flos": 20631327120000.0,
      "grad_norm": 1.6491853169325177,
      "language_loss": 0.7816999,
      "learning_rate": 9.066529038054805e-07,
      "loss": 0.80362821,
      "num_input_tokens_seen": 124092875,
      "step": 5768,
      "time_per_iteration": 2.691490411758423
    },
    {
      "auxiliary_loss_clip": 0.01138282,
      "auxiliary_loss_mlp": 0.01033634,
      "balance_loss_clip": 1.04946756,
      "balance_loss_mlp": 1.02607632,
      "epoch": 0.6936812360969158,
      "flos": 18254184019200.0,
      "grad_norm": 1.769680593593471,
      "language_loss": 0.74089575,
      "learning_rate": 9.060007180947071e-07,
      "loss": 0.76261497,
      "num_input_tokens_seen": 124110930,
      "step": 5769,
      "time_per_iteration": 2.6513495445251465
    },
    {
      "auxiliary_loss_clip": 0.01179369,
      "auxiliary_loss_mlp": 0.01026629,
      "balance_loss_clip": 1.05354977,
      "balance_loss_mlp": 1.01843333,
      "epoch": 0.6938014789875548,
      "flos": 31317336368640.0,
      "grad_norm": 1.7580139626195297,
      "language_loss": 0.73079199,
      "learning_rate": 9.053486983427534e-07,
      "loss": 0.75285196,
      "num_input_tokens_seen": 124132180,
      "step": 5770,
      "time_per_iteration": 2.701890230178833
    },
    {
      "auxiliary_loss_clip": 0.01176518,
      "auxiliary_loss_mlp": 0.01026879,
      "balance_loss_clip": 1.05256331,
      "balance_loss_mlp": 1.01860559,
      "epoch": 0.6939217218781939,
      "flos": 17528429721600.0,
      "grad_norm": 2.375958222826987,
      "language_loss": 0.70810103,
      "learning_rate": 9.046968446485326e-07,
      "loss": 0.73013496,
      "num_input_tokens_seen": 124150585,
      "step": 5771,
      "time_per_iteration": 2.5565407276153564
    },
    {
      "auxiliary_loss_clip": 0.01158497,
      "auxiliary_loss_mlp": 0.01030664,
      "balance_loss_clip": 1.04932165,
      "balance_loss_mlp": 1.02202749,
      "epoch": 0.6940419647688331,
      "flos": 18551776199040.0,
      "grad_norm": 2.5625744341077614,
      "language_loss": 0.70555675,
      "learning_rate": 9.040451571109295e-07,
      "loss": 0.72744834,
      "num_input_tokens_seen": 124166205,
      "step": 5772,
      "time_per_iteration": 2.6700685024261475
    },
    {
      "auxiliary_loss_clip": 0.01054519,
      "auxiliary_loss_mlp": 0.01002446,
      "balance_loss_clip": 1.01147819,
      "balance_loss_mlp": 1.00143278,
      "epoch": 0.6941622076594721,
      "flos": 66926286829440.0,
      "grad_norm": 0.8317512896965218,
      "language_loss": 0.60305226,
      "learning_rate": 9.033936358288042e-07,
      "loss": 0.62362194,
      "num_input_tokens_seen": 124219940,
      "step": 5773,
      "time_per_iteration": 3.1285159587860107
    },
    {
      "auxiliary_loss_clip": 0.01144392,
      "auxiliary_loss_mlp": 0.01028102,
      "balance_loss_clip": 1.04752684,
      "balance_loss_mlp": 1.01973963,
      "epoch": 0.6942824505501112,
      "flos": 26578062051840.0,
      "grad_norm": 5.338349844828294,
      "language_loss": 0.82529587,
      "learning_rate": 9.027422809009937e-07,
      "loss": 0.84702086,
      "num_input_tokens_seen": 124239885,
      "step": 5774,
      "time_per_iteration": 2.72232723236084
    },
    {
      "auxiliary_loss_clip": 0.01158912,
      "auxiliary_loss_mlp": 0.01030136,
      "balance_loss_clip": 1.05101633,
      "balance_loss_mlp": 1.02198267,
      "epoch": 0.6944026934407503,
      "flos": 21248308056960.0,
      "grad_norm": 3.4422433365847684,
      "language_loss": 0.83535504,
      "learning_rate": 9.020910924263054e-07,
      "loss": 0.8572455,
      "num_input_tokens_seen": 124258410,
      "step": 5775,
      "time_per_iteration": 3.609708786010742
    },
    {
      "auxiliary_loss_clip": 0.01043171,
      "auxiliary_loss_mlp": 0.01002295,
      "balance_loss_clip": 1.01214159,
      "balance_loss_mlp": 1.00134706,
      "epoch": 0.6945229363313894,
      "flos": 70677191537280.0,
      "grad_norm": 0.8205863848890587,
      "language_loss": 0.58057159,
      "learning_rate": 9.014400705035261e-07,
      "loss": 0.60102624,
      "num_input_tokens_seen": 124315315,
      "step": 5776,
      "time_per_iteration": 3.2749640941619873
    },
    {
      "auxiliary_loss_clip": 0.01145808,
      "auxiliary_loss_mlp": 0.01038696,
      "balance_loss_clip": 1.04986262,
      "balance_loss_mlp": 1.03050053,
      "epoch": 0.6946431792220285,
      "flos": 18952934267520.0,
      "grad_norm": 1.9467157399498505,
      "language_loss": 0.7726382,
      "learning_rate": 9.00789215231414e-07,
      "loss": 0.79448318,
      "num_input_tokens_seen": 124333710,
      "step": 5777,
      "time_per_iteration": 2.6991047859191895
    },
    {
      "auxiliary_loss_clip": 0.01163197,
      "auxiliary_loss_mlp": 0.01029509,
      "balance_loss_clip": 1.05277061,
      "balance_loss_mlp": 1.02161717,
      "epoch": 0.6947634221126676,
      "flos": 20338834671360.0,
      "grad_norm": 2.019949639181017,
      "language_loss": 0.81994402,
      "learning_rate": 9.001385267087056e-07,
      "loss": 0.84187108,
      "num_input_tokens_seen": 124352855,
      "step": 5778,
      "time_per_iteration": 2.6499557495117188
    },
    {
      "auxiliary_loss_clip": 0.01160595,
      "auxiliary_loss_mlp": 0.01029952,
      "balance_loss_clip": 1.0521822,
      "balance_loss_mlp": 1.02139914,
      "epoch": 0.6948836650033067,
      "flos": 21833723917440.0,
      "grad_norm": 1.9505470755780847,
      "language_loss": 0.7073341,
      "learning_rate": 8.994880050341072e-07,
      "loss": 0.72923958,
      "num_input_tokens_seen": 124372960,
      "step": 5779,
      "time_per_iteration": 2.6590425968170166
    },
    {
      "auxiliary_loss_clip": 0.01171227,
      "auxiliary_loss_mlp": 0.0102953,
      "balance_loss_clip": 1.05808198,
      "balance_loss_mlp": 1.02144766,
      "epoch": 0.6950039078939457,
      "flos": 23657519024640.0,
      "grad_norm": 1.8143413688436076,
      "language_loss": 0.77745932,
      "learning_rate": 8.988376503063026e-07,
      "loss": 0.79946691,
      "num_input_tokens_seen": 124394220,
      "step": 5780,
      "time_per_iteration": 2.696352005004883
    },
    {
      "auxiliary_loss_clip": 0.01181505,
      "auxiliary_loss_mlp": 0.01029629,
      "balance_loss_clip": 1.05434513,
      "balance_loss_mlp": 1.02094507,
      "epoch": 0.6951241507845849,
      "flos": 21792462168960.0,
      "grad_norm": 1.989538493887388,
      "language_loss": 0.81675589,
      "learning_rate": 8.981874626239521e-07,
      "loss": 0.83886731,
      "num_input_tokens_seen": 124412795,
      "step": 5781,
      "time_per_iteration": 2.6349904537200928
    },
    {
      "auxiliary_loss_clip": 0.01181077,
      "auxiliary_loss_mlp": 0.01030058,
      "balance_loss_clip": 1.05486691,
      "balance_loss_mlp": 1.02202916,
      "epoch": 0.695244393675224,
      "flos": 14647568244480.0,
      "grad_norm": 1.9910031116947318,
      "language_loss": 0.88527423,
      "learning_rate": 8.975374420856872e-07,
      "loss": 0.90738559,
      "num_input_tokens_seen": 124429690,
      "step": 5782,
      "time_per_iteration": 2.6448771953582764
    },
    {
      "auxiliary_loss_clip": 0.01176403,
      "auxiliary_loss_mlp": 0.01032789,
      "balance_loss_clip": 1.05289984,
      "balance_loss_mlp": 1.02493358,
      "epoch": 0.695364636565863,
      "flos": 16873203778560.0,
      "grad_norm": 2.000319915203572,
      "language_loss": 0.7273829,
      "learning_rate": 8.968875887901157e-07,
      "loss": 0.74947476,
      "num_input_tokens_seen": 124447070,
      "step": 5783,
      "time_per_iteration": 2.602802276611328
    },
    {
      "auxiliary_loss_clip": 0.0114885,
      "auxiliary_loss_mlp": 0.01028459,
      "balance_loss_clip": 1.05041027,
      "balance_loss_mlp": 1.01998961,
      "epoch": 0.6954848794565022,
      "flos": 19354523299200.0,
      "grad_norm": 2.2091019203356463,
      "language_loss": 0.62650472,
      "learning_rate": 8.9623790283582e-07,
      "loss": 0.64827776,
      "num_input_tokens_seen": 124464950,
      "step": 5784,
      "time_per_iteration": 2.7812881469726562
    },
    {
      "auxiliary_loss_clip": 0.01172793,
      "auxiliary_loss_mlp": 0.0103194,
      "balance_loss_clip": 1.05630851,
      "balance_loss_mlp": 1.02382827,
      "epoch": 0.6956051223471412,
      "flos": 18990209606400.0,
      "grad_norm": 2.1830490388335972,
      "language_loss": 0.76583171,
      "learning_rate": 8.955883843213561e-07,
      "loss": 0.78787905,
      "num_input_tokens_seen": 124483965,
      "step": 5785,
      "time_per_iteration": 2.61442494392395
    },
    {
      "auxiliary_loss_clip": 0.01181409,
      "auxiliary_loss_mlp": 0.01033207,
      "balance_loss_clip": 1.05340743,
      "balance_loss_mlp": 1.02513099,
      "epoch": 0.6957253652377803,
      "flos": 16107229226880.0,
      "grad_norm": 2.826910890451639,
      "language_loss": 0.86701906,
      "learning_rate": 8.949390333452569e-07,
      "loss": 0.88916528,
      "num_input_tokens_seen": 124501910,
      "step": 5786,
      "time_per_iteration": 2.626389980316162
    },
    {
      "auxiliary_loss_clip": 0.01176485,
      "auxiliary_loss_mlp": 0.01028637,
      "balance_loss_clip": 1.05362177,
      "balance_loss_mlp": 1.02050114,
      "epoch": 0.6958456081284194,
      "flos": 29388646569600.0,
      "grad_norm": 1.788208817391169,
      "language_loss": 0.6768682,
      "learning_rate": 8.942898500060279e-07,
      "loss": 0.69891936,
      "num_input_tokens_seen": 124521625,
      "step": 5787,
      "time_per_iteration": 2.6488423347473145
    },
    {
      "auxiliary_loss_clip": 0.01143915,
      "auxiliary_loss_mlp": 0.01030343,
      "balance_loss_clip": 1.04932296,
      "balance_loss_mlp": 1.0219568,
      "epoch": 0.6959658510190585,
      "flos": 25154850395520.0,
      "grad_norm": 2.9453186836857275,
      "language_loss": 0.72186291,
      "learning_rate": 8.936408344021493e-07,
      "loss": 0.74360549,
      "num_input_tokens_seen": 124538540,
      "step": 5788,
      "time_per_iteration": 2.7466185092926025
    },
    {
      "auxiliary_loss_clip": 0.01184553,
      "auxiliary_loss_mlp": 0.01032796,
      "balance_loss_clip": 1.05613971,
      "balance_loss_mlp": 1.02421904,
      "epoch": 0.6960860939096976,
      "flos": 42814388759040.0,
      "grad_norm": 2.078166838427911,
      "language_loss": 0.71259475,
      "learning_rate": 8.929919866320765e-07,
      "loss": 0.73476827,
      "num_input_tokens_seen": 124559355,
      "step": 5789,
      "time_per_iteration": 3.7507970333099365
    },
    {
      "auxiliary_loss_clip": 0.01180097,
      "auxiliary_loss_mlp": 0.01032707,
      "balance_loss_clip": 1.05347323,
      "balance_loss_mlp": 1.02440953,
      "epoch": 0.6962063368003367,
      "flos": 17566566986880.0,
      "grad_norm": 2.0011931249514228,
      "language_loss": 0.81470066,
      "learning_rate": 8.923433067942385e-07,
      "loss": 0.83682871,
      "num_input_tokens_seen": 124577920,
      "step": 5790,
      "time_per_iteration": 2.566204786300659
    },
    {
      "auxiliary_loss_clip": 0.01175605,
      "auxiliary_loss_mlp": 0.01030803,
      "balance_loss_clip": 1.05352318,
      "balance_loss_mlp": 1.02267337,
      "epoch": 0.6963265796909758,
      "flos": 21251648021760.0,
      "grad_norm": 7.379329364314472,
      "language_loss": 0.68577069,
      "learning_rate": 8.916947949870417e-07,
      "loss": 0.70783478,
      "num_input_tokens_seen": 124597585,
      "step": 5791,
      "time_per_iteration": 3.519261360168457
    },
    {
      "auxiliary_loss_clip": 0.0104697,
      "auxiliary_loss_mlp": 0.01003455,
      "balance_loss_clip": 1.01115942,
      "balance_loss_mlp": 1.00251329,
      "epoch": 0.6964468225816148,
      "flos": 68828295801600.0,
      "grad_norm": 0.7406206054641298,
      "language_loss": 0.58083582,
      "learning_rate": 8.910464513088615e-07,
      "loss": 0.60134006,
      "num_input_tokens_seen": 124661625,
      "step": 5792,
      "time_per_iteration": 3.3357372283935547
    },
    {
      "auxiliary_loss_clip": 0.0116712,
      "auxiliary_loss_mlp": 0.01025191,
      "balance_loss_clip": 1.05163932,
      "balance_loss_mlp": 1.01725137,
      "epoch": 0.696567065472254,
      "flos": 18950887192320.0,
      "grad_norm": 2.087131239514999,
      "language_loss": 0.78494978,
      "learning_rate": 8.903982758580542e-07,
      "loss": 0.80687284,
      "num_input_tokens_seen": 124680565,
      "step": 5793,
      "time_per_iteration": 3.575833559036255
    },
    {
      "auxiliary_loss_clip": 0.0118086,
      "auxiliary_loss_mlp": 0.01030362,
      "balance_loss_clip": 1.05392098,
      "balance_loss_mlp": 1.02186298,
      "epoch": 0.696687308362893,
      "flos": 22856675345280.0,
      "grad_norm": 1.8472971260002995,
      "language_loss": 0.80142641,
      "learning_rate": 8.897502687329457e-07,
      "loss": 0.82353866,
      "num_input_tokens_seen": 124700365,
      "step": 5794,
      "time_per_iteration": 2.682075023651123
    },
    {
      "auxiliary_loss_clip": 0.01143913,
      "auxiliary_loss_mlp": 0.01028763,
      "balance_loss_clip": 1.04698622,
      "balance_loss_mlp": 1.0208534,
      "epoch": 0.6968075512535321,
      "flos": 24972926987520.0,
      "grad_norm": 5.634050205771925,
      "language_loss": 0.80070913,
      "learning_rate": 8.891024300318382e-07,
      "loss": 0.82243586,
      "num_input_tokens_seen": 124718935,
      "step": 5795,
      "time_per_iteration": 2.689330577850342
    },
    {
      "auxiliary_loss_clip": 0.01146848,
      "auxiliary_loss_mlp": 0.01028681,
      "balance_loss_clip": 1.05039334,
      "balance_loss_mlp": 1.02066386,
      "epoch": 0.6969277941441713,
      "flos": 21030438113280.0,
      "grad_norm": 1.510550619246858,
      "language_loss": 0.75858152,
      "learning_rate": 8.884547598530103e-07,
      "loss": 0.78033686,
      "num_input_tokens_seen": 124739505,
      "step": 5796,
      "time_per_iteration": 2.7548627853393555
    },
    {
      "auxiliary_loss_clip": 0.0116444,
      "auxiliary_loss_mlp": 0.01027419,
      "balance_loss_clip": 1.05062556,
      "balance_loss_mlp": 1.0190922,
      "epoch": 0.6970480370348103,
      "flos": 21579404647680.0,
      "grad_norm": 18.07587321016974,
      "language_loss": 0.75273907,
      "learning_rate": 8.8780725829471e-07,
      "loss": 0.77465773,
      "num_input_tokens_seen": 124757410,
      "step": 5797,
      "time_per_iteration": 2.6291558742523193
    },
    {
      "auxiliary_loss_clip": 0.01165797,
      "auxiliary_loss_mlp": 0.01026582,
      "balance_loss_clip": 1.0501864,
      "balance_loss_mlp": 1.01844025,
      "epoch": 0.6971682799254494,
      "flos": 22419175691520.0,
      "grad_norm": 2.0659810841647164,
      "language_loss": 0.77889705,
      "learning_rate": 8.87159925455165e-07,
      "loss": 0.80082083,
      "num_input_tokens_seen": 124777240,
      "step": 5798,
      "time_per_iteration": 2.744903087615967
    },
    {
      "auxiliary_loss_clip": 0.01178861,
      "auxiliary_loss_mlp": 0.01028518,
      "balance_loss_clip": 1.05408847,
      "balance_loss_mlp": 1.02017927,
      "epoch": 0.6972885228160886,
      "flos": 20005834659840.0,
      "grad_norm": 2.932232720470556,
      "language_loss": 0.73092467,
      "learning_rate": 8.865127614325738e-07,
      "loss": 0.75299847,
      "num_input_tokens_seen": 124795670,
      "step": 5799,
      "time_per_iteration": 2.5708587169647217
    },
    {
      "auxiliary_loss_clip": 0.01159218,
      "auxiliary_loss_mlp": 0.01031065,
      "balance_loss_clip": 1.05033219,
      "balance_loss_mlp": 1.02213621,
      "epoch": 0.6974087657067276,
      "flos": 37853437656960.0,
      "grad_norm": 2.2067364132765483,
      "language_loss": 0.66948676,
      "learning_rate": 8.85865766325113e-07,
      "loss": 0.69138962,
      "num_input_tokens_seen": 124819600,
      "step": 5800,
      "time_per_iteration": 2.7647366523742676
    },
    {
      "auxiliary_loss_clip": 0.01148203,
      "auxiliary_loss_mlp": 0.01031644,
      "balance_loss_clip": 1.04990411,
      "balance_loss_mlp": 1.02292967,
      "epoch": 0.6975290085973667,
      "flos": 29489267543040.0,
      "grad_norm": 2.9494527037133182,
      "language_loss": 0.72145849,
      "learning_rate": 8.852189402309287e-07,
      "loss": 0.74325699,
      "num_input_tokens_seen": 124838785,
      "step": 5801,
      "time_per_iteration": 2.810045003890991
    },
    {
      "auxiliary_loss_clip": 0.01168838,
      "auxiliary_loss_mlp": 0.01031468,
      "balance_loss_clip": 1.05433214,
      "balance_loss_mlp": 1.02346265,
      "epoch": 0.6976492514880057,
      "flos": 12895630295040.0,
      "grad_norm": 2.364109293179616,
      "language_loss": 0.74231243,
      "learning_rate": 8.845722832481441e-07,
      "loss": 0.76431549,
      "num_input_tokens_seen": 124854215,
      "step": 5802,
      "time_per_iteration": 3.5276412963867188
    },
    {
      "auxiliary_loss_clip": 0.01164343,
      "auxiliary_loss_mlp": 0.01028428,
      "balance_loss_clip": 1.04840708,
      "balance_loss_mlp": 1.02041733,
      "epoch": 0.6977694943786449,
      "flos": 24352929308160.0,
      "grad_norm": 2.631215163246301,
      "language_loss": 0.77859402,
      "learning_rate": 8.83925795474858e-07,
      "loss": 0.80052173,
      "num_input_tokens_seen": 124874340,
      "step": 5803,
      "time_per_iteration": 2.7341132164001465
    },
    {
      "auxiliary_loss_clip": 0.01160515,
      "auxiliary_loss_mlp": 0.01029534,
      "balance_loss_clip": 1.05354977,
      "balance_loss_mlp": 1.02036071,
      "epoch": 0.6978897372692839,
      "flos": 29898470257920.0,
      "grad_norm": 2.485165691443787,
      "language_loss": 0.59717613,
      "learning_rate": 8.832794770091414e-07,
      "loss": 0.61907661,
      "num_input_tokens_seen": 124895175,
      "step": 5804,
      "time_per_iteration": 2.7213337421417236
    },
    {
      "auxiliary_loss_clip": 0.01164248,
      "auxiliary_loss_mlp": 0.01032095,
      "balance_loss_clip": 1.0506959,
      "balance_loss_mlp": 1.02390552,
      "epoch": 0.698009980159923,
      "flos": 21761579450880.0,
      "grad_norm": 1.9965499917224316,
      "language_loss": 0.82846469,
      "learning_rate": 8.826333279490401e-07,
      "loss": 0.8504281,
      "num_input_tokens_seen": 124915810,
      "step": 5805,
      "time_per_iteration": 2.7289493083953857
    },
    {
      "auxiliary_loss_clip": 0.01169596,
      "auxiliary_loss_mlp": 0.01030846,
      "balance_loss_clip": 1.05642915,
      "balance_loss_mlp": 1.02351761,
      "epoch": 0.6981302230505622,
      "flos": 19857164267520.0,
      "grad_norm": 2.2943854743266203,
      "language_loss": 0.68312579,
      "learning_rate": 8.819873483925748e-07,
      "loss": 0.70513016,
      "num_input_tokens_seen": 124932930,
      "step": 5806,
      "time_per_iteration": 2.682157516479492
    },
    {
      "auxiliary_loss_clip": 0.01178083,
      "auxiliary_loss_mlp": 0.01030563,
      "balance_loss_clip": 1.05300355,
      "balance_loss_mlp": 1.02243018,
      "epoch": 0.6982504659412012,
      "flos": 22198648141440.0,
      "grad_norm": 2.6287812331559084,
      "language_loss": 0.74541968,
      "learning_rate": 8.81341538437739e-07,
      "loss": 0.76750612,
      "num_input_tokens_seen": 124951220,
      "step": 5807,
      "time_per_iteration": 2.6406655311584473
    },
    {
      "auxiliary_loss_clip": 0.01161888,
      "auxiliary_loss_mlp": 0.01029863,
      "balance_loss_clip": 1.05123234,
      "balance_loss_mlp": 1.02139354,
      "epoch": 0.6983707088318403,
      "flos": 35588479708800.0,
      "grad_norm": 1.6512355884015142,
      "language_loss": 0.6839385,
      "learning_rate": 8.80695898182503e-07,
      "loss": 0.70585603,
      "num_input_tokens_seen": 124972200,
      "step": 5808,
      "time_per_iteration": 2.7774879932403564
    },
    {
      "auxiliary_loss_clip": 0.01048706,
      "auxiliary_loss_mlp": 0.01000124,
      "balance_loss_clip": 1.0108099,
      "balance_loss_mlp": 0.99918842,
      "epoch": 0.6984909517224794,
      "flos": 65440052760960.0,
      "grad_norm": 0.8157283125226592,
      "language_loss": 0.65010428,
      "learning_rate": 8.800504277248093e-07,
      "loss": 0.67059255,
      "num_input_tokens_seen": 125036950,
      "step": 5809,
      "time_per_iteration": 3.277247428894043
    },
    {
      "auxiliary_loss_clip": 0.01179769,
      "auxiliary_loss_mlp": 0.0102553,
      "balance_loss_clip": 1.05469823,
      "balance_loss_mlp": 1.01769805,
      "epoch": 0.6986111946131185,
      "flos": 18546927863040.0,
      "grad_norm": 1.9521288620736064,
      "language_loss": 0.75119805,
      "learning_rate": 8.794051271625753e-07,
      "loss": 0.77325106,
      "num_input_tokens_seen": 125054585,
      "step": 5810,
      "time_per_iteration": 2.6166181564331055
    },
    {
      "auxiliary_loss_clip": 0.01158505,
      "auxiliary_loss_mlp": 0.010261,
      "balance_loss_clip": 1.04715681,
      "balance_loss_mlp": 1.01816726,
      "epoch": 0.6987314375037575,
      "flos": 23039173370880.0,
      "grad_norm": 1.7845103781078315,
      "language_loss": 0.83419293,
      "learning_rate": 8.787599965936925e-07,
      "loss": 0.85603893,
      "num_input_tokens_seen": 125075515,
      "step": 5811,
      "time_per_iteration": 2.657285451889038
    },
    {
      "auxiliary_loss_clip": 0.01175864,
      "auxiliary_loss_mlp": 0.01025636,
      "balance_loss_clip": 1.05421901,
      "balance_loss_mlp": 1.01754189,
      "epoch": 0.6988516803943967,
      "flos": 38400393029760.0,
      "grad_norm": 1.7465783285347005,
      "language_loss": 0.72115099,
      "learning_rate": 8.781150361160261e-07,
      "loss": 0.74316609,
      "num_input_tokens_seen": 125097425,
      "step": 5812,
      "time_per_iteration": 2.817347764968872
    },
    {
      "auxiliary_loss_clip": 0.01163443,
      "auxiliary_loss_mlp": 0.01032448,
      "balance_loss_clip": 1.05526698,
      "balance_loss_mlp": 1.02400208,
      "epoch": 0.6989719232850358,
      "flos": 24096993926400.0,
      "grad_norm": 1.7465320945949145,
      "language_loss": 0.73857588,
      "learning_rate": 8.774702458274181e-07,
      "loss": 0.76053476,
      "num_input_tokens_seen": 125117830,
      "step": 5813,
      "time_per_iteration": 2.642341136932373
    },
    {
      "auxiliary_loss_clip": 0.01178933,
      "auxiliary_loss_mlp": 0.01026407,
      "balance_loss_clip": 1.05230653,
      "balance_loss_mlp": 1.01788092,
      "epoch": 0.6990921661756748,
      "flos": 14866838818560.0,
      "grad_norm": 2.1660896291756333,
      "language_loss": 0.70750463,
      "learning_rate": 8.768256258256799e-07,
      "loss": 0.72955799,
      "num_input_tokens_seen": 125134455,
      "step": 5814,
      "time_per_iteration": 2.618055582046509
    },
    {
      "auxiliary_loss_clip": 0.01179412,
      "auxiliary_loss_mlp": 0.01030808,
      "balance_loss_clip": 1.05467296,
      "balance_loss_mlp": 1.0229342,
      "epoch": 0.699212409066314,
      "flos": 20193719725440.0,
      "grad_norm": 1.813664870936869,
      "language_loss": 0.74153996,
      "learning_rate": 8.76181176208602e-07,
      "loss": 0.76364219,
      "num_input_tokens_seen": 125152555,
      "step": 5815,
      "time_per_iteration": 2.609572410583496
    },
    {
      "auxiliary_loss_clip": 0.0117872,
      "auxiliary_loss_mlp": 0.01030168,
      "balance_loss_clip": 1.05324626,
      "balance_loss_mlp": 1.02164435,
      "epoch": 0.699332651956953,
      "flos": 19427888828160.0,
      "grad_norm": 2.321427473849262,
      "language_loss": 0.73591888,
      "learning_rate": 8.755368970739461e-07,
      "loss": 0.75800782,
      "num_input_tokens_seen": 125171915,
      "step": 5816,
      "time_per_iteration": 3.585986375808716
    },
    {
      "auxiliary_loss_clip": 0.01161234,
      "auxiliary_loss_mlp": 0.01030519,
      "balance_loss_clip": 1.05008316,
      "balance_loss_mlp": 1.02175117,
      "epoch": 0.6994528948475921,
      "flos": 16143714466560.0,
      "grad_norm": 2.113026736766962,
      "language_loss": 0.61828959,
      "learning_rate": 8.748927885194479e-07,
      "loss": 0.64020705,
      "num_input_tokens_seen": 125190220,
      "step": 5817,
      "time_per_iteration": 2.6978399753570557
    },
    {
      "auxiliary_loss_clip": 0.01034488,
      "auxiliary_loss_mlp": 0.01004766,
      "balance_loss_clip": 1.0115782,
      "balance_loss_mlp": 1.00375843,
      "epoch": 0.6995731377382313,
      "flos": 64952420699520.0,
      "grad_norm": 0.7909133440984383,
      "language_loss": 0.57303536,
      "learning_rate": 8.742488506428209e-07,
      "loss": 0.5934279,
      "num_input_tokens_seen": 125249310,
      "step": 5818,
      "time_per_iteration": 4.145409345626831
    },
    {
      "auxiliary_loss_clip": 0.01158874,
      "auxiliary_loss_mlp": 0.01022317,
      "balance_loss_clip": 1.05165994,
      "balance_loss_mlp": 1.01422822,
      "epoch": 0.6996933806288703,
      "flos": 24900136076160.0,
      "grad_norm": 2.100100370583147,
      "language_loss": 0.78311974,
      "learning_rate": 8.736050835417466e-07,
      "loss": 0.80493164,
      "num_input_tokens_seen": 125269350,
      "step": 5819,
      "time_per_iteration": 3.688305377960205
    },
    {
      "auxiliary_loss_clip": 0.01134463,
      "auxiliary_loss_mlp": 0.01028003,
      "balance_loss_clip": 1.04582489,
      "balance_loss_mlp": 1.01995015,
      "epoch": 0.6998136235195094,
      "flos": 20777806782720.0,
      "grad_norm": 2.0693200750026777,
      "language_loss": 0.61684257,
      "learning_rate": 8.729614873138862e-07,
      "loss": 0.63846725,
      "num_input_tokens_seen": 125286985,
      "step": 5820,
      "time_per_iteration": 2.730644464492798
    },
    {
      "auxiliary_loss_clip": 0.01168448,
      "auxiliary_loss_mlp": 0.01035132,
      "balance_loss_clip": 1.05159068,
      "balance_loss_mlp": 1.02653742,
      "epoch": 0.6999338664101485,
      "flos": 23733470332800.0,
      "grad_norm": 3.439029911727737,
      "language_loss": 0.77984667,
      "learning_rate": 8.723180620568716e-07,
      "loss": 0.80188251,
      "num_input_tokens_seen": 125306240,
      "step": 5821,
      "time_per_iteration": 2.649409532546997
    },
    {
      "auxiliary_loss_clip": 0.01163972,
      "auxiliary_loss_mlp": 0.01027842,
      "balance_loss_clip": 1.05086172,
      "balance_loss_mlp": 1.02026033,
      "epoch": 0.7000541093007876,
      "flos": 19864598382720.0,
      "grad_norm": 1.9932773923769926,
      "language_loss": 0.85153854,
      "learning_rate": 8.716748078683116e-07,
      "loss": 0.8734566,
      "num_input_tokens_seen": 125323015,
      "step": 5822,
      "time_per_iteration": 2.701429605484009
    },
    {
      "auxiliary_loss_clip": 0.01169422,
      "auxiliary_loss_mlp": 0.01034462,
      "balance_loss_clip": 1.05198765,
      "balance_loss_mlp": 1.02614689,
      "epoch": 0.7001743521914267,
      "flos": 29679056029440.0,
      "grad_norm": 8.048433823199476,
      "language_loss": 0.68726772,
      "learning_rate": 8.710317248457855e-07,
      "loss": 0.7093066,
      "num_input_tokens_seen": 125342630,
      "step": 5823,
      "time_per_iteration": 2.6828134059906006
    },
    {
      "auxiliary_loss_clip": 0.01156815,
      "auxiliary_loss_mlp": 0.01025501,
      "balance_loss_clip": 1.05117941,
      "balance_loss_mlp": 1.01715004,
      "epoch": 0.7002945950820658,
      "flos": 27489762080640.0,
      "grad_norm": 1.6424364482532925,
      "language_loss": 0.72191054,
      "learning_rate": 8.703888130868482e-07,
      "loss": 0.74373376,
      "num_input_tokens_seen": 125364480,
      "step": 5824,
      "time_per_iteration": 2.7643072605133057
    },
    {
      "auxiliary_loss_clip": 0.01176137,
      "auxiliary_loss_mlp": 0.01029395,
      "balance_loss_clip": 1.05423105,
      "balance_loss_mlp": 1.02134299,
      "epoch": 0.7004148379727049,
      "flos": 22158463800960.0,
      "grad_norm": 2.0392308063048126,
      "language_loss": 0.82417333,
      "learning_rate": 8.697460726890307e-07,
      "loss": 0.8462286,
      "num_input_tokens_seen": 125381625,
      "step": 5825,
      "time_per_iteration": 2.542341947555542
    },
    {
      "auxiliary_loss_clip": 0.01140207,
      "auxiliary_loss_mlp": 0.01025522,
      "balance_loss_clip": 1.04664707,
      "balance_loss_mlp": 1.01715338,
      "epoch": 0.7005350808633439,
      "flos": 19423758764160.0,
      "grad_norm": 2.163479253409252,
      "language_loss": 0.90345544,
      "learning_rate": 8.691035037498354e-07,
      "loss": 0.92511272,
      "num_input_tokens_seen": 125397615,
      "step": 5826,
      "time_per_iteration": 2.740231990814209
    },
    {
      "auxiliary_loss_clip": 0.01175221,
      "auxiliary_loss_mlp": 0.01025758,
      "balance_loss_clip": 1.05250287,
      "balance_loss_mlp": 1.01808083,
      "epoch": 0.7006553237539831,
      "flos": 23476708938240.0,
      "grad_norm": 1.6824257148577855,
      "language_loss": 0.72546083,
      "learning_rate": 8.684611063667391e-07,
      "loss": 0.74747062,
      "num_input_tokens_seen": 125418080,
      "step": 5827,
      "time_per_iteration": 2.605900287628174
    },
    {
      "auxiliary_loss_clip": 0.01178814,
      "auxiliary_loss_mlp": 0.01023977,
      "balance_loss_clip": 1.05432403,
      "balance_loss_mlp": 1.0162524,
      "epoch": 0.7007755666446221,
      "flos": 31212872640000.0,
      "grad_norm": 4.858775921608301,
      "language_loss": 0.77152056,
      "learning_rate": 8.678188806371935e-07,
      "loss": 0.79354846,
      "num_input_tokens_seen": 125440115,
      "step": 5828,
      "time_per_iteration": 3.7124335765838623
    },
    {
      "auxiliary_loss_clip": 0.01161832,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.05047727,
      "balance_loss_mlp": 1.01942217,
      "epoch": 0.7008958095352612,
      "flos": 18149899858560.0,
      "grad_norm": 2.062356350650976,
      "language_loss": 0.85629976,
      "learning_rate": 8.671768266586228e-07,
      "loss": 0.87818712,
      "num_input_tokens_seen": 125458240,
      "step": 5829,
      "time_per_iteration": 2.6817381381988525
    },
    {
      "auxiliary_loss_clip": 0.01168763,
      "auxiliary_loss_mlp": 0.01023541,
      "balance_loss_clip": 1.05340004,
      "balance_loss_mlp": 1.01541686,
      "epoch": 0.7010160524259004,
      "flos": 27452307173760.0,
      "grad_norm": 2.077008270223516,
      "language_loss": 0.78283077,
      "learning_rate": 8.665349445284275e-07,
      "loss": 0.80475378,
      "num_input_tokens_seen": 125477980,
      "step": 5830,
      "time_per_iteration": 2.7064015865325928
    },
    {
      "auxiliary_loss_clip": 0.01147536,
      "auxiliary_loss_mlp": 0.01027564,
      "balance_loss_clip": 1.04846513,
      "balance_loss_mlp": 1.01960361,
      "epoch": 0.7011362953165394,
      "flos": 23842064125440.0,
      "grad_norm": 1.505931676185916,
      "language_loss": 0.80937821,
      "learning_rate": 8.658932343439799e-07,
      "loss": 0.83112919,
      "num_input_tokens_seen": 125497765,
      "step": 5831,
      "time_per_iteration": 2.748213768005371
    },
    {
      "auxiliary_loss_clip": 0.01177319,
      "auxiliary_loss_mlp": 0.01031193,
      "balance_loss_clip": 1.05356598,
      "balance_loss_mlp": 1.02288413,
      "epoch": 0.7012565382071785,
      "flos": 24823430582400.0,
      "grad_norm": 2.158184649059573,
      "language_loss": 0.77842039,
      "learning_rate": 8.65251696202627e-07,
      "loss": 0.80050552,
      "num_input_tokens_seen": 125514145,
      "step": 5832,
      "time_per_iteration": 2.667508840560913
    },
    {
      "auxiliary_loss_clip": 0.01123037,
      "auxiliary_loss_mlp": 0.01033667,
      "balance_loss_clip": 1.04105306,
      "balance_loss_mlp": 1.02517366,
      "epoch": 0.7013767810978175,
      "flos": 21397445326080.0,
      "grad_norm": 2.1840525968082067,
      "language_loss": 0.87823176,
      "learning_rate": 8.646103302016896e-07,
      "loss": 0.89979881,
      "num_input_tokens_seen": 125533115,
      "step": 5833,
      "time_per_iteration": 2.8297934532165527
    },
    {
      "auxiliary_loss_clip": 0.01167939,
      "auxiliary_loss_mlp": 0.010314,
      "balance_loss_clip": 1.05075407,
      "balance_loss_mlp": 1.02295399,
      "epoch": 0.7014970239884567,
      "flos": 16687150306560.0,
      "grad_norm": 1.872660358476087,
      "language_loss": 0.88691527,
      "learning_rate": 8.639691364384614e-07,
      "loss": 0.90890861,
      "num_input_tokens_seen": 125550740,
      "step": 5834,
      "time_per_iteration": 2.6225979328155518
    },
    {
      "auxiliary_loss_clip": 0.01176715,
      "auxiliary_loss_mlp": 0.0102436,
      "balance_loss_clip": 1.05107105,
      "balance_loss_mlp": 1.01638162,
      "epoch": 0.7016172668790958,
      "flos": 12568268718720.0,
      "grad_norm": 4.418209739655172,
      "language_loss": 0.73094594,
      "learning_rate": 8.633281150102136e-07,
      "loss": 0.75295663,
      "num_input_tokens_seen": 125567590,
      "step": 5835,
      "time_per_iteration": 2.628688097000122
    },
    {
      "auxiliary_loss_clip": 0.01160978,
      "auxiliary_loss_mlp": 0.01025985,
      "balance_loss_clip": 1.05064583,
      "balance_loss_mlp": 1.01838505,
      "epoch": 0.7017375097697348,
      "flos": 17452729808640.0,
      "grad_norm": 2.454610515968616,
      "language_loss": 0.6838665,
      "learning_rate": 8.626872660141855e-07,
      "loss": 0.70573616,
      "num_input_tokens_seen": 125585500,
      "step": 5836,
      "time_per_iteration": 2.6504836082458496
    },
    {
      "auxiliary_loss_clip": 0.01139312,
      "auxiliary_loss_mlp": 0.01026923,
      "balance_loss_clip": 1.04774892,
      "balance_loss_mlp": 1.01912057,
      "epoch": 0.701857752660374,
      "flos": 18513028402560.0,
      "grad_norm": 1.6394392587469389,
      "language_loss": 0.75122255,
      "learning_rate": 8.620465895475957e-07,
      "loss": 0.77288485,
      "num_input_tokens_seen": 125603720,
      "step": 5837,
      "time_per_iteration": 2.643099546432495
    },
    {
      "auxiliary_loss_clip": 0.01160984,
      "auxiliary_loss_mlp": 0.01031993,
      "balance_loss_clip": 1.05349803,
      "balance_loss_mlp": 1.0244174,
      "epoch": 0.701977995551013,
      "flos": 24425971614720.0,
      "grad_norm": 2.1054785190297363,
      "language_loss": 0.75663996,
      "learning_rate": 8.614060857076333e-07,
      "loss": 0.7785697,
      "num_input_tokens_seen": 125624390,
      "step": 5838,
      "time_per_iteration": 2.6738245487213135
    },
    {
      "auxiliary_loss_clip": 0.01177533,
      "auxiliary_loss_mlp": 0.01026393,
      "balance_loss_clip": 1.05329633,
      "balance_loss_mlp": 1.01884127,
      "epoch": 0.7020982384416521,
      "flos": 23002759958400.0,
      "grad_norm": 1.9744255060236882,
      "language_loss": 0.7527473,
      "learning_rate": 8.60765754591462e-07,
      "loss": 0.77478659,
      "num_input_tokens_seen": 125644085,
      "step": 5839,
      "time_per_iteration": 2.677264451980591
    },
    {
      "auxiliary_loss_clip": 0.01146966,
      "auxiliary_loss_mlp": 0.010336,
      "balance_loss_clip": 1.04954088,
      "balance_loss_mlp": 1.02562463,
      "epoch": 0.7022184813322913,
      "flos": 20449080489600.0,
      "grad_norm": 2.2395807658785745,
      "language_loss": 0.72908914,
      "learning_rate": 8.601255962962211e-07,
      "loss": 0.75089478,
      "num_input_tokens_seen": 125663095,
      "step": 5840,
      "time_per_iteration": 2.7654597759246826
    },
    {
      "auxiliary_loss_clip": 0.01162696,
      "auxiliary_loss_mlp": 0.01037528,
      "balance_loss_clip": 1.05042219,
      "balance_loss_mlp": 1.02773464,
      "epoch": 0.7023387242229303,
      "flos": 19790514581760.0,
      "grad_norm": 2.416949224767392,
      "language_loss": 0.72579324,
      "learning_rate": 8.594856109190194e-07,
      "loss": 0.74779546,
      "num_input_tokens_seen": 125680125,
      "step": 5841,
      "time_per_iteration": 2.618622064590454
    },
    {
      "auxiliary_loss_clip": 0.0117622,
      "auxiliary_loss_mlp": 0.01024396,
      "balance_loss_clip": 1.05215192,
      "balance_loss_mlp": 1.01672816,
      "epoch": 0.7024589671135694,
      "flos": 33259278286080.0,
      "grad_norm": 2.6289212127317203,
      "language_loss": 0.69406164,
      "learning_rate": 8.588457985569446e-07,
      "loss": 0.71606779,
      "num_input_tokens_seen": 125703035,
      "step": 5842,
      "time_per_iteration": 3.620039701461792
    },
    {
      "auxiliary_loss_clip": 0.01167631,
      "auxiliary_loss_mlp": 0.01030667,
      "balance_loss_clip": 1.05265319,
      "balance_loss_mlp": 1.0228529,
      "epoch": 0.7025792100042085,
      "flos": 19098982967040.0,
      "grad_norm": 2.2176885422674326,
      "language_loss": 0.72043014,
      "learning_rate": 8.582061593070542e-07,
      "loss": 0.7424131,
      "num_input_tokens_seen": 125723765,
      "step": 5843,
      "time_per_iteration": 2.53833270072937
    },
    {
      "auxiliary_loss_clip": 0.01130202,
      "auxiliary_loss_mlp": 0.010279,
      "balance_loss_clip": 1.04943359,
      "balance_loss_mlp": 1.01921618,
      "epoch": 0.7026994528948476,
      "flos": 18952611045120.0,
      "grad_norm": 2.6284301016270515,
      "language_loss": 0.77013135,
      "learning_rate": 8.57566693266383e-07,
      "loss": 0.79171228,
      "num_input_tokens_seen": 125741455,
      "step": 5844,
      "time_per_iteration": 3.4775888919830322
    },
    {
      "auxiliary_loss_clip": 0.01177604,
      "auxiliary_loss_mlp": 0.01029826,
      "balance_loss_clip": 1.05238557,
      "balance_loss_mlp": 1.02121305,
      "epoch": 0.7028196957854866,
      "flos": 19536662188800.0,
      "grad_norm": 2.349595864254986,
      "language_loss": 0.69835305,
      "learning_rate": 8.569274005319354e-07,
      "loss": 0.72042739,
      "num_input_tokens_seen": 125759855,
      "step": 5845,
      "time_per_iteration": 2.5070059299468994
    },
    {
      "auxiliary_loss_clip": 0.01161203,
      "auxiliary_loss_mlp": 0.01029401,
      "balance_loss_clip": 1.05267859,
      "balance_loss_mlp": 1.0208056,
      "epoch": 0.7029399386761258,
      "flos": 20845318394880.0,
      "grad_norm": 1.9557923214286264,
      "language_loss": 0.79822737,
      "learning_rate": 8.562882812006913e-07,
      "loss": 0.82013339,
      "num_input_tokens_seen": 125777345,
      "step": 5846,
      "time_per_iteration": 3.436566114425659
    },
    {
      "auxiliary_loss_clip": 0.01176168,
      "auxiliary_loss_mlp": 0.0103153,
      "balance_loss_clip": 1.05248237,
      "balance_loss_mlp": 1.02405,
      "epoch": 0.7030601815667649,
      "flos": 22055005653120.0,
      "grad_norm": 23.92587094879223,
      "language_loss": 0.77997637,
      "learning_rate": 8.556493353696066e-07,
      "loss": 0.80205333,
      "num_input_tokens_seen": 125796345,
      "step": 5847,
      "time_per_iteration": 2.6132566928863525
    },
    {
      "auxiliary_loss_clip": 0.01179584,
      "auxiliary_loss_mlp": 0.01028716,
      "balance_loss_clip": 1.05363011,
      "balance_loss_mlp": 1.02034187,
      "epoch": 0.7031804244574039,
      "flos": 27198742089600.0,
      "grad_norm": 2.2015797798173344,
      "language_loss": 0.68098533,
      "learning_rate": 8.550105631356077e-07,
      "loss": 0.70306838,
      "num_input_tokens_seen": 125816070,
      "step": 5848,
      "time_per_iteration": 2.6982293128967285
    },
    {
      "auxiliary_loss_clip": 0.0115193,
      "auxiliary_loss_mlp": 0.01025899,
      "balance_loss_clip": 1.04962611,
      "balance_loss_mlp": 1.01797771,
      "epoch": 0.7033006673480431,
      "flos": 22379853277440.0,
      "grad_norm": 2.1222468883250505,
      "language_loss": 0.77196997,
      "learning_rate": 8.543719645955961e-07,
      "loss": 0.79374826,
      "num_input_tokens_seen": 125834400,
      "step": 5849,
      "time_per_iteration": 2.8955600261688232
    },
    {
      "auxiliary_loss_clip": 0.01166072,
      "auxiliary_loss_mlp": 0.01025963,
      "balance_loss_clip": 1.04976368,
      "balance_loss_mlp": 1.0178864,
      "epoch": 0.7034209102386821,
      "flos": 24715986024960.0,
      "grad_norm": 1.6645135002780236,
      "language_loss": 0.74525499,
      "learning_rate": 8.537335398464467e-07,
      "loss": 0.76717532,
      "num_input_tokens_seen": 125854720,
      "step": 5850,
      "time_per_iteration": 2.738523483276367
    },
    {
      "auxiliary_loss_clip": 0.01180593,
      "auxiliary_loss_mlp": 0.01026668,
      "balance_loss_clip": 1.05365014,
      "balance_loss_mlp": 1.01807332,
      "epoch": 0.7035411531293212,
      "flos": 22556174163840.0,
      "grad_norm": 3.0531435824079396,
      "language_loss": 0.85243917,
      "learning_rate": 8.53095288985007e-07,
      "loss": 0.87451172,
      "num_input_tokens_seen": 125868455,
      "step": 5851,
      "time_per_iteration": 2.5758652687072754
    },
    {
      "auxiliary_loss_clip": 0.01130979,
      "auxiliary_loss_mlp": 0.01024272,
      "balance_loss_clip": 1.0486114,
      "balance_loss_mlp": 1.01722062,
      "epoch": 0.7036613960199604,
      "flos": 22674967418880.0,
      "grad_norm": 1.716376151379196,
      "language_loss": 0.82420051,
      "learning_rate": 8.524572121081009e-07,
      "loss": 0.84575301,
      "num_input_tokens_seen": 125888555,
      "step": 5852,
      "time_per_iteration": 2.7484896183013916
    },
    {
      "auxiliary_loss_clip": 0.01179238,
      "auxiliary_loss_mlp": 0.01030529,
      "balance_loss_clip": 1.05372989,
      "balance_loss_mlp": 1.0218389,
      "epoch": 0.7037816389105994,
      "flos": 22492146170880.0,
      "grad_norm": 2.052855782534073,
      "language_loss": 0.62061775,
      "learning_rate": 8.518193093125232e-07,
      "loss": 0.64271539,
      "num_input_tokens_seen": 125907610,
      "step": 5853,
      "time_per_iteration": 2.5995304584503174
    },
    {
      "auxiliary_loss_clip": 0.01153123,
      "auxiliary_loss_mlp": 0.01029827,
      "balance_loss_clip": 1.04773629,
      "balance_loss_mlp": 1.02175033,
      "epoch": 0.7039018818012385,
      "flos": 27087490690560.0,
      "grad_norm": 1.8296447158891223,
      "language_loss": 0.80957866,
      "learning_rate": 8.511815806950436e-07,
      "loss": 0.8314082,
      "num_input_tokens_seen": 125928640,
      "step": 5854,
      "time_per_iteration": 2.699218273162842
    },
    {
      "auxiliary_loss_clip": 0.01141888,
      "auxiliary_loss_mlp": 0.01022106,
      "balance_loss_clip": 1.04634607,
      "balance_loss_mlp": 1.01457763,
      "epoch": 0.7040221246918776,
      "flos": 17749819198080.0,
      "grad_norm": 1.7864852878027242,
      "language_loss": 0.77937162,
      "learning_rate": 8.505440263524044e-07,
      "loss": 0.80101156,
      "num_input_tokens_seen": 125947485,
      "step": 5855,
      "time_per_iteration": 3.692690372467041
    },
    {
      "auxiliary_loss_clip": 0.01181047,
      "auxiliary_loss_mlp": 0.01030653,
      "balance_loss_clip": 1.05363083,
      "balance_loss_mlp": 1.02138472,
      "epoch": 0.7041423675825167,
      "flos": 16279851012480.0,
      "grad_norm": 2.6226200360536542,
      "language_loss": 0.8765741,
      "learning_rate": 8.49906646381322e-07,
      "loss": 0.89869112,
      "num_input_tokens_seen": 125960320,
      "step": 5856,
      "time_per_iteration": 2.6450376510620117
    },
    {
      "auxiliary_loss_clip": 0.01174693,
      "auxiliary_loss_mlp": 0.01027449,
      "balance_loss_clip": 1.05253971,
      "balance_loss_mlp": 1.01957572,
      "epoch": 0.7042626104731557,
      "flos": 25483181639040.0,
      "grad_norm": 1.9011314503040386,
      "language_loss": 0.7238546,
      "learning_rate": 8.492694408784884e-07,
      "loss": 0.74587607,
      "num_input_tokens_seen": 125980575,
      "step": 5857,
      "time_per_iteration": 2.588945150375366
    },
    {
      "auxiliary_loss_clip": 0.01177044,
      "auxiliary_loss_mlp": 0.01031234,
      "balance_loss_clip": 1.05366325,
      "balance_loss_mlp": 1.02343225,
      "epoch": 0.7043828533637949,
      "flos": 17857622891520.0,
      "grad_norm": 2.2902011458297045,
      "language_loss": 0.62821198,
      "learning_rate": 8.486324099405642e-07,
      "loss": 0.65029478,
      "num_input_tokens_seen": 125997420,
      "step": 5858,
      "time_per_iteration": 2.6441006660461426
    },
    {
      "auxiliary_loss_clip": 0.01174767,
      "auxiliary_loss_mlp": 0.01027205,
      "balance_loss_clip": 1.05147302,
      "balance_loss_mlp": 1.02005267,
      "epoch": 0.704503096254434,
      "flos": 29494259533440.0,
      "grad_norm": 2.3695773485987357,
      "language_loss": 0.75166237,
      "learning_rate": 8.479955536641887e-07,
      "loss": 0.77368212,
      "num_input_tokens_seen": 126018915,
      "step": 5859,
      "time_per_iteration": 2.6787028312683105
    },
    {
      "auxiliary_loss_clip": 0.01162216,
      "auxiliary_loss_mlp": 0.01028782,
      "balance_loss_clip": 1.04850304,
      "balance_loss_mlp": 1.0214777,
      "epoch": 0.704623339145073,
      "flos": 30920739327360.0,
      "grad_norm": 1.9888745347338748,
      "language_loss": 0.66362441,
      "learning_rate": 8.473588721459716e-07,
      "loss": 0.68553442,
      "num_input_tokens_seen": 126038825,
      "step": 5860,
      "time_per_iteration": 2.7460732460021973
    },
    {
      "auxiliary_loss_clip": 0.0113603,
      "auxiliary_loss_mlp": 0.01032057,
      "balance_loss_clip": 1.0514406,
      "balance_loss_mlp": 1.02293134,
      "epoch": 0.7047435820357122,
      "flos": 23914747296000.0,
      "grad_norm": 2.3474913337810315,
      "language_loss": 0.70922881,
      "learning_rate": 8.467223654824967e-07,
      "loss": 0.73090971,
      "num_input_tokens_seen": 126058280,
      "step": 5861,
      "time_per_iteration": 2.804276704788208
    },
    {
      "auxiliary_loss_clip": 0.01148485,
      "auxiliary_loss_mlp": 0.01028406,
      "balance_loss_clip": 1.05130053,
      "balance_loss_mlp": 1.02022815,
      "epoch": 0.7048638249263512,
      "flos": 46494010926720.0,
      "grad_norm": 2.901209552815667,
      "language_loss": 0.62512577,
      "learning_rate": 8.460860337703233e-07,
      "loss": 0.64689469,
      "num_input_tokens_seen": 126078885,
      "step": 5862,
      "time_per_iteration": 2.824909210205078
    },
    {
      "auxiliary_loss_clip": 0.01183085,
      "auxiliary_loss_mlp": 0.01029434,
      "balance_loss_clip": 1.05356836,
      "balance_loss_mlp": 1.02022505,
      "epoch": 0.7049840678169903,
      "flos": 21689219502720.0,
      "grad_norm": 2.269409268506623,
      "language_loss": 0.70451641,
      "learning_rate": 8.454498771059797e-07,
      "loss": 0.7266416,
      "num_input_tokens_seen": 126098260,
      "step": 5863,
      "time_per_iteration": 2.6721389293670654
    },
    {
      "auxiliary_loss_clip": 0.01180184,
      "auxiliary_loss_mlp": 0.01029709,
      "balance_loss_clip": 1.05278587,
      "balance_loss_mlp": 1.02171636,
      "epoch": 0.7051043107076294,
      "flos": 18405081054720.0,
      "grad_norm": 2.3555452103488284,
      "language_loss": 0.83075178,
      "learning_rate": 8.448138955859725e-07,
      "loss": 0.85285074,
      "num_input_tokens_seen": 126114845,
      "step": 5864,
      "time_per_iteration": 2.5703067779541016
    },
    {
      "auxiliary_loss_clip": 0.01148014,
      "auxiliary_loss_mlp": 0.00762379,
      "balance_loss_clip": 1.05099285,
      "balance_loss_mlp": 1.00057769,
      "epoch": 0.7052245535982685,
      "flos": 19319043640320.0,
      "grad_norm": 1.8674799712713197,
      "language_loss": 0.9031499,
      "learning_rate": 8.44178089306778e-07,
      "loss": 0.92225385,
      "num_input_tokens_seen": 126132780,
      "step": 5865,
      "time_per_iteration": 2.6811652183532715
    },
    {
      "auxiliary_loss_clip": 0.01167007,
      "auxiliary_loss_mlp": 0.01038919,
      "balance_loss_clip": 1.05273747,
      "balance_loss_mlp": 1.03145957,
      "epoch": 0.7053447964889076,
      "flos": 19062138591360.0,
      "grad_norm": 1.9225693020023797,
      "language_loss": 0.7683351,
      "learning_rate": 8.4354245836485e-07,
      "loss": 0.79039431,
      "num_input_tokens_seen": 126151225,
      "step": 5866,
      "time_per_iteration": 2.626455783843994
    },
    {
      "auxiliary_loss_clip": 0.01182969,
      "auxiliary_loss_mlp": 0.01031275,
      "balance_loss_clip": 1.05462122,
      "balance_loss_mlp": 1.02249503,
      "epoch": 0.7054650393795466,
      "flos": 27379228953600.0,
      "grad_norm": 1.5847395251758434,
      "language_loss": 0.73040688,
      "learning_rate": 8.429070028566108e-07,
      "loss": 0.75254935,
      "num_input_tokens_seen": 126172535,
      "step": 5867,
      "time_per_iteration": 2.7255423069000244
    },
    {
      "auxiliary_loss_clip": 0.01176851,
      "auxiliary_loss_mlp": 0.01024451,
      "balance_loss_clip": 1.05308068,
      "balance_loss_mlp": 1.01658916,
      "epoch": 0.7055852822701858,
      "flos": 16102201322880.0,
      "grad_norm": 2.0062539732169498,
      "language_loss": 0.75042164,
      "learning_rate": 8.422717228784586e-07,
      "loss": 0.77243471,
      "num_input_tokens_seen": 126189410,
      "step": 5868,
      "time_per_iteration": 3.588979482650757
    },
    {
      "auxiliary_loss_clip": 0.01164037,
      "auxiliary_loss_mlp": 0.01025451,
      "balance_loss_clip": 1.05400515,
      "balance_loss_mlp": 1.01744068,
      "epoch": 0.7057055251608249,
      "flos": 11692299744000.0,
      "grad_norm": 2.302144879705966,
      "language_loss": 0.69620478,
      "learning_rate": 8.416366185267663e-07,
      "loss": 0.71809959,
      "num_input_tokens_seen": 126206910,
      "step": 5869,
      "time_per_iteration": 3.5206656455993652
    },
    {
      "auxiliary_loss_clip": 0.01155465,
      "auxiliary_loss_mlp": 0.01026845,
      "balance_loss_clip": 1.04691732,
      "balance_loss_mlp": 1.01905429,
      "epoch": 0.7058257680514639,
      "flos": 22711560399360.0,
      "grad_norm": 2.929674370298565,
      "language_loss": 0.77671945,
      "learning_rate": 8.410016898978778e-07,
      "loss": 0.7985425,
      "num_input_tokens_seen": 126224385,
      "step": 5870,
      "time_per_iteration": 2.7021963596343994
    },
    {
      "auxiliary_loss_clip": 0.01159646,
      "auxiliary_loss_mlp": 0.01027488,
      "balance_loss_clip": 1.0495466,
      "balance_loss_mlp": 1.01983476,
      "epoch": 0.7059460109421031,
      "flos": 17529543043200.0,
      "grad_norm": 1.9706453761104548,
      "language_loss": 0.7938894,
      "learning_rate": 8.403669370881115e-07,
      "loss": 0.81576073,
      "num_input_tokens_seen": 126243120,
      "step": 5871,
      "time_per_iteration": 2.6481292247772217
    },
    {
      "auxiliary_loss_clip": 0.01153342,
      "auxiliary_loss_mlp": 0.0102552,
      "balance_loss_clip": 1.04824424,
      "balance_loss_mlp": 1.01835513,
      "epoch": 0.7060662538327421,
      "flos": 23544687427200.0,
      "grad_norm": 1.7235503932136043,
      "language_loss": 0.7856102,
      "learning_rate": 8.397323601937587e-07,
      "loss": 0.8073988,
      "num_input_tokens_seen": 126263020,
      "step": 5872,
      "time_per_iteration": 3.746919631958008
    },
    {
      "auxiliary_loss_clip": 0.01176928,
      "auxiliary_loss_mlp": 0.0103266,
      "balance_loss_clip": 1.05276251,
      "balance_loss_mlp": 1.0247978,
      "epoch": 0.7061864967233812,
      "flos": 30260736875520.0,
      "grad_norm": 2.271453821499919,
      "language_loss": 0.7732994,
      "learning_rate": 8.390979593110838e-07,
      "loss": 0.79539526,
      "num_input_tokens_seen": 126285150,
      "step": 5873,
      "time_per_iteration": 2.7126121520996094
    },
    {
      "auxiliary_loss_clip": 0.01156554,
      "auxiliary_loss_mlp": 0.01028234,
      "balance_loss_clip": 1.04855323,
      "balance_loss_mlp": 1.01976407,
      "epoch": 0.7063067396140204,
      "flos": 20701460424960.0,
      "grad_norm": 1.8528314783661446,
      "language_loss": 0.82039452,
      "learning_rate": 8.384637345363262e-07,
      "loss": 0.84224236,
      "num_input_tokens_seen": 126304340,
      "step": 5874,
      "time_per_iteration": 2.7224509716033936
    },
    {
      "auxiliary_loss_clip": 0.01173778,
      "auxiliary_loss_mlp": 0.01030925,
      "balance_loss_clip": 1.04996371,
      "balance_loss_mlp": 1.02280653,
      "epoch": 0.7064269825046594,
      "flos": 32266168081920.0,
      "grad_norm": 1.906426445082898,
      "language_loss": 0.7685461,
      "learning_rate": 8.378296859656964e-07,
      "loss": 0.79059315,
      "num_input_tokens_seen": 126325495,
      "step": 5875,
      "time_per_iteration": 2.7130632400512695
    },
    {
      "auxiliary_loss_clip": 0.01163281,
      "auxiliary_loss_mlp": 0.00761841,
      "balance_loss_clip": 1.05059993,
      "balance_loss_mlp": 1.00066233,
      "epoch": 0.7065472253952985,
      "flos": 30227124723840.0,
      "grad_norm": 2.028310254684216,
      "language_loss": 0.68420815,
      "learning_rate": 8.371958136953792e-07,
      "loss": 0.70345932,
      "num_input_tokens_seen": 126345525,
      "step": 5876,
      "time_per_iteration": 2.737495183944702
    },
    {
      "auxiliary_loss_clip": 0.01163841,
      "auxiliary_loss_mlp": 0.01032453,
      "balance_loss_clip": 1.05091143,
      "balance_loss_mlp": 1.02413869,
      "epoch": 0.7066674682859376,
      "flos": 16216720859520.0,
      "grad_norm": 16.589003385169576,
      "language_loss": 0.66387665,
      "learning_rate": 8.365621178215326e-07,
      "loss": 0.68583965,
      "num_input_tokens_seen": 126361995,
      "step": 5877,
      "time_per_iteration": 2.6853578090667725
    },
    {
      "auxiliary_loss_clip": 0.01147613,
      "auxiliary_loss_mlp": 0.01024725,
      "balance_loss_clip": 1.04949439,
      "balance_loss_mlp": 1.01721227,
      "epoch": 0.7067877111765767,
      "flos": 14830461319680.0,
      "grad_norm": 2.04160442451351,
      "language_loss": 0.75286758,
      "learning_rate": 8.359285984402871e-07,
      "loss": 0.77459097,
      "num_input_tokens_seen": 126379260,
      "step": 5878,
      "time_per_iteration": 2.623723268508911
    },
    {
      "auxiliary_loss_clip": 0.01141257,
      "auxiliary_loss_mlp": 0.01032155,
      "balance_loss_clip": 1.04708064,
      "balance_loss_mlp": 1.02439451,
      "epoch": 0.7069079540672157,
      "flos": 25440196037760.0,
      "grad_norm": 1.8456294591473976,
      "language_loss": 0.7445457,
      "learning_rate": 8.352952556477489e-07,
      "loss": 0.76627982,
      "num_input_tokens_seen": 126397170,
      "step": 5879,
      "time_per_iteration": 2.734790563583374
    },
    {
      "auxiliary_loss_clip": 0.01176194,
      "auxiliary_loss_mlp": 0.01032462,
      "balance_loss_clip": 1.05308509,
      "balance_loss_mlp": 1.02490735,
      "epoch": 0.7070281969578549,
      "flos": 24607751368320.0,
      "grad_norm": 1.8194239760324227,
      "language_loss": 0.76535976,
      "learning_rate": 8.34662089539993e-07,
      "loss": 0.78744638,
      "num_input_tokens_seen": 126416680,
      "step": 5880,
      "time_per_iteration": 2.618048906326294
    },
    {
      "auxiliary_loss_clip": 0.01174846,
      "auxiliary_loss_mlp": 0.00761532,
      "balance_loss_clip": 1.05253625,
      "balance_loss_mlp": 1.00067365,
      "epoch": 0.707148439848494,
      "flos": 26724469887360.0,
      "grad_norm": 2.8716016797972004,
      "language_loss": 0.7935819,
      "learning_rate": 8.340291002130722e-07,
      "loss": 0.81294566,
      "num_input_tokens_seen": 126435870,
      "step": 5881,
      "time_per_iteration": 3.628298044204712
    },
    {
      "auxiliary_loss_clip": 0.01129833,
      "auxiliary_loss_mlp": 0.01026407,
      "balance_loss_clip": 1.04577124,
      "balance_loss_mlp": 1.01854467,
      "epoch": 0.707268682739133,
      "flos": 15085750256640.0,
      "grad_norm": 2.705644078772536,
      "language_loss": 0.80105233,
      "learning_rate": 8.3339628776301e-07,
      "loss": 0.82261479,
      "num_input_tokens_seen": 126454010,
      "step": 5882,
      "time_per_iteration": 2.7344186305999756
    },
    {
      "auxiliary_loss_clip": 0.01173102,
      "auxiliary_loss_mlp": 0.01024424,
      "balance_loss_clip": 1.05035329,
      "balance_loss_mlp": 1.01633,
      "epoch": 0.7073889256297722,
      "flos": 34313148345600.0,
      "grad_norm": 2.1631145046436915,
      "language_loss": 0.57143295,
      "learning_rate": 8.327636522858033e-07,
      "loss": 0.59340823,
      "num_input_tokens_seen": 126473615,
      "step": 5883,
      "time_per_iteration": 2.7670538425445557
    },
    {
      "auxiliary_loss_clip": 0.0116438,
      "auxiliary_loss_mlp": 0.0103331,
      "balance_loss_clip": 1.05318117,
      "balance_loss_mlp": 1.02525735,
      "epoch": 0.7075091685204112,
      "flos": 20083940784000.0,
      "grad_norm": 2.003757751797666,
      "language_loss": 0.77122378,
      "learning_rate": 8.321311938774225e-07,
      "loss": 0.79320073,
      "num_input_tokens_seen": 126492705,
      "step": 5884,
      "time_per_iteration": 2.700756311416626
    },
    {
      "auxiliary_loss_clip": 0.01154912,
      "auxiliary_loss_mlp": 0.01028758,
      "balance_loss_clip": 1.0479269,
      "balance_loss_mlp": 1.02081227,
      "epoch": 0.7076294114110503,
      "flos": 20777124424320.0,
      "grad_norm": 1.836537002977465,
      "language_loss": 0.79178119,
      "learning_rate": 8.314989126338104e-07,
      "loss": 0.81361783,
      "num_input_tokens_seen": 126512715,
      "step": 5885,
      "time_per_iteration": 2.643430709838867
    },
    {
      "auxiliary_loss_clip": 0.01162239,
      "auxiliary_loss_mlp": 0.01025164,
      "balance_loss_clip": 1.05268323,
      "balance_loss_mlp": 1.01706076,
      "epoch": 0.7077496543016895,
      "flos": 17967689141760.0,
      "grad_norm": 1.9245458095535901,
      "language_loss": 0.84176588,
      "learning_rate": 8.308668086508847e-07,
      "loss": 0.86363989,
      "num_input_tokens_seen": 126530795,
      "step": 5886,
      "time_per_iteration": 2.6732280254364014
    },
    {
      "auxiliary_loss_clip": 0.01178474,
      "auxiliary_loss_mlp": 0.01028841,
      "balance_loss_clip": 1.05280685,
      "balance_loss_mlp": 1.01999545,
      "epoch": 0.7078698971923285,
      "flos": 45478098564480.0,
      "grad_norm": 2.0898257305282724,
      "language_loss": 0.7394532,
      "learning_rate": 8.302348820245342e-07,
      "loss": 0.76152635,
      "num_input_tokens_seen": 126553360,
      "step": 5887,
      "time_per_iteration": 2.809868574142456
    },
    {
      "auxiliary_loss_clip": 0.01167977,
      "auxiliary_loss_mlp": 0.010287,
      "balance_loss_clip": 1.05146337,
      "balance_loss_mlp": 1.02039719,
      "epoch": 0.7079901400829676,
      "flos": 26943704547840.0,
      "grad_norm": 4.103889710825421,
      "language_loss": 0.70156074,
      "learning_rate": 8.296031328506232e-07,
      "loss": 0.72352755,
      "num_input_tokens_seen": 126573110,
      "step": 5888,
      "time_per_iteration": 2.713674306869507
    },
    {
      "auxiliary_loss_clip": 0.01146814,
      "auxiliary_loss_mlp": 0.01024787,
      "balance_loss_clip": 1.05234051,
      "balance_loss_mlp": 1.01708651,
      "epoch": 0.7081103829736067,
      "flos": 24423206267520.0,
      "grad_norm": 1.8620841190093462,
      "language_loss": 0.75762999,
      "learning_rate": 8.289715612249857e-07,
      "loss": 0.77934599,
      "num_input_tokens_seen": 126593725,
      "step": 5889,
      "time_per_iteration": 2.6975295543670654
    },
    {
      "auxiliary_loss_clip": 0.01149699,
      "auxiliary_loss_mlp": 0.01024379,
      "balance_loss_clip": 1.05003881,
      "balance_loss_mlp": 1.01656747,
      "epoch": 0.7082306258642458,
      "flos": 18543300589440.0,
      "grad_norm": 2.512587634809376,
      "language_loss": 0.78355366,
      "learning_rate": 8.283401672434305e-07,
      "loss": 0.80529445,
      "num_input_tokens_seen": 126608950,
      "step": 5890,
      "time_per_iteration": 2.6855833530426025
    },
    {
      "auxiliary_loss_clip": 0.01153579,
      "auxiliary_loss_mlp": 0.0102546,
      "balance_loss_clip": 1.04907012,
      "balance_loss_mlp": 1.0174998,
      "epoch": 0.7083508687548848,
      "flos": 23477534951040.0,
      "grad_norm": 2.3864738134553463,
      "language_loss": 0.70733809,
      "learning_rate": 8.277089510017412e-07,
      "loss": 0.72912848,
      "num_input_tokens_seen": 126629755,
      "step": 5891,
      "time_per_iteration": 2.7278618812561035
    },
    {
      "auxiliary_loss_clip": 0.0114592,
      "auxiliary_loss_mlp": 0.0103138,
      "balance_loss_clip": 1.04689133,
      "balance_loss_mlp": 1.02405488,
      "epoch": 0.708471111645524,
      "flos": 22419463000320.0,
      "grad_norm": 1.7306502179296,
      "language_loss": 0.82397264,
      "learning_rate": 8.270779125956719e-07,
      "loss": 0.84574562,
      "num_input_tokens_seen": 126650135,
      "step": 5892,
      "time_per_iteration": 2.6762335300445557
    },
    {
      "auxiliary_loss_clip": 0.01162314,
      "auxiliary_loss_mlp": 0.01026291,
      "balance_loss_clip": 1.05175364,
      "balance_loss_mlp": 1.01886988,
      "epoch": 0.7085913545361631,
      "flos": 20922885815040.0,
      "grad_norm": 2.6549815855180072,
      "language_loss": 0.80172241,
      "learning_rate": 8.264470521209505e-07,
      "loss": 0.82360846,
      "num_input_tokens_seen": 126668500,
      "step": 5893,
      "time_per_iteration": 2.6917495727539062
    },
    {
      "auxiliary_loss_clip": 0.01174408,
      "auxiliary_loss_mlp": 0.01030218,
      "balance_loss_clip": 1.04997277,
      "balance_loss_mlp": 1.02169454,
      "epoch": 0.7087115974268021,
      "flos": 15012384727680.0,
      "grad_norm": 4.375318784508954,
      "language_loss": 0.76864946,
      "learning_rate": 8.258163696732785e-07,
      "loss": 0.79069573,
      "num_input_tokens_seen": 126686090,
      "step": 5894,
      "time_per_iteration": 3.5082812309265137
    },
    {
      "auxiliary_loss_clip": 0.01164852,
      "auxiliary_loss_mlp": 0.01028078,
      "balance_loss_clip": 1.0512259,
      "balance_loss_mlp": 1.02054989,
      "epoch": 0.7088318403174413,
      "flos": 21539040739200.0,
      "grad_norm": 2.1868759882380853,
      "language_loss": 0.77063239,
      "learning_rate": 8.251858653483288e-07,
      "loss": 0.79256165,
      "num_input_tokens_seen": 126704255,
      "step": 5895,
      "time_per_iteration": 3.633455276489258
    },
    {
      "auxiliary_loss_clip": 0.01163547,
      "auxiliary_loss_mlp": 0.00762005,
      "balance_loss_clip": 1.05221868,
      "balance_loss_mlp": 1.00067592,
      "epoch": 0.7089520832080803,
      "flos": 15516785462400.0,
      "grad_norm": 2.014653431653575,
      "language_loss": 0.86106849,
      "learning_rate": 8.245555392417501e-07,
      "loss": 0.88032401,
      "num_input_tokens_seen": 126718910,
      "step": 5896,
      "time_per_iteration": 2.6826324462890625
    },
    {
      "auxiliary_loss_clip": 0.01165534,
      "auxiliary_loss_mlp": 0.01029351,
      "balance_loss_clip": 1.05138159,
      "balance_loss_mlp": 1.02220428,
      "epoch": 0.7090723260987194,
      "flos": 20412667077120.0,
      "grad_norm": 2.416717214998456,
      "language_loss": 0.79286015,
      "learning_rate": 8.239253914491613e-07,
      "loss": 0.81480908,
      "num_input_tokens_seen": 126737235,
      "step": 5897,
      "time_per_iteration": 2.636178493499756
    },
    {
      "auxiliary_loss_clip": 0.01166326,
      "auxiliary_loss_mlp": 0.01027503,
      "balance_loss_clip": 1.04955244,
      "balance_loss_mlp": 1.01943874,
      "epoch": 0.7091925689893585,
      "flos": 25668337271040.0,
      "grad_norm": 1.8910242595542184,
      "language_loss": 0.75141418,
      "learning_rate": 8.232954220661556e-07,
      "loss": 0.77335244,
      "num_input_tokens_seen": 126759970,
      "step": 5898,
      "time_per_iteration": 3.7266979217529297
    },
    {
      "auxiliary_loss_clip": 0.01159964,
      "auxiliary_loss_mlp": 0.01023867,
      "balance_loss_clip": 1.05274534,
      "balance_loss_mlp": 1.01632452,
      "epoch": 0.7093128118799976,
      "flos": 24206629213440.0,
      "grad_norm": 9.896593396697996,
      "language_loss": 0.70343882,
      "learning_rate": 8.226656311882989e-07,
      "loss": 0.72527713,
      "num_input_tokens_seen": 126779280,
      "step": 5899,
      "time_per_iteration": 2.6779863834381104
    },
    {
      "auxiliary_loss_clip": 0.01158646,
      "auxiliary_loss_mlp": 0.0102529,
      "balance_loss_clip": 1.04989076,
      "balance_loss_mlp": 1.01825964,
      "epoch": 0.7094330547706367,
      "flos": 16646786398080.0,
      "grad_norm": 2.3139090521193464,
      "language_loss": 0.76732224,
      "learning_rate": 8.22036018911129e-07,
      "loss": 0.78916168,
      "num_input_tokens_seen": 126797310,
      "step": 5900,
      "time_per_iteration": 2.6724958419799805
    },
    {
      "auxiliary_loss_clip": 0.01149719,
      "auxiliary_loss_mlp": 0.01029722,
      "balance_loss_clip": 1.04851162,
      "balance_loss_mlp": 1.02135372,
      "epoch": 0.7095532976612757,
      "flos": 16283370545280.0,
      "grad_norm": 2.2646234427287326,
      "language_loss": 0.80875254,
      "learning_rate": 8.214065853301599e-07,
      "loss": 0.83054698,
      "num_input_tokens_seen": 126812840,
      "step": 5901,
      "time_per_iteration": 2.7047362327575684
    },
    {
      "auxiliary_loss_clip": 0.01066093,
      "auxiliary_loss_mlp": 0.0100516,
      "balance_loss_clip": 1.01047266,
      "balance_loss_mlp": 1.0042603,
      "epoch": 0.7096735405519149,
      "flos": 70722080559360.0,
      "grad_norm": 0.8515931842405786,
      "language_loss": 0.58166689,
      "learning_rate": 8.207773305408734e-07,
      "loss": 0.60237944,
      "num_input_tokens_seen": 126880060,
      "step": 5902,
      "time_per_iteration": 3.335688829421997
    },
    {
      "auxiliary_loss_clip": 0.01150025,
      "auxiliary_loss_mlp": 0.01031821,
      "balance_loss_clip": 1.05000412,
      "balance_loss_mlp": 1.02370906,
      "epoch": 0.709793783442554,
      "flos": 23621500661760.0,
      "grad_norm": 14.885048658123244,
      "language_loss": 0.80410957,
      "learning_rate": 8.201482546387288e-07,
      "loss": 0.82592809,
      "num_input_tokens_seen": 126899535,
      "step": 5903,
      "time_per_iteration": 2.7282793521881104
    },
    {
      "auxiliary_loss_clip": 0.01175497,
      "auxiliary_loss_mlp": 0.01028045,
      "balance_loss_clip": 1.05363071,
      "balance_loss_mlp": 1.02040982,
      "epoch": 0.709914026333193,
      "flos": 25993472204160.0,
      "grad_norm": 1.7805251487328817,
      "language_loss": 0.91618395,
      "learning_rate": 8.195193577191553e-07,
      "loss": 0.93821943,
      "num_input_tokens_seen": 126921365,
      "step": 5904,
      "time_per_iteration": 2.6785149574279785
    },
    {
      "auxiliary_loss_clip": 0.01175249,
      "auxiliary_loss_mlp": 0.01027196,
      "balance_loss_clip": 1.05182743,
      "balance_loss_mlp": 1.01952481,
      "epoch": 0.7100342692238322,
      "flos": 24861531934080.0,
      "grad_norm": 1.8587864460171648,
      "language_loss": 0.84548002,
      "learning_rate": 8.188906398775579e-07,
      "loss": 0.86750448,
      "num_input_tokens_seen": 126941910,
      "step": 5905,
      "time_per_iteration": 2.7209982872009277
    },
    {
      "auxiliary_loss_clip": 0.01166069,
      "auxiliary_loss_mlp": 0.01033015,
      "balance_loss_clip": 1.04964399,
      "balance_loss_mlp": 1.02417612,
      "epoch": 0.7101545121144712,
      "flos": 24932203943040.0,
      "grad_norm": 2.539957890988475,
      "language_loss": 0.69227278,
      "learning_rate": 8.18262101209311e-07,
      "loss": 0.71426356,
      "num_input_tokens_seen": 126961120,
      "step": 5906,
      "time_per_iteration": 2.637113571166992
    },
    {
      "auxiliary_loss_clip": 0.01174578,
      "auxiliary_loss_mlp": 0.01024913,
      "balance_loss_clip": 1.05213618,
      "balance_loss_mlp": 1.01682496,
      "epoch": 0.7102747550051103,
      "flos": 23768842250880.0,
      "grad_norm": 2.4183361065206905,
      "language_loss": 0.70105398,
      "learning_rate": 8.176337418097626e-07,
      "loss": 0.72304881,
      "num_input_tokens_seen": 126981590,
      "step": 5907,
      "time_per_iteration": 2.6698477268218994
    },
    {
      "auxiliary_loss_clip": 0.01166672,
      "auxiliary_loss_mlp": 0.01028025,
      "balance_loss_clip": 1.05157256,
      "balance_loss_mlp": 1.02055657,
      "epoch": 0.7103949978957494,
      "flos": 15303907509120.0,
      "grad_norm": 2.3618262841425373,
      "language_loss": 0.79743421,
      "learning_rate": 8.170055617742364e-07,
      "loss": 0.81938118,
      "num_input_tokens_seen": 126998870,
      "step": 5908,
      "time_per_iteration": 3.5334458351135254
    },
    {
      "auxiliary_loss_clip": 0.01159734,
      "auxiliary_loss_mlp": 0.01024728,
      "balance_loss_clip": 1.04888034,
      "balance_loss_mlp": 1.0169493,
      "epoch": 0.7105152407863885,
      "flos": 22638805401600.0,
      "grad_norm": 1.8139117019846793,
      "language_loss": 0.70748019,
      "learning_rate": 8.163775611980252e-07,
      "loss": 0.72932476,
      "num_input_tokens_seen": 127017980,
      "step": 5909,
      "time_per_iteration": 2.6568968296051025
    },
    {
      "auxiliary_loss_clip": 0.0116342,
      "auxiliary_loss_mlp": 0.01034849,
      "balance_loss_clip": 1.05061579,
      "balance_loss_mlp": 1.02667117,
      "epoch": 0.7106354836770276,
      "flos": 17238594879360.0,
      "grad_norm": 1.9452460661714996,
      "language_loss": 0.79182315,
      "learning_rate": 8.157497401763982e-07,
      "loss": 0.81380588,
      "num_input_tokens_seen": 127035645,
      "step": 5910,
      "time_per_iteration": 2.6408755779266357
    },
    {
      "auxiliary_loss_clip": 0.01159369,
      "auxiliary_loss_mlp": 0.01030616,
      "balance_loss_clip": 1.05133319,
      "balance_loss_mlp": 1.02262902,
      "epoch": 0.7107557265676667,
      "flos": 20193647898240.0,
      "grad_norm": 1.7247945028827778,
      "language_loss": 0.7798388,
      "learning_rate": 8.151220988045935e-07,
      "loss": 0.80173868,
      "num_input_tokens_seen": 127054900,
      "step": 5911,
      "time_per_iteration": 2.627246141433716
    },
    {
      "auxiliary_loss_clip": 0.01159816,
      "auxiliary_loss_mlp": 0.01027398,
      "balance_loss_clip": 1.04883444,
      "balance_loss_mlp": 1.01992369,
      "epoch": 0.7108759694583058,
      "flos": 21507080613120.0,
      "grad_norm": 2.846856271520589,
      "language_loss": 0.83035922,
      "learning_rate": 8.144946371778234e-07,
      "loss": 0.85223138,
      "num_input_tokens_seen": 127075010,
      "step": 5912,
      "time_per_iteration": 2.704153060913086
    },
    {
      "auxiliary_loss_clip": 0.01161487,
      "auxiliary_loss_mlp": 0.01023084,
      "balance_loss_clip": 1.053128,
      "balance_loss_mlp": 1.01518381,
      "epoch": 0.7109962123489448,
      "flos": 24061909317120.0,
      "grad_norm": 4.82029927110014,
      "language_loss": 0.78381956,
      "learning_rate": 8.138673553912751e-07,
      "loss": 0.80566531,
      "num_input_tokens_seen": 127095570,
      "step": 5913,
      "time_per_iteration": 2.6583025455474854
    },
    {
      "auxiliary_loss_clip": 0.01151982,
      "auxiliary_loss_mlp": 0.01023927,
      "balance_loss_clip": 1.05189228,
      "balance_loss_mlp": 1.01561236,
      "epoch": 0.711116455239584,
      "flos": 30480474326400.0,
      "grad_norm": 2.4716975285187246,
      "language_loss": 0.56590676,
      "learning_rate": 8.132402535401059e-07,
      "loss": 0.5876658,
      "num_input_tokens_seen": 127116825,
      "step": 5914,
      "time_per_iteration": 2.7788474559783936
    },
    {
      "auxiliary_loss_clip": 0.01135569,
      "auxiliary_loss_mlp": 0.01028624,
      "balance_loss_clip": 1.04509926,
      "balance_loss_mlp": 1.02070212,
      "epoch": 0.711236698130223,
      "flos": 25045610158080.0,
      "grad_norm": 2.030086799443954,
      "language_loss": 0.74424469,
      "learning_rate": 8.126133317194465e-07,
      "loss": 0.76588655,
      "num_input_tokens_seen": 127137015,
      "step": 5915,
      "time_per_iteration": 2.6902756690979004
    },
    {
      "auxiliary_loss_clip": 0.01166792,
      "auxiliary_loss_mlp": 0.0102483,
      "balance_loss_clip": 1.05074692,
      "balance_loss_mlp": 1.0165391,
      "epoch": 0.7113569410208621,
      "flos": 24206701040640.0,
      "grad_norm": 3.9101027859623207,
      "language_loss": 0.74840391,
      "learning_rate": 8.11986590024401e-07,
      "loss": 0.77032018,
      "num_input_tokens_seen": 127156755,
      "step": 5916,
      "time_per_iteration": 2.737870693206787
    },
    {
      "auxiliary_loss_clip": 0.01143611,
      "auxiliary_loss_mlp": 0.01027758,
      "balance_loss_clip": 1.04802585,
      "balance_loss_mlp": 1.01903176,
      "epoch": 0.7114771839115013,
      "flos": 35439306526080.0,
      "grad_norm": 2.0310975583660564,
      "language_loss": 0.68949056,
      "learning_rate": 8.113600285500442e-07,
      "loss": 0.71120423,
      "num_input_tokens_seen": 127176965,
      "step": 5917,
      "time_per_iteration": 2.7896878719329834
    },
    {
      "auxiliary_loss_clip": 0.01175358,
      "auxiliary_loss_mlp": 0.01026763,
      "balance_loss_clip": 1.05095518,
      "balance_loss_mlp": 1.01915121,
      "epoch": 0.7115974268021403,
      "flos": 21099458096640.0,
      "grad_norm": 1.8009314893696078,
      "language_loss": 0.74772125,
      "learning_rate": 8.107336473914268e-07,
      "loss": 0.76974249,
      "num_input_tokens_seen": 127195595,
      "step": 5918,
      "time_per_iteration": 2.6169395446777344
    },
    {
      "auxiliary_loss_clip": 0.01065736,
      "auxiliary_loss_mlp": 0.01002147,
      "balance_loss_clip": 1.0101068,
      "balance_loss_mlp": 1.00124729,
      "epoch": 0.7117176696927794,
      "flos": 56752866616320.0,
      "grad_norm": 0.7677068967376143,
      "language_loss": 0.55679631,
      "learning_rate": 8.101074466435694e-07,
      "loss": 0.57747513,
      "num_input_tokens_seen": 127255070,
      "step": 5919,
      "time_per_iteration": 3.17972993850708
    },
    {
      "auxiliary_loss_clip": 0.01162569,
      "auxiliary_loss_mlp": 0.01029635,
      "balance_loss_clip": 1.05474901,
      "balance_loss_mlp": 1.02149928,
      "epoch": 0.7118379125834186,
      "flos": 15925269905280.0,
      "grad_norm": 1.9277727035427883,
      "language_loss": 0.67815822,
      "learning_rate": 8.094814264014662e-07,
      "loss": 0.70008022,
      "num_input_tokens_seen": 127273825,
      "step": 5920,
      "time_per_iteration": 3.6154286861419678
    },
    {
      "auxiliary_loss_clip": 0.01074758,
      "auxiliary_loss_mlp": 0.01031618,
      "balance_loss_clip": 1.04156184,
      "balance_loss_mlp": 1.02304053,
      "epoch": 0.7119581554740576,
      "flos": 20193360589440.0,
      "grad_norm": 7.800544822313216,
      "language_loss": 0.81693339,
      "learning_rate": 8.088555867600844e-07,
      "loss": 0.8379972,
      "num_input_tokens_seen": 127289990,
      "step": 5921,
      "time_per_iteration": 3.8201045989990234
    },
    {
      "auxiliary_loss_clip": 0.0116163,
      "auxiliary_loss_mlp": 0.01027016,
      "balance_loss_clip": 1.0521903,
      "balance_loss_mlp": 1.01958299,
      "epoch": 0.7120783983646967,
      "flos": 34715383822080.0,
      "grad_norm": 1.9451886970141659,
      "language_loss": 0.60628843,
      "learning_rate": 8.08229927814362e-07,
      "loss": 0.6281749,
      "num_input_tokens_seen": 127312880,
      "step": 5922,
      "time_per_iteration": 2.989457845687866
    },
    {
      "auxiliary_loss_clip": 0.01173233,
      "auxiliary_loss_mlp": 0.01026816,
      "balance_loss_clip": 1.05171597,
      "balance_loss_mlp": 1.01885259,
      "epoch": 0.7121986412553358,
      "flos": 26359114700160.0,
      "grad_norm": 1.9239608563173725,
      "language_loss": 0.65149313,
      "learning_rate": 8.076044496592134e-07,
      "loss": 0.67349362,
      "num_input_tokens_seen": 127334730,
      "step": 5923,
      "time_per_iteration": 2.6839725971221924
    },
    {
      "auxiliary_loss_clip": 0.0115506,
      "auxiliary_loss_mlp": 0.01033496,
      "balance_loss_clip": 1.04914689,
      "balance_loss_mlp": 1.0247041,
      "epoch": 0.7123188841459749,
      "flos": 11145344371200.0,
      "grad_norm": 2.609708682440157,
      "language_loss": 0.78057128,
      "learning_rate": 8.069791523895204e-07,
      "loss": 0.80245686,
      "num_input_tokens_seen": 127351180,
      "step": 5924,
      "time_per_iteration": 2.6749014854431152
    },
    {
      "auxiliary_loss_clip": 0.0116,
      "auxiliary_loss_mlp": 0.01025026,
      "balance_loss_clip": 1.05104887,
      "balance_loss_mlp": 1.01704216,
      "epoch": 0.7124391270366139,
      "flos": 20811670329600.0,
      "grad_norm": 2.479960189210455,
      "language_loss": 0.77566147,
      "learning_rate": 8.063540361001422e-07,
      "loss": 0.7975117,
      "num_input_tokens_seen": 127369750,
      "step": 5925,
      "time_per_iteration": 3.61000919342041
    },
    {
      "auxiliary_loss_clip": 0.01149645,
      "auxiliary_loss_mlp": 0.01030506,
      "balance_loss_clip": 1.04744494,
      "balance_loss_mlp": 1.02284074,
      "epoch": 0.7125593699272531,
      "flos": 17603734584960.0,
      "grad_norm": 2.0884095109949805,
      "language_loss": 0.79626048,
      "learning_rate": 8.057291008859069e-07,
      "loss": 0.81806195,
      "num_input_tokens_seen": 127387910,
      "step": 5926,
      "time_per_iteration": 2.7274746894836426
    },
    {
      "auxiliary_loss_clip": 0.01171927,
      "auxiliary_loss_mlp": 0.01026935,
      "balance_loss_clip": 1.05176997,
      "balance_loss_mlp": 1.01943398,
      "epoch": 0.7126796128178922,
      "flos": 28654057526400.0,
      "grad_norm": 7.724587159311088,
      "language_loss": 0.68062514,
      "learning_rate": 8.051043468416187e-07,
      "loss": 0.70261371,
      "num_input_tokens_seen": 127409160,
      "step": 5927,
      "time_per_iteration": 2.727977752685547
    },
    {
      "auxiliary_loss_clip": 0.01172786,
      "auxiliary_loss_mlp": 0.01024513,
      "balance_loss_clip": 1.05120218,
      "balance_loss_mlp": 1.01685059,
      "epoch": 0.7127998557085312,
      "flos": 16034438315520.0,
      "grad_norm": 2.13889583328419,
      "language_loss": 0.82307804,
      "learning_rate": 8.044797740620506e-07,
      "loss": 0.84505105,
      "num_input_tokens_seen": 127427765,
      "step": 5928,
      "time_per_iteration": 2.557039976119995
    },
    {
      "auxiliary_loss_clip": 0.01166976,
      "auxiliary_loss_mlp": 0.01028097,
      "balance_loss_clip": 1.05273342,
      "balance_loss_mlp": 1.01975799,
      "epoch": 0.7129200985991703,
      "flos": 23403271582080.0,
      "grad_norm": 2.812047570150256,
      "language_loss": 0.78408217,
      "learning_rate": 8.038553826419494e-07,
      "loss": 0.8060329,
      "num_input_tokens_seen": 127446475,
      "step": 5929,
      "time_per_iteration": 2.710737943649292
    },
    {
      "auxiliary_loss_clip": 0.01148401,
      "auxiliary_loss_mlp": 0.01025061,
      "balance_loss_clip": 1.04843211,
      "balance_loss_mlp": 1.01700234,
      "epoch": 0.7130403414898094,
      "flos": 21397445326080.0,
      "grad_norm": 1.6748820853882171,
      "language_loss": 0.81520116,
      "learning_rate": 8.032311726760364e-07,
      "loss": 0.83693576,
      "num_input_tokens_seen": 127467695,
      "step": 5930,
      "time_per_iteration": 2.701262950897217
    },
    {
      "auxiliary_loss_clip": 0.01177679,
      "auxiliary_loss_mlp": 0.01030986,
      "balance_loss_clip": 1.05302775,
      "balance_loss_mlp": 1.02279007,
      "epoch": 0.7131605843804485,
      "flos": 74739045306240.0,
      "grad_norm": 3.88751084947152,
      "language_loss": 0.69096339,
      "learning_rate": 8.026071442590022e-07,
      "loss": 0.71305001,
      "num_input_tokens_seen": 127494590,
      "step": 5931,
      "time_per_iteration": 3.037511110305786
    },
    {
      "auxiliary_loss_clip": 0.01155904,
      "auxiliary_loss_mlp": 0.01025254,
      "balance_loss_clip": 1.05046165,
      "balance_loss_mlp": 1.01750553,
      "epoch": 0.7132808272710875,
      "flos": 18368739469440.0,
      "grad_norm": 2.1188936286782263,
      "language_loss": 0.80517888,
      "learning_rate": 8.019832974855134e-07,
      "loss": 0.82699049,
      "num_input_tokens_seen": 127512550,
      "step": 5932,
      "time_per_iteration": 2.6343815326690674
    },
    {
      "auxiliary_loss_clip": 0.01181698,
      "auxiliary_loss_mlp": 0.01028074,
      "balance_loss_clip": 1.05686879,
      "balance_loss_mlp": 1.02021766,
      "epoch": 0.7134010701617267,
      "flos": 23253380127360.0,
      "grad_norm": 2.5112117559473637,
      "language_loss": 0.82963634,
      "learning_rate": 8.013596324502052e-07,
      "loss": 0.85173404,
      "num_input_tokens_seen": 127531015,
      "step": 5933,
      "time_per_iteration": 2.6473331451416016
    },
    {
      "auxiliary_loss_clip": 0.01173447,
      "auxiliary_loss_mlp": 0.0102728,
      "balance_loss_clip": 1.05328739,
      "balance_loss_mlp": 1.01928091,
      "epoch": 0.7135213130523658,
      "flos": 23653137565440.0,
      "grad_norm": 1.8383866927596764,
      "language_loss": 0.79224098,
      "learning_rate": 8.007361492476872e-07,
      "loss": 0.81424832,
      "num_input_tokens_seen": 127550340,
      "step": 5934,
      "time_per_iteration": 3.623512029647827
    },
    {
      "auxiliary_loss_clip": 0.01175246,
      "auxiliary_loss_mlp": 0.01022254,
      "balance_loss_clip": 1.05058122,
      "balance_loss_mlp": 1.01474392,
      "epoch": 0.7136415559430048,
      "flos": 24790644443520.0,
      "grad_norm": 1.5633352729215153,
      "language_loss": 0.7931155,
      "learning_rate": 8.001128479725426e-07,
      "loss": 0.81509054,
      "num_input_tokens_seen": 127572245,
      "step": 5935,
      "time_per_iteration": 2.6557071208953857
    },
    {
      "auxiliary_loss_clip": 0.01176627,
      "auxiliary_loss_mlp": 0.01026294,
      "balance_loss_clip": 1.05150592,
      "balance_loss_mlp": 1.01797271,
      "epoch": 0.713761798833644,
      "flos": 18296954138880.0,
      "grad_norm": 1.6262617349186252,
      "language_loss": 0.80822021,
      "learning_rate": 7.994897287193248e-07,
      "loss": 0.83024943,
      "num_input_tokens_seen": 127591625,
      "step": 5936,
      "time_per_iteration": 2.58022141456604
    },
    {
      "auxiliary_loss_clip": 0.011534,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.04781938,
      "balance_loss_mlp": 1.01952803,
      "epoch": 0.713882041724283,
      "flos": 15558262692480.0,
      "grad_norm": 2.591990841261045,
      "language_loss": 0.83612138,
      "learning_rate": 7.988667915825605e-07,
      "loss": 0.8579309,
      "num_input_tokens_seen": 127608690,
      "step": 5937,
      "time_per_iteration": 2.5436315536499023
    },
    {
      "auxiliary_loss_clip": 0.0116375,
      "auxiliary_loss_mlp": 0.0103119,
      "balance_loss_clip": 1.05240321,
      "balance_loss_mlp": 1.0227263,
      "epoch": 0.7140022846149221,
      "flos": 24061011477120.0,
      "grad_norm": 2.816408010814988,
      "language_loss": 0.75699818,
      "learning_rate": 7.982440366567491e-07,
      "loss": 0.77894747,
      "num_input_tokens_seen": 127627180,
      "step": 5938,
      "time_per_iteration": 2.6795692443847656
    },
    {
      "auxiliary_loss_clip": 0.01173357,
      "auxiliary_loss_mlp": 0.01028473,
      "balance_loss_clip": 1.05118406,
      "balance_loss_mlp": 1.0210638,
      "epoch": 0.7141225275055613,
      "flos": 27891710248320.0,
      "grad_norm": 2.605890730610887,
      "language_loss": 0.75190997,
      "learning_rate": 7.97621464036361e-07,
      "loss": 0.77392823,
      "num_input_tokens_seen": 127648940,
      "step": 5939,
      "time_per_iteration": 2.650007724761963
    },
    {
      "auxiliary_loss_clip": 0.01176229,
      "auxiliary_loss_mlp": 0.01032498,
      "balance_loss_clip": 1.05096221,
      "balance_loss_mlp": 1.02412391,
      "epoch": 0.7142427703962003,
      "flos": 19682603147520.0,
      "grad_norm": 1.8179744275053953,
      "language_loss": 0.67972374,
      "learning_rate": 7.969990738158417e-07,
      "loss": 0.70181102,
      "num_input_tokens_seen": 127667350,
      "step": 5940,
      "time_per_iteration": 2.696608304977417
    },
    {
      "auxiliary_loss_clip": 0.0118004,
      "auxiliary_loss_mlp": 0.010333,
      "balance_loss_clip": 1.05368626,
      "balance_loss_mlp": 1.02492583,
      "epoch": 0.7143630132868394,
      "flos": 21032377447680.0,
      "grad_norm": 2.0510503874649926,
      "language_loss": 0.85076749,
      "learning_rate": 7.963768660896062e-07,
      "loss": 0.8729009,
      "num_input_tokens_seen": 127685760,
      "step": 5941,
      "time_per_iteration": 2.5951106548309326
    },
    {
      "auxiliary_loss_clip": 0.01151946,
      "auxiliary_loss_mlp": 0.01025836,
      "balance_loss_clip": 1.05084109,
      "balance_loss_mlp": 1.01820016,
      "epoch": 0.7144832561774785,
      "flos": 24129923719680.0,
      "grad_norm": 2.0791748485047203,
      "language_loss": 0.82299149,
      "learning_rate": 7.957548409520432e-07,
      "loss": 0.8447693,
      "num_input_tokens_seen": 127704985,
      "step": 5942,
      "time_per_iteration": 2.699798583984375
    },
    {
      "auxiliary_loss_clip": 0.01174851,
      "auxiliary_loss_mlp": 0.01026563,
      "balance_loss_clip": 1.0513922,
      "balance_loss_mlp": 1.01886487,
      "epoch": 0.7146034990681176,
      "flos": 16325817442560.0,
      "grad_norm": 4.598883428826336,
      "language_loss": 0.84060264,
      "learning_rate": 7.951329984975135e-07,
      "loss": 0.86261684,
      "num_input_tokens_seen": 127721925,
      "step": 5943,
      "time_per_iteration": 2.5656983852386475
    },
    {
      "auxiliary_loss_clip": 0.01065825,
      "auxiliary_loss_mlp": 0.01000878,
      "balance_loss_clip": 1.00994933,
      "balance_loss_mlp": 0.99997181,
      "epoch": 0.7147237419587567,
      "flos": 69627164232960.0,
      "grad_norm": 0.7088270869680954,
      "language_loss": 0.54181731,
      "learning_rate": 7.94511338820349e-07,
      "loss": 0.56248438,
      "num_input_tokens_seen": 127784230,
      "step": 5944,
      "time_per_iteration": 3.1804707050323486
    },
    {
      "auxiliary_loss_clip": 0.01149201,
      "auxiliary_loss_mlp": 0.01030703,
      "balance_loss_clip": 1.04607821,
      "balance_loss_mlp": 1.02167845,
      "epoch": 0.7148439848493958,
      "flos": 22266806198400.0,
      "grad_norm": 2.2080203534367686,
      "language_loss": 0.78259039,
      "learning_rate": 7.938898620148575e-07,
      "loss": 0.80438942,
      "num_input_tokens_seen": 127801990,
      "step": 5945,
      "time_per_iteration": 2.6507840156555176
    },
    {
      "auxiliary_loss_clip": 0.01144246,
      "auxiliary_loss_mlp": 0.01026061,
      "balance_loss_clip": 1.04610264,
      "balance_loss_mlp": 1.01794863,
      "epoch": 0.7149642277400349,
      "flos": 17931383470080.0,
      "grad_norm": 2.183527981081573,
      "language_loss": 0.71431947,
      "learning_rate": 7.932685681753135e-07,
      "loss": 0.73602259,
      "num_input_tokens_seen": 127819270,
      "step": 5946,
      "time_per_iteration": 2.6554296016693115
    },
    {
      "auxiliary_loss_clip": 0.01150049,
      "auxiliary_loss_mlp": 0.01025151,
      "balance_loss_clip": 1.04545355,
      "balance_loss_mlp": 1.0173552,
      "epoch": 0.7150844706306739,
      "flos": 31681937370240.0,
      "grad_norm": 2.0320880058174335,
      "language_loss": 0.63151628,
      "learning_rate": 7.92647457395969e-07,
      "loss": 0.65326828,
      "num_input_tokens_seen": 127841095,
      "step": 5947,
      "time_per_iteration": 3.8034284114837646
    },
    {
      "auxiliary_loss_clip": 0.01135161,
      "auxiliary_loss_mlp": 0.0102959,
      "balance_loss_clip": 1.05137706,
      "balance_loss_mlp": 1.02165949,
      "epoch": 0.7152047135213131,
      "flos": 10926217451520.0,
      "grad_norm": 2.1928734129988308,
      "language_loss": 0.74268448,
      "learning_rate": 7.920265297710444e-07,
      "loss": 0.76433194,
      "num_input_tokens_seen": 127858485,
      "step": 5948,
      "time_per_iteration": 3.613180160522461
    },
    {
      "auxiliary_loss_clip": 0.01081924,
      "auxiliary_loss_mlp": 0.00761349,
      "balance_loss_clip": 1.0437181,
      "balance_loss_mlp": 1.00074899,
      "epoch": 0.7153249564119522,
      "flos": 20995640812800.0,
      "grad_norm": 2.9838733946197262,
      "language_loss": 0.73580253,
      "learning_rate": 7.914057853947363e-07,
      "loss": 0.75423527,
      "num_input_tokens_seen": 127877665,
      "step": 5949,
      "time_per_iteration": 3.1335971355438232
    },
    {
      "auxiliary_loss_clip": 0.01092646,
      "auxiliary_loss_mlp": 0.01030922,
      "balance_loss_clip": 1.0423497,
      "balance_loss_mlp": 1.02266407,
      "epoch": 0.7154451993025912,
      "flos": 24243114453120.0,
      "grad_norm": 1.7783418429037714,
      "language_loss": 0.62900937,
      "learning_rate": 7.907852243612089e-07,
      "loss": 0.65024507,
      "num_input_tokens_seen": 127898070,
      "step": 5950,
      "time_per_iteration": 3.133176565170288
    },
    {
      "auxiliary_loss_clip": 0.01128574,
      "auxiliary_loss_mlp": 0.01023603,
      "balance_loss_clip": 1.0502007,
      "balance_loss_mlp": 1.01617634,
      "epoch": 0.7155654421932304,
      "flos": 23330947547520.0,
      "grad_norm": 2.401329821404744,
      "language_loss": 0.72502005,
      "learning_rate": 7.901648467646009e-07,
      "loss": 0.74654174,
      "num_input_tokens_seen": 127917010,
      "step": 5951,
      "time_per_iteration": 3.9714195728302
    },
    {
      "auxiliary_loss_clip": 0.01115574,
      "auxiliary_loss_mlp": 0.01031696,
      "balance_loss_clip": 1.04281008,
      "balance_loss_mlp": 1.02372098,
      "epoch": 0.7156856850838694,
      "flos": 22711883621760.0,
      "grad_norm": 2.1040318608374164,
      "language_loss": 0.72590977,
      "learning_rate": 7.895446526990244e-07,
      "loss": 0.74738252,
      "num_input_tokens_seen": 127937025,
      "step": 5952,
      "time_per_iteration": 2.8240067958831787
    },
    {
      "auxiliary_loss_clip": 0.01131198,
      "auxiliary_loss_mlp": 0.01033935,
      "balance_loss_clip": 1.04450846,
      "balance_loss_mlp": 1.02597213,
      "epoch": 0.7158059279745085,
      "flos": 19865424395520.0,
      "grad_norm": 1.8645871196850508,
      "language_loss": 0.75816619,
      "learning_rate": 7.889246422585609e-07,
      "loss": 0.77981746,
      "num_input_tokens_seen": 127956410,
      "step": 5953,
      "time_per_iteration": 2.6862733364105225
    },
    {
      "auxiliary_loss_clip": 0.01164979,
      "auxiliary_loss_mlp": 0.01026023,
      "balance_loss_clip": 1.05066466,
      "balance_loss_mlp": 1.01799107,
      "epoch": 0.7159261708651476,
      "flos": 24134772055680.0,
      "grad_norm": 1.9739582334795083,
      "language_loss": 0.73533857,
      "learning_rate": 7.883048155372675e-07,
      "loss": 0.75724852,
      "num_input_tokens_seen": 127974925,
      "step": 5954,
      "time_per_iteration": 2.669032096862793
    },
    {
      "auxiliary_loss_clip": 0.01130413,
      "auxiliary_loss_mlp": 0.01022494,
      "balance_loss_clip": 1.04968786,
      "balance_loss_mlp": 1.01471531,
      "epoch": 0.7160464137557867,
      "flos": 16983198201600.0,
      "grad_norm": 2.1551078970417077,
      "language_loss": 0.72278494,
      "learning_rate": 7.876851726291698e-07,
      "loss": 0.74431401,
      "num_input_tokens_seen": 127993225,
      "step": 5955,
      "time_per_iteration": 2.7399590015411377
    },
    {
      "auxiliary_loss_clip": 0.011494,
      "auxiliary_loss_mlp": 0.01029371,
      "balance_loss_clip": 1.04796553,
      "balance_loss_mlp": 1.0220809,
      "epoch": 0.7161666566464258,
      "flos": 25228251838080.0,
      "grad_norm": 1.799662040093095,
      "language_loss": 0.78662586,
      "learning_rate": 7.870657136282666e-07,
      "loss": 0.80841357,
      "num_input_tokens_seen": 128012085,
      "step": 5956,
      "time_per_iteration": 2.685518503189087
    },
    {
      "auxiliary_loss_clip": 0.01136486,
      "auxiliary_loss_mlp": 0.01031382,
      "balance_loss_clip": 1.04936683,
      "balance_loss_mlp": 1.02359724,
      "epoch": 0.7162868995370649,
      "flos": 26468390851200.0,
      "grad_norm": 1.6305868898740596,
      "language_loss": 0.82122582,
      "learning_rate": 7.86446438628531e-07,
      "loss": 0.84290451,
      "num_input_tokens_seen": 128033155,
      "step": 5957,
      "time_per_iteration": 2.812095880508423
    },
    {
      "auxiliary_loss_clip": 0.01040755,
      "auxiliary_loss_mlp": 0.01001137,
      "balance_loss_clip": 1.01083469,
      "balance_loss_mlp": 1.00026071,
      "epoch": 0.716407142427704,
      "flos": 69998912040960.0,
      "grad_norm": 0.7809520189644259,
      "language_loss": 0.56791413,
      "learning_rate": 7.858273477239059e-07,
      "loss": 0.58833307,
      "num_input_tokens_seen": 128101575,
      "step": 5958,
      "time_per_iteration": 3.234980344772339
    },
    {
      "auxiliary_loss_clip": 0.0110648,
      "auxiliary_loss_mlp": 0.01025527,
      "balance_loss_clip": 1.04912233,
      "balance_loss_mlp": 1.01726568,
      "epoch": 0.716527385318343,
      "flos": 20740459616640.0,
      "grad_norm": 3.117949667015653,
      "language_loss": 0.71365553,
      "learning_rate": 7.852084410083067e-07,
      "loss": 0.73497558,
      "num_input_tokens_seen": 128120395,
      "step": 5959,
      "time_per_iteration": 3.750579833984375
    },
    {
      "auxiliary_loss_clip": 0.01171573,
      "auxiliary_loss_mlp": 0.0102432,
      "balance_loss_clip": 1.05051303,
      "balance_loss_mlp": 1.01713133,
      "epoch": 0.7166476282089821,
      "flos": 25371966153600.0,
      "grad_norm": 1.7593634451210285,
      "language_loss": 0.63592601,
      "learning_rate": 7.84589718575621e-07,
      "loss": 0.6578849,
      "num_input_tokens_seen": 128140840,
      "step": 5960,
      "time_per_iteration": 2.675260305404663
    },
    {
      "auxiliary_loss_clip": 0.0117621,
      "auxiliary_loss_mlp": 0.01027276,
      "balance_loss_clip": 1.05131888,
      "balance_loss_mlp": 1.01929474,
      "epoch": 0.7167678710996213,
      "flos": 24133730561280.0,
      "grad_norm": 2.4408865691080504,
      "language_loss": 0.68858105,
      "learning_rate": 7.83971180519708e-07,
      "loss": 0.71061593,
      "num_input_tokens_seen": 128159695,
      "step": 5961,
      "time_per_iteration": 2.6197164058685303
    },
    {
      "auxiliary_loss_clip": 0.01109074,
      "auxiliary_loss_mlp": 0.01027888,
      "balance_loss_clip": 1.04448605,
      "balance_loss_mlp": 1.01967454,
      "epoch": 0.7168881139902603,
      "flos": 30226586019840.0,
      "grad_norm": 2.6537893676537774,
      "language_loss": 0.75211465,
      "learning_rate": 7.833528269344008e-07,
      "loss": 0.77348423,
      "num_input_tokens_seen": 128179600,
      "step": 5962,
      "time_per_iteration": 2.8583767414093018
    },
    {
      "auxiliary_loss_clip": 0.01168929,
      "auxiliary_loss_mlp": 0.01029708,
      "balance_loss_clip": 1.05337501,
      "balance_loss_mlp": 1.02098203,
      "epoch": 0.7170083568808994,
      "flos": 14606414236800.0,
      "grad_norm": 2.049010124415216,
      "language_loss": 0.77830112,
      "learning_rate": 7.827346579135023e-07,
      "loss": 0.80028749,
      "num_input_tokens_seen": 128196940,
      "step": 5963,
      "time_per_iteration": 2.573758125305176
    },
    {
      "auxiliary_loss_clip": 0.0113169,
      "auxiliary_loss_mlp": 0.00762213,
      "balance_loss_clip": 1.04438269,
      "balance_loss_mlp": 1.00071239,
      "epoch": 0.7171285997715385,
      "flos": 23331091201920.0,
      "grad_norm": 1.8308436510797468,
      "language_loss": 0.83502597,
      "learning_rate": 7.821166735507885e-07,
      "loss": 0.85396504,
      "num_input_tokens_seen": 128215970,
      "step": 5964,
      "time_per_iteration": 2.7664449214935303
    },
    {
      "auxiliary_loss_clip": 0.01127895,
      "auxiliary_loss_mlp": 0.01020774,
      "balance_loss_clip": 1.04679394,
      "balance_loss_mlp": 1.01343417,
      "epoch": 0.7172488426621776,
      "flos": 16543543731840.0,
      "grad_norm": 2.1137629559728692,
      "language_loss": 0.68649817,
      "learning_rate": 7.81498873940007e-07,
      "loss": 0.70798481,
      "num_input_tokens_seen": 128233185,
      "step": 5965,
      "time_per_iteration": 2.648540496826172
    },
    {
      "auxiliary_loss_clip": 0.01166485,
      "auxiliary_loss_mlp": 0.01025726,
      "balance_loss_clip": 1.05000758,
      "balance_loss_mlp": 1.01703584,
      "epoch": 0.7173690855528166,
      "flos": 26541612725760.0,
      "grad_norm": 2.3764515746994457,
      "language_loss": 0.77653849,
      "learning_rate": 7.808812591748768e-07,
      "loss": 0.79846054,
      "num_input_tokens_seen": 128253565,
      "step": 5966,
      "time_per_iteration": 2.7391889095306396
    },
    {
      "auxiliary_loss_clip": 0.0117997,
      "auxiliary_loss_mlp": 0.01028076,
      "balance_loss_clip": 1.05352652,
      "balance_loss_mlp": 1.01968944,
      "epoch": 0.7174893284434558,
      "flos": 22784099915520.0,
      "grad_norm": 3.467034283452036,
      "language_loss": 0.6501137,
      "learning_rate": 7.802638293490915e-07,
      "loss": 0.67219412,
      "num_input_tokens_seen": 128273210,
      "step": 5967,
      "time_per_iteration": 2.6215691566467285
    },
    {
      "auxiliary_loss_clip": 0.01162055,
      "auxiliary_loss_mlp": 0.01029308,
      "balance_loss_clip": 1.05564296,
      "balance_loss_mlp": 1.0211364,
      "epoch": 0.7176095713340949,
      "flos": 23293564467840.0,
      "grad_norm": 3.4201718005802815,
      "language_loss": 0.76902938,
      "learning_rate": 7.796465845563123e-07,
      "loss": 0.79094303,
      "num_input_tokens_seen": 128292085,
      "step": 5968,
      "time_per_iteration": 2.6854147911071777
    },
    {
      "auxiliary_loss_clip": 0.01138599,
      "auxiliary_loss_mlp": 0.00761674,
      "balance_loss_clip": 1.04860079,
      "balance_loss_mlp": 1.00072598,
      "epoch": 0.7177298142247339,
      "flos": 25591631777280.0,
      "grad_norm": 1.9649410638550273,
      "language_loss": 0.79546845,
      "learning_rate": 7.790295248901766e-07,
      "loss": 0.81447119,
      "num_input_tokens_seen": 128313215,
      "step": 5969,
      "time_per_iteration": 2.8386666774749756
    },
    {
      "auxiliary_loss_clip": 0.01126068,
      "auxiliary_loss_mlp": 0.01028074,
      "balance_loss_clip": 1.04607904,
      "balance_loss_mlp": 1.01975346,
      "epoch": 0.7178500571153731,
      "flos": 31652778504960.0,
      "grad_norm": 1.875014585983829,
      "language_loss": 0.6251179,
      "learning_rate": 7.784126504442902e-07,
      "loss": 0.64665931,
      "num_input_tokens_seen": 128336445,
      "step": 5970,
      "time_per_iteration": 2.801791191101074
    },
    {
      "auxiliary_loss_clip": 0.01117003,
      "auxiliary_loss_mlp": 0.01029802,
      "balance_loss_clip": 1.0431689,
      "balance_loss_mlp": 1.02202952,
      "epoch": 0.7179703000060121,
      "flos": 19427242383360.0,
      "grad_norm": 3.9945368449704226,
      "language_loss": 0.68238389,
      "learning_rate": 7.777959613122351e-07,
      "loss": 0.703852,
      "num_input_tokens_seen": 128356270,
      "step": 5971,
      "time_per_iteration": 2.7845680713653564
    },
    {
      "auxiliary_loss_clip": 0.01119869,
      "auxiliary_loss_mlp": 0.01031836,
      "balance_loss_clip": 1.04618788,
      "balance_loss_mlp": 1.02433181,
      "epoch": 0.7180905428966512,
      "flos": 28839249072000.0,
      "grad_norm": 1.7410372716337752,
      "language_loss": 0.78170121,
      "learning_rate": 7.771794575875604e-07,
      "loss": 0.80321825,
      "num_input_tokens_seen": 128378140,
      "step": 5972,
      "time_per_iteration": 3.6145286560058594
    },
    {
      "auxiliary_loss_clip": 0.01132778,
      "auxiliary_loss_mlp": 0.01029957,
      "balance_loss_clip": 1.04463863,
      "balance_loss_mlp": 1.02122533,
      "epoch": 0.7182107857872904,
      "flos": 20047563285120.0,
      "grad_norm": 2.4296123175390942,
      "language_loss": 0.77800262,
      "learning_rate": 7.765631393637888e-07,
      "loss": 0.79962999,
      "num_input_tokens_seen": 128396335,
      "step": 5973,
      "time_per_iteration": 3.6025328636169434
    },
    {
      "auxiliary_loss_clip": 0.01146368,
      "auxiliary_loss_mlp": 0.01024039,
      "balance_loss_clip": 1.049173,
      "balance_loss_mlp": 1.01610613,
      "epoch": 0.7183310286779294,
      "flos": 22747686503040.0,
      "grad_norm": 3.3256684321408487,
      "language_loss": 0.48410228,
      "learning_rate": 7.75947006734417e-07,
      "loss": 0.50580633,
      "num_input_tokens_seen": 128414115,
      "step": 5974,
      "time_per_iteration": 2.714958667755127
    },
    {
      "auxiliary_loss_clip": 0.01159535,
      "auxiliary_loss_mlp": 0.01027487,
      "balance_loss_clip": 1.05048144,
      "balance_loss_mlp": 1.01946473,
      "epoch": 0.7184512715685685,
      "flos": 17158262112000.0,
      "grad_norm": 2.5054908743621875,
      "language_loss": 0.829301,
      "learning_rate": 7.753310597929101e-07,
      "loss": 0.85117126,
      "num_input_tokens_seen": 128430755,
      "step": 5975,
      "time_per_iteration": 2.602342128753662
    },
    {
      "auxiliary_loss_clip": 0.01027059,
      "auxiliary_loss_mlp": 0.01000213,
      "balance_loss_clip": 1.01051307,
      "balance_loss_mlp": 0.99927104,
      "epoch": 0.7185715144592076,
      "flos": 65509611448320.0,
      "grad_norm": 0.751308511307516,
      "language_loss": 0.55085135,
      "learning_rate": 7.747152986327095e-07,
      "loss": 0.57112408,
      "num_input_tokens_seen": 128491300,
      "step": 5976,
      "time_per_iteration": 3.2623488903045654
    },
    {
      "auxiliary_loss_clip": 0.01147417,
      "auxiliary_loss_mlp": 0.01032954,
      "balance_loss_clip": 1.050946,
      "balance_loss_mlp": 1.02565575,
      "epoch": 0.7186917573498467,
      "flos": 16180522928640.0,
      "grad_norm": 1.9749649136740497,
      "language_loss": 0.67997926,
      "learning_rate": 7.740997233472228e-07,
      "loss": 0.701783,
      "num_input_tokens_seen": 128508920,
      "step": 5977,
      "time_per_iteration": 3.6468207836151123
    },
    {
      "auxiliary_loss_clip": 0.01143778,
      "auxiliary_loss_mlp": 0.01027998,
      "balance_loss_clip": 1.04737115,
      "balance_loss_mlp": 1.02027345,
      "epoch": 0.7188120002404857,
      "flos": 29242274647680.0,
      "grad_norm": 2.428661989824715,
      "language_loss": 0.70788586,
      "learning_rate": 7.734843340298329e-07,
      "loss": 0.72960365,
      "num_input_tokens_seen": 128528745,
      "step": 5978,
      "time_per_iteration": 2.716818332672119
    },
    {
      "auxiliary_loss_clip": 0.01145874,
      "auxiliary_loss_mlp": 0.01028857,
      "balance_loss_clip": 1.04620731,
      "balance_loss_mlp": 1.02023256,
      "epoch": 0.7189322431311249,
      "flos": 33401161008000.0,
      "grad_norm": 1.9675607144250014,
      "language_loss": 0.75196171,
      "learning_rate": 7.72869130773895e-07,
      "loss": 0.77370906,
      "num_input_tokens_seen": 128549345,
      "step": 5979,
      "time_per_iteration": 2.8492836952209473
    },
    {
      "auxiliary_loss_clip": 0.01035528,
      "auxiliary_loss_mlp": 0.01003384,
      "balance_loss_clip": 1.01208782,
      "balance_loss_mlp": 1.00245404,
      "epoch": 0.719052486021764,
      "flos": 61351263792000.0,
      "grad_norm": 0.7842388684541819,
      "language_loss": 0.59285891,
      "learning_rate": 7.722541136727343e-07,
      "loss": 0.61324805,
      "num_input_tokens_seen": 128605360,
      "step": 5980,
      "time_per_iteration": 3.1432173252105713
    },
    {
      "auxiliary_loss_clip": 0.01099892,
      "auxiliary_loss_mlp": 0.01030987,
      "balance_loss_clip": 1.04190087,
      "balance_loss_mlp": 1.02313089,
      "epoch": 0.719172728912403,
      "flos": 15596795007360.0,
      "grad_norm": 2.06624006246961,
      "language_loss": 0.80747283,
      "learning_rate": 7.716392828196483e-07,
      "loss": 0.8287816,
      "num_input_tokens_seen": 128623160,
      "step": 5981,
      "time_per_iteration": 2.8167059421539307
    },
    {
      "auxiliary_loss_clip": 0.01084836,
      "auxiliary_loss_mlp": 0.01027843,
      "balance_loss_clip": 1.03709698,
      "balance_loss_mlp": 1.01988018,
      "epoch": 0.7192929718030422,
      "flos": 15553162961280.0,
      "grad_norm": 2.3876764639487456,
      "language_loss": 0.77405995,
      "learning_rate": 7.710246383079064e-07,
      "loss": 0.79518676,
      "num_input_tokens_seen": 128638545,
      "step": 5982,
      "time_per_iteration": 2.9247798919677734
    },
    {
      "auxiliary_loss_clip": 0.01117509,
      "auxiliary_loss_mlp": 0.0102678,
      "balance_loss_clip": 1.04542637,
      "balance_loss_mlp": 1.01897812,
      "epoch": 0.7194132146936812,
      "flos": 21862487733120.0,
      "grad_norm": 2.7667452519670346,
      "language_loss": 0.92374742,
      "learning_rate": 7.704101802307492e-07,
      "loss": 0.94519031,
      "num_input_tokens_seen": 128650845,
      "step": 5983,
      "time_per_iteration": 3.4936769008636475
    },
    {
      "auxiliary_loss_clip": 0.01123752,
      "auxiliary_loss_mlp": 0.01030326,
      "balance_loss_clip": 1.0487839,
      "balance_loss_mlp": 1.02168894,
      "epoch": 0.7195334575843203,
      "flos": 27338900958720.0,
      "grad_norm": 2.0906719105881595,
      "language_loss": 0.87055308,
      "learning_rate": 7.697959086813912e-07,
      "loss": 0.89209384,
      "num_input_tokens_seen": 128667010,
      "step": 5984,
      "time_per_iteration": 2.9324207305908203
    },
    {
      "auxiliary_loss_clip": 0.01159766,
      "auxiliary_loss_mlp": 0.01027384,
      "balance_loss_clip": 1.05074286,
      "balance_loss_mlp": 1.01982021,
      "epoch": 0.7196537004749595,
      "flos": 18770615809920.0,
      "grad_norm": 1.7245491205810937,
      "language_loss": 0.80626458,
      "learning_rate": 7.691818237530145e-07,
      "loss": 0.82813609,
      "num_input_tokens_seen": 128685870,
      "step": 5985,
      "time_per_iteration": 3.5799553394317627
    },
    {
      "auxiliary_loss_clip": 0.01084441,
      "auxiliary_loss_mlp": 0.01029498,
      "balance_loss_clip": 1.04027152,
      "balance_loss_mlp": 1.02103376,
      "epoch": 0.7197739433655985,
      "flos": 24531009960960.0,
      "grad_norm": 2.6613118419920907,
      "language_loss": 0.77269256,
      "learning_rate": 7.685679255387774e-07,
      "loss": 0.79383194,
      "num_input_tokens_seen": 128704185,
      "step": 5986,
      "time_per_iteration": 2.848447799682617
    },
    {
      "auxiliary_loss_clip": 0.01119925,
      "auxiliary_loss_mlp": 0.00761445,
      "balance_loss_clip": 1.04761243,
      "balance_loss_mlp": 1.00065696,
      "epoch": 0.7198941862562376,
      "flos": 18040587793920.0,
      "grad_norm": 2.3101039760934,
      "language_loss": 0.77163959,
      "learning_rate": 7.679542141318065e-07,
      "loss": 0.79045331,
      "num_input_tokens_seen": 128721290,
      "step": 5987,
      "time_per_iteration": 2.7902231216430664
    },
    {
      "auxiliary_loss_clip": 0.01115213,
      "auxiliary_loss_mlp": 0.01027579,
      "balance_loss_clip": 1.04683971,
      "balance_loss_mlp": 1.01966953,
      "epoch": 0.7200144291468767,
      "flos": 29022393542400.0,
      "grad_norm": 1.912239850623922,
      "language_loss": 0.76022768,
      "learning_rate": 7.673406896252013e-07,
      "loss": 0.78165567,
      "num_input_tokens_seen": 128742665,
      "step": 5988,
      "time_per_iteration": 2.784099578857422
    },
    {
      "auxiliary_loss_clip": 0.01164952,
      "auxiliary_loss_mlp": 0.01036626,
      "balance_loss_clip": 1.05282557,
      "balance_loss_mlp": 1.02840018,
      "epoch": 0.7201346720375158,
      "flos": 25374264624000.0,
      "grad_norm": 1.7733607330845707,
      "language_loss": 0.7841742,
      "learning_rate": 7.667273521120347e-07,
      "loss": 0.80619001,
      "num_input_tokens_seen": 128762225,
      "step": 5989,
      "time_per_iteration": 2.746147871017456
    },
    {
      "auxiliary_loss_clip": 0.01135914,
      "auxiliary_loss_mlp": 0.01030731,
      "balance_loss_clip": 1.04445708,
      "balance_loss_mlp": 1.02190912,
      "epoch": 0.7202549149281549,
      "flos": 14355614499840.0,
      "grad_norm": 1.8698319727057517,
      "language_loss": 0.7921347,
      "learning_rate": 7.661142016853468e-07,
      "loss": 0.81380111,
      "num_input_tokens_seen": 128779585,
      "step": 5990,
      "time_per_iteration": 2.6670000553131104
    },
    {
      "auxiliary_loss_clip": 0.01126519,
      "auxiliary_loss_mlp": 0.01025562,
      "balance_loss_clip": 1.04767084,
      "balance_loss_mlp": 1.01770639,
      "epoch": 0.7203751578187939,
      "flos": 23001682550400.0,
      "grad_norm": 1.751121775232293,
      "language_loss": 0.75000638,
      "learning_rate": 7.655012384381543e-07,
      "loss": 0.77152717,
      "num_input_tokens_seen": 128799070,
      "step": 5991,
      "time_per_iteration": 2.791024923324585
    },
    {
      "auxiliary_loss_clip": 0.01152403,
      "auxiliary_loss_mlp": 0.01030979,
      "balance_loss_clip": 1.05158615,
      "balance_loss_mlp": 1.02218127,
      "epoch": 0.7204954007094331,
      "flos": 23692424065920.0,
      "grad_norm": 2.4067523804173785,
      "language_loss": 0.82015342,
      "learning_rate": 7.648884624634415e-07,
      "loss": 0.84198725,
      "num_input_tokens_seen": 128817620,
      "step": 5992,
      "time_per_iteration": 2.6515281200408936
    },
    {
      "auxiliary_loss_clip": 0.01144493,
      "auxiliary_loss_mlp": 0.01030664,
      "balance_loss_clip": 1.04863167,
      "balance_loss_mlp": 1.02305841,
      "epoch": 0.7206156436000721,
      "flos": 16253026531200.0,
      "grad_norm": 3.5046107185293507,
      "language_loss": 0.89079905,
      "learning_rate": 7.642758738541683e-07,
      "loss": 0.91255057,
      "num_input_tokens_seen": 128834200,
      "step": 5993,
      "time_per_iteration": 2.7419590950012207
    },
    {
      "auxiliary_loss_clip": 0.01043252,
      "auxiliary_loss_mlp": 0.01004469,
      "balance_loss_clip": 1.00993037,
      "balance_loss_mlp": 1.00360513,
      "epoch": 0.7207358864907112,
      "flos": 54377806504320.0,
      "grad_norm": 0.7590765284402982,
      "language_loss": 0.6067425,
      "learning_rate": 7.636634727032621e-07,
      "loss": 0.62721974,
      "num_input_tokens_seen": 128891305,
      "step": 5994,
      "time_per_iteration": 3.124150514602661
    },
    {
      "auxiliary_loss_clip": 0.01103755,
      "auxiliary_loss_mlp": 0.01031741,
      "balance_loss_clip": 1.04435802,
      "balance_loss_mlp": 1.02341139,
      "epoch": 0.7208561293813504,
      "flos": 19135540033920.0,
      "grad_norm": 1.982472919048864,
      "language_loss": 0.78792602,
      "learning_rate": 7.630512591036231e-07,
      "loss": 0.80928099,
      "num_input_tokens_seen": 128910615,
      "step": 5995,
      "time_per_iteration": 2.8003463745117188
    },
    {
      "auxiliary_loss_clip": 0.011453,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.04731584,
      "balance_loss_mlp": 1.01876497,
      "epoch": 0.7209763722719894,
      "flos": 17748526308480.0,
      "grad_norm": 2.7848060816626092,
      "language_loss": 0.64898741,
      "learning_rate": 7.624392331481255e-07,
      "loss": 0.67070633,
      "num_input_tokens_seen": 128928270,
      "step": 5996,
      "time_per_iteration": 2.68149733543396
    },
    {
      "auxiliary_loss_clip": 0.01048251,
      "auxiliary_loss_mlp": 0.0099952,
      "balance_loss_clip": 1.0186913,
      "balance_loss_mlp": 0.99861968,
      "epoch": 0.7210966151626285,
      "flos": 66819488716800.0,
      "grad_norm": 0.7463644830250836,
      "language_loss": 0.51773727,
      "learning_rate": 7.618273949296115e-07,
      "loss": 0.53821492,
      "num_input_tokens_seen": 128987780,
      "step": 5997,
      "time_per_iteration": 3.190070867538452
    },
    {
      "auxiliary_loss_clip": 0.01175576,
      "auxiliary_loss_mlp": 0.01026162,
      "balance_loss_clip": 1.05169499,
      "balance_loss_mlp": 1.01833034,
      "epoch": 0.7212168580532676,
      "flos": 21141869080320.0,
      "grad_norm": 1.9261579641132154,
      "language_loss": 0.68548644,
      "learning_rate": 7.612157445408987e-07,
      "loss": 0.7075038,
      "num_input_tokens_seen": 129005590,
      "step": 5998,
      "time_per_iteration": 2.655216693878174
    },
    {
      "auxiliary_loss_clip": 0.01148982,
      "auxiliary_loss_mlp": 0.01025809,
      "balance_loss_clip": 1.0512712,
      "balance_loss_mlp": 1.01743472,
      "epoch": 0.7213371009439067,
      "flos": 22345738335360.0,
      "grad_norm": 2.537876802720384,
      "language_loss": 0.74597132,
      "learning_rate": 7.606042820747716e-07,
      "loss": 0.76771921,
      "num_input_tokens_seen": 129021995,
      "step": 5999,
      "time_per_iteration": 4.6311211585998535
    },
    {
      "auxiliary_loss_clip": 0.01166325,
      "auxiliary_loss_mlp": 0.01032974,
      "balance_loss_clip": 1.05576825,
      "balance_loss_mlp": 1.02490985,
      "epoch": 0.7214573438345457,
      "flos": 18515901490560.0,
      "grad_norm": 1.8456706299084602,
      "language_loss": 0.85608071,
      "learning_rate": 7.599930076239889e-07,
      "loss": 0.87807369,
      "num_input_tokens_seen": 129039280,
      "step": 6000,
      "time_per_iteration": 2.627577543258667
    },
    {
      "auxiliary_loss_clip": 0.01164196,
      "auxiliary_loss_mlp": 0.00762173,
      "balance_loss_clip": 1.05328298,
      "balance_loss_mlp": 1.00071025,
      "epoch": 0.7215775867251849,
      "flos": 35736108606720.0,
      "grad_norm": 2.8261844982843125,
      "language_loss": 0.70977616,
      "learning_rate": 7.593819212812818e-07,
      "loss": 0.72903979,
      "num_input_tokens_seen": 129060860,
      "step": 6001,
      "time_per_iteration": 2.8348746299743652
    },
    {
      "auxiliary_loss_clip": 0.01161051,
      "auxiliary_loss_mlp": 0.01025826,
      "balance_loss_clip": 1.05252898,
      "balance_loss_mlp": 1.01833427,
      "epoch": 0.721697829615824,
      "flos": 20372410909440.0,
      "grad_norm": 3.4038872615921716,
      "language_loss": 0.72438967,
      "learning_rate": 7.587710231393508e-07,
      "loss": 0.74625838,
      "num_input_tokens_seen": 129079215,
      "step": 6002,
      "time_per_iteration": 2.585327386856079
    },
    {
      "auxiliary_loss_clip": 0.01173744,
      "auxiliary_loss_mlp": 0.01030455,
      "balance_loss_clip": 1.05143714,
      "balance_loss_mlp": 1.02282596,
      "epoch": 0.721818072506463,
      "flos": 20229809915520.0,
      "grad_norm": 2.3111485812140344,
      "language_loss": 0.83735824,
      "learning_rate": 7.581603132908685e-07,
      "loss": 0.85940015,
      "num_input_tokens_seen": 129097185,
      "step": 6003,
      "time_per_iteration": 3.63535475730896
    },
    {
      "auxiliary_loss_clip": 0.01159956,
      "auxiliary_loss_mlp": 0.0102824,
      "balance_loss_clip": 1.04772282,
      "balance_loss_mlp": 1.01963937,
      "epoch": 0.7219383153971022,
      "flos": 18186887888640.0,
      "grad_norm": 2.3576927185092553,
      "language_loss": 0.79193389,
      "learning_rate": 7.575497918284795e-07,
      "loss": 0.81381583,
      "num_input_tokens_seen": 129114730,
      "step": 6004,
      "time_per_iteration": 2.6979503631591797
    },
    {
      "auxiliary_loss_clip": 0.01143464,
      "auxiliary_loss_mlp": 0.01025946,
      "balance_loss_clip": 1.04771304,
      "balance_loss_mlp": 1.01730943,
      "epoch": 0.7220585582877412,
      "flos": 17342124854400.0,
      "grad_norm": 2.1616602849259805,
      "language_loss": 0.74735272,
      "learning_rate": 7.569394588447984e-07,
      "loss": 0.76904678,
      "num_input_tokens_seen": 129131745,
      "step": 6005,
      "time_per_iteration": 2.759875535964966
    },
    {
      "auxiliary_loss_clip": 0.01129845,
      "auxiliary_loss_mlp": 0.01030482,
      "balance_loss_clip": 1.04786444,
      "balance_loss_mlp": 1.0223217,
      "epoch": 0.7221788011783803,
      "flos": 16976338704000.0,
      "grad_norm": 2.7959205969967176,
      "language_loss": 0.78585351,
      "learning_rate": 7.563293144324146e-07,
      "loss": 0.80745679,
      "num_input_tokens_seen": 129147295,
      "step": 6006,
      "time_per_iteration": 2.73078989982605
    },
    {
      "auxiliary_loss_clip": 0.01173838,
      "auxiliary_loss_mlp": 0.01024771,
      "balance_loss_clip": 1.05216026,
      "balance_loss_mlp": 1.01765418,
      "epoch": 0.7222990440690195,
      "flos": 26286359702400.0,
      "grad_norm": 1.92542911068581,
      "language_loss": 0.80499148,
      "learning_rate": 7.557193586838834e-07,
      "loss": 0.82697749,
      "num_input_tokens_seen": 129162660,
      "step": 6007,
      "time_per_iteration": 2.619342088699341
    },
    {
      "auxiliary_loss_clip": 0.01149458,
      "auxiliary_loss_mlp": 0.01028431,
      "balance_loss_clip": 1.04859757,
      "balance_loss_mlp": 1.02043259,
      "epoch": 0.7224192869596585,
      "flos": 17601687509760.0,
      "grad_norm": 3.87547572657334,
      "language_loss": 0.71245003,
      "learning_rate": 7.551095916917371e-07,
      "loss": 0.73422891,
      "num_input_tokens_seen": 129179990,
      "step": 6008,
      "time_per_iteration": 2.6805214881896973
    },
    {
      "auxiliary_loss_clip": 0.0113214,
      "auxiliary_loss_mlp": 0.01029041,
      "balance_loss_clip": 1.04930377,
      "balance_loss_mlp": 1.02028489,
      "epoch": 0.7225395298502976,
      "flos": 12932331016320.0,
      "grad_norm": 3.280723275004061,
      "language_loss": 0.6694327,
      "learning_rate": 7.545000135484758e-07,
      "loss": 0.69104451,
      "num_input_tokens_seen": 129197425,
      "step": 6009,
      "time_per_iteration": 2.712510585784912
    },
    {
      "auxiliary_loss_clip": 0.01139861,
      "auxiliary_loss_mlp": 0.01029663,
      "balance_loss_clip": 1.04719567,
      "balance_loss_mlp": 1.02168167,
      "epoch": 0.7226597727409367,
      "flos": 29643899592960.0,
      "grad_norm": 2.591582905671445,
      "language_loss": 0.62657589,
      "learning_rate": 7.538906243465714e-07,
      "loss": 0.64827108,
      "num_input_tokens_seen": 129217560,
      "step": 6010,
      "time_per_iteration": 3.7661633491516113
    },
    {
      "auxiliary_loss_clip": 0.01138149,
      "auxiliary_loss_mlp": 0.01030619,
      "balance_loss_clip": 1.04832792,
      "balance_loss_mlp": 1.02238214,
      "epoch": 0.7227800156315758,
      "flos": 13771635183360.0,
      "grad_norm": 1.9927260947977894,
      "language_loss": 0.78693044,
      "learning_rate": 7.5328142417847e-07,
      "loss": 0.80861813,
      "num_input_tokens_seen": 129234325,
      "step": 6011,
      "time_per_iteration": 2.8367698192596436
    },
    {
      "auxiliary_loss_clip": 0.01131235,
      "auxiliary_loss_mlp": 0.01027533,
      "balance_loss_clip": 1.04867125,
      "balance_loss_mlp": 1.02017164,
      "epoch": 0.7229002585222148,
      "flos": 20301882554880.0,
      "grad_norm": 1.809864589995202,
      "language_loss": 0.69437706,
      "learning_rate": 7.526724131365838e-07,
      "loss": 0.71596473,
      "num_input_tokens_seen": 129255280,
      "step": 6012,
      "time_per_iteration": 2.7276840209960938
    },
    {
      "auxiliary_loss_clip": 0.01182166,
      "auxiliary_loss_mlp": 0.00762316,
      "balance_loss_clip": 1.05561876,
      "balance_loss_mlp": 1.00071096,
      "epoch": 0.723020501412854,
      "flos": 16581250033920.0,
      "grad_norm": 1.7976391890015062,
      "language_loss": 0.7095083,
      "learning_rate": 7.520635913133017e-07,
      "loss": 0.72895312,
      "num_input_tokens_seen": 129273910,
      "step": 6013,
      "time_per_iteration": 2.6880040168762207
    },
    {
      "auxiliary_loss_clip": 0.01153744,
      "auxiliary_loss_mlp": 0.01028129,
      "balance_loss_clip": 1.04988325,
      "balance_loss_mlp": 1.01937854,
      "epoch": 0.7231407443034931,
      "flos": 28548300908160.0,
      "grad_norm": 7.036816834493344,
      "language_loss": 0.82745481,
      "learning_rate": 7.514549588009798e-07,
      "loss": 0.84927356,
      "num_input_tokens_seen": 129294785,
      "step": 6014,
      "time_per_iteration": 2.7878992557525635
    },
    {
      "auxiliary_loss_clip": 0.011385,
      "auxiliary_loss_mlp": 0.0103009,
      "balance_loss_clip": 1.04835367,
      "balance_loss_mlp": 1.02106571,
      "epoch": 0.7232609871941321,
      "flos": 30008536508160.0,
      "grad_norm": 1.9630433048622626,
      "language_loss": 0.70579469,
      "learning_rate": 7.508465156919492e-07,
      "loss": 0.72748065,
      "num_input_tokens_seen": 129318295,
      "step": 6015,
      "time_per_iteration": 2.855742931365967
    },
    {
      "auxiliary_loss_clip": 0.01102322,
      "auxiliary_loss_mlp": 0.01029542,
      "balance_loss_clip": 1.0440619,
      "balance_loss_mlp": 1.02058947,
      "epoch": 0.7233812300847713,
      "flos": 16654005031680.0,
      "grad_norm": 3.4901091505612767,
      "language_loss": 0.61476409,
      "learning_rate": 7.502382620785083e-07,
      "loss": 0.63608277,
      "num_input_tokens_seen": 129334845,
      "step": 6016,
      "time_per_iteration": 2.7247204780578613
    },
    {
      "auxiliary_loss_clip": 0.0105884,
      "auxiliary_loss_mlp": 0.00751984,
      "balance_loss_clip": 1.01106882,
      "balance_loss_mlp": 1.00013006,
      "epoch": 0.7235014729754103,
      "flos": 67258784050560.0,
      "grad_norm": 0.8050577414729131,
      "language_loss": 0.62459111,
      "learning_rate": 7.496301980529289e-07,
      "loss": 0.64269936,
      "num_input_tokens_seen": 129398055,
      "step": 6017,
      "time_per_iteration": 3.271754264831543
    },
    {
      "auxiliary_loss_clip": 0.01158179,
      "auxiliary_loss_mlp": 0.01029893,
      "balance_loss_clip": 1.04774213,
      "balance_loss_mlp": 1.02154827,
      "epoch": 0.7236217158660494,
      "flos": 26943237671040.0,
      "grad_norm": 4.091232614937147,
      "language_loss": 0.74497896,
      "learning_rate": 7.490223237074547e-07,
      "loss": 0.76685965,
      "num_input_tokens_seen": 129417765,
      "step": 6018,
      "time_per_iteration": 2.727036952972412
    },
    {
      "auxiliary_loss_clip": 0.01115506,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 1.0439136,
      "balance_loss_mlp": 1.01677346,
      "epoch": 0.7237419587566886,
      "flos": 29423372042880.0,
      "grad_norm": 2.4170026029692186,
      "language_loss": 0.65913928,
      "learning_rate": 7.484146391342989e-07,
      "loss": 0.68054056,
      "num_input_tokens_seen": 129437560,
      "step": 6019,
      "time_per_iteration": 2.830057382583618
    },
    {
      "auxiliary_loss_clip": 0.01107475,
      "auxiliary_loss_mlp": 0.01026521,
      "balance_loss_clip": 1.0428797,
      "balance_loss_mlp": 1.01877201,
      "epoch": 0.7238622016473276,
      "flos": 17821496787840.0,
      "grad_norm": 2.3356474920088797,
      "language_loss": 0.56878936,
      "learning_rate": 7.478071444256484e-07,
      "loss": 0.59012938,
      "num_input_tokens_seen": 129455320,
      "step": 6020,
      "time_per_iteration": 2.7839579582214355
    },
    {
      "auxiliary_loss_clip": 0.0116549,
      "auxiliary_loss_mlp": 0.0076232,
      "balance_loss_clip": 1.05027652,
      "balance_loss_mlp": 1.00065231,
      "epoch": 0.7239824445379667,
      "flos": 25739117020800.0,
      "grad_norm": 1.8336968040648784,
      "language_loss": 0.79968673,
      "learning_rate": 7.471998396736579e-07,
      "loss": 0.81896484,
      "num_input_tokens_seen": 129475700,
      "step": 6021,
      "time_per_iteration": 2.679194927215576
    },
    {
      "auxiliary_loss_clip": 0.01119,
      "auxiliary_loss_mlp": 0.01026193,
      "balance_loss_clip": 1.05010045,
      "balance_loss_mlp": 1.01775908,
      "epoch": 0.7241026874286057,
      "flos": 23148916398720.0,
      "grad_norm": 1.9037001700682208,
      "language_loss": 0.76194584,
      "learning_rate": 7.465927249704549e-07,
      "loss": 0.78339779,
      "num_input_tokens_seen": 129493585,
      "step": 6022,
      "time_per_iteration": 2.7994608879089355
    },
    {
      "auxiliary_loss_clip": 0.0113444,
      "auxiliary_loss_mlp": 0.00762212,
      "balance_loss_clip": 1.04698658,
      "balance_loss_mlp": 1.00068426,
      "epoch": 0.7242229303192449,
      "flos": 20266905686400.0,
      "grad_norm": 1.7915205423201568,
      "language_loss": 0.78058481,
      "learning_rate": 7.459858004081398e-07,
      "loss": 0.79955125,
      "num_input_tokens_seen": 129511555,
      "step": 6023,
      "time_per_iteration": 2.7361936569213867
    },
    {
      "auxiliary_loss_clip": 0.01048524,
      "auxiliary_loss_mlp": 0.0100281,
      "balance_loss_clip": 1.01110268,
      "balance_loss_mlp": 1.00188601,
      "epoch": 0.724343173209884,
      "flos": 62311659684480.0,
      "grad_norm": 0.6521089106850416,
      "language_loss": 0.57938659,
      "learning_rate": 7.453790660787815e-07,
      "loss": 0.59989995,
      "num_input_tokens_seen": 129579650,
      "step": 6024,
      "time_per_iteration": 3.3909544944763184
    },
    {
      "auxiliary_loss_clip": 0.01148154,
      "auxiliary_loss_mlp": 0.01028689,
      "balance_loss_clip": 1.05033445,
      "balance_loss_mlp": 1.01984954,
      "epoch": 0.724463416100523,
      "flos": 35006403813120.0,
      "grad_norm": 2.3956970331629432,
      "language_loss": 0.63602853,
      "learning_rate": 7.447725220744214e-07,
      "loss": 0.65779698,
      "num_input_tokens_seen": 129601895,
      "step": 6025,
      "time_per_iteration": 3.8576576709747314
    },
    {
      "auxiliary_loss_clip": 0.01161089,
      "auxiliary_loss_mlp": 0.01024611,
      "balance_loss_clip": 1.04969156,
      "balance_loss_mlp": 1.01660037,
      "epoch": 0.7245836589911622,
      "flos": 21871968923520.0,
      "grad_norm": 2.4643984460361565,
      "language_loss": 0.77549469,
      "learning_rate": 7.441661684870717e-07,
      "loss": 0.79735166,
      "num_input_tokens_seen": 129622150,
      "step": 6026,
      "time_per_iteration": 2.672684907913208
    },
    {
      "auxiliary_loss_clip": 0.01148917,
      "auxiliary_loss_mlp": 0.01031164,
      "balance_loss_clip": 1.04664981,
      "balance_loss_mlp": 1.02352834,
      "epoch": 0.7247039018818012,
      "flos": 23006494972800.0,
      "grad_norm": 1.7294167973409977,
      "language_loss": 0.81985492,
      "learning_rate": 7.435600054087152e-07,
      "loss": 0.84165573,
      "num_input_tokens_seen": 129644315,
      "step": 6027,
      "time_per_iteration": 2.7719759941101074
    },
    {
      "auxiliary_loss_clip": 0.01157266,
      "auxiliary_loss_mlp": 0.01031091,
      "balance_loss_clip": 1.05013692,
      "balance_loss_mlp": 1.02275825,
      "epoch": 0.7248241447724403,
      "flos": 31722588587520.0,
      "grad_norm": 2.1176449679534404,
      "language_loss": 0.74598879,
      "learning_rate": 7.42954032931308e-07,
      "loss": 0.76787239,
      "num_input_tokens_seen": 129665355,
      "step": 6028,
      "time_per_iteration": 3.694380760192871
    },
    {
      "auxiliary_loss_clip": 0.01114633,
      "auxiliary_loss_mlp": 0.01034292,
      "balance_loss_clip": 1.04298735,
      "balance_loss_mlp": 1.02644825,
      "epoch": 0.7249443876630794,
      "flos": 34896984007680.0,
      "grad_norm": 2.2929331095008947,
      "language_loss": 0.74419838,
      "learning_rate": 7.423482511467733e-07,
      "loss": 0.76568764,
      "num_input_tokens_seen": 129686125,
      "step": 6029,
      "time_per_iteration": 2.838621139526367
    },
    {
      "auxiliary_loss_clip": 0.01102726,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.04563415,
      "balance_loss_mlp": 1.02047658,
      "epoch": 0.7250646305537185,
      "flos": 26359294268160.0,
      "grad_norm": 1.9726361764998253,
      "language_loss": 0.64481562,
      "learning_rate": 7.417426601470099e-07,
      "loss": 0.6661256,
      "num_input_tokens_seen": 129706485,
      "step": 6030,
      "time_per_iteration": 2.857137441635132
    },
    {
      "auxiliary_loss_clip": 0.01126954,
      "auxiliary_loss_mlp": 0.01029753,
      "balance_loss_clip": 1.04502749,
      "balance_loss_mlp": 1.02096128,
      "epoch": 0.7251848734443576,
      "flos": 30081614728320.0,
      "grad_norm": 2.3821126302674833,
      "language_loss": 0.78484619,
      "learning_rate": 7.411372600238841e-07,
      "loss": 0.80641323,
      "num_input_tokens_seen": 129727100,
      "step": 6031,
      "time_per_iteration": 2.779844045639038
    },
    {
      "auxiliary_loss_clip": 0.01115052,
      "auxiliary_loss_mlp": 0.01027439,
      "balance_loss_clip": 1.04946661,
      "balance_loss_mlp": 1.01971424,
      "epoch": 0.7253051163349967,
      "flos": 17785262943360.0,
      "grad_norm": 2.548333140878703,
      "language_loss": 0.74217528,
      "learning_rate": 7.405320508692346e-07,
      "loss": 0.76360023,
      "num_input_tokens_seen": 129745840,
      "step": 6032,
      "time_per_iteration": 2.741956949234009
    },
    {
      "auxiliary_loss_clip": 0.01131844,
      "auxiliary_loss_mlp": 0.01031371,
      "balance_loss_clip": 1.0494957,
      "balance_loss_mlp": 1.02378011,
      "epoch": 0.7254253592256358,
      "flos": 12641346938880.0,
      "grad_norm": 3.254868415428386,
      "language_loss": 0.7547735,
      "learning_rate": 7.399270327748727e-07,
      "loss": 0.77640569,
      "num_input_tokens_seen": 129763500,
      "step": 6033,
      "time_per_iteration": 2.7002816200256348
    },
    {
      "auxiliary_loss_clip": 0.01103499,
      "auxiliary_loss_mlp": 0.01033998,
      "balance_loss_clip": 1.04416013,
      "balance_loss_mlp": 1.02643466,
      "epoch": 0.7255456021162748,
      "flos": 27199208966400.0,
      "grad_norm": 1.9026438200869837,
      "language_loss": 0.74440682,
      "learning_rate": 7.39322205832577e-07,
      "loss": 0.76578176,
      "num_input_tokens_seen": 129784390,
      "step": 6034,
      "time_per_iteration": 2.8803040981292725
    },
    {
      "auxiliary_loss_clip": 0.0116007,
      "auxiliary_loss_mlp": 0.00761717,
      "balance_loss_clip": 1.05104017,
      "balance_loss_mlp": 1.00066876,
      "epoch": 0.725665845006914,
      "flos": 21288205088640.0,
      "grad_norm": 1.964117969411425,
      "language_loss": 0.81164086,
      "learning_rate": 7.387175701341009e-07,
      "loss": 0.83085871,
      "num_input_tokens_seen": 129803060,
      "step": 6035,
      "time_per_iteration": 2.7267613410949707
    },
    {
      "auxiliary_loss_clip": 0.01140265,
      "auxiliary_loss_mlp": 0.01026267,
      "balance_loss_clip": 1.04461312,
      "balance_loss_mlp": 1.01822615,
      "epoch": 0.7257860878975531,
      "flos": 16033684129920.0,
      "grad_norm": 2.2332463808222216,
      "language_loss": 0.72462183,
      "learning_rate": 7.381131257711659e-07,
      "loss": 0.74628711,
      "num_input_tokens_seen": 129820165,
      "step": 6036,
      "time_per_iteration": 3.6514010429382324
    },
    {
      "auxiliary_loss_clip": 0.01133465,
      "auxiliary_loss_mlp": 0.01033883,
      "balance_loss_clip": 1.04415298,
      "balance_loss_mlp": 1.0262239,
      "epoch": 0.7259063307881921,
      "flos": 12129943052160.0,
      "grad_norm": 1.9787021082804446,
      "language_loss": 0.83851516,
      "learning_rate": 7.375088728354677e-07,
      "loss": 0.8601886,
      "num_input_tokens_seen": 129835195,
      "step": 6037,
      "time_per_iteration": 2.8084986209869385
    },
    {
      "auxiliary_loss_clip": 0.01096045,
      "auxiliary_loss_mlp": 0.01029143,
      "balance_loss_clip": 1.04254866,
      "balance_loss_mlp": 1.02116156,
      "epoch": 0.7260265736788313,
      "flos": 30443845432320.0,
      "grad_norm": 1.9959693893644601,
      "language_loss": 0.6739983,
      "learning_rate": 7.369048114186691e-07,
      "loss": 0.69525015,
      "num_input_tokens_seen": 129856240,
      "step": 6038,
      "time_per_iteration": 2.867002010345459
    },
    {
      "auxiliary_loss_clip": 0.0117569,
      "auxiliary_loss_mlp": 0.00761591,
      "balance_loss_clip": 1.05406559,
      "balance_loss_mlp": 1.00088906,
      "epoch": 0.7261468165694703,
      "flos": 21142264129920.0,
      "grad_norm": 1.7570623302256227,
      "language_loss": 0.832304,
      "learning_rate": 7.363009416124055e-07,
      "loss": 0.85167682,
      "num_input_tokens_seen": 129875565,
      "step": 6039,
      "time_per_iteration": 2.735856056213379
    },
    {
      "auxiliary_loss_clip": 0.01135147,
      "auxiliary_loss_mlp": 0.01035698,
      "balance_loss_clip": 1.04514408,
      "balance_loss_mlp": 1.02749074,
      "epoch": 0.7262670594601094,
      "flos": 22306308180480.0,
      "grad_norm": 2.2602359613987972,
      "language_loss": 0.63094258,
      "learning_rate": 7.356972635082852e-07,
      "loss": 0.65265101,
      "num_input_tokens_seen": 129894420,
      "step": 6040,
      "time_per_iteration": 2.6898953914642334
    },
    {
      "auxiliary_loss_clip": 0.01095765,
      "auxiliary_loss_mlp": 0.01025321,
      "balance_loss_clip": 1.04279637,
      "balance_loss_mlp": 1.01720285,
      "epoch": 0.7263873023507486,
      "flos": 25335049950720.0,
      "grad_norm": 1.9475832751290247,
      "language_loss": 0.75586224,
      "learning_rate": 7.35093777197884e-07,
      "loss": 0.77707303,
      "num_input_tokens_seen": 129914490,
      "step": 6041,
      "time_per_iteration": 3.0125691890716553
    },
    {
      "auxiliary_loss_clip": 0.01134298,
      "auxiliary_loss_mlp": 0.01033313,
      "balance_loss_clip": 1.04707313,
      "balance_loss_mlp": 1.02577901,
      "epoch": 0.7265075452413876,
      "flos": 23878621192320.0,
      "grad_norm": 2.2438975166087953,
      "language_loss": 0.85936987,
      "learning_rate": 7.344904827727525e-07,
      "loss": 0.88104606,
      "num_input_tokens_seen": 129931670,
      "step": 6042,
      "time_per_iteration": 3.0183253288269043
    },
    {
      "auxiliary_loss_clip": 0.0114451,
      "auxiliary_loss_mlp": 0.01023386,
      "balance_loss_clip": 1.05144477,
      "balance_loss_mlp": 1.0151788,
      "epoch": 0.7266277881320267,
      "flos": 28724549967360.0,
      "grad_norm": 2.3904328388127225,
      "language_loss": 0.73830664,
      "learning_rate": 7.338873803244076e-07,
      "loss": 0.75998563,
      "num_input_tokens_seen": 129946905,
      "step": 6043,
      "time_per_iteration": 2.757354497909546
    },
    {
      "auxiliary_loss_clip": 0.01130667,
      "auxiliary_loss_mlp": 0.01023717,
      "balance_loss_clip": 1.04793429,
      "balance_loss_mlp": 1.01589382,
      "epoch": 0.7267480310226658,
      "flos": 24863507182080.0,
      "grad_norm": 1.7575999902685562,
      "language_loss": 0.80850118,
      "learning_rate": 7.332844699443401e-07,
      "loss": 0.83004498,
      "num_input_tokens_seen": 129965505,
      "step": 6044,
      "time_per_iteration": 2.809776544570923
    },
    {
      "auxiliary_loss_clip": 0.01150768,
      "auxiliary_loss_mlp": 0.0076151,
      "balance_loss_clip": 1.05241251,
      "balance_loss_mlp": 1.00074232,
      "epoch": 0.7268682739133049,
      "flos": 27198490694400.0,
      "grad_norm": 2.4080340920935295,
      "language_loss": 0.75642651,
      "learning_rate": 7.326817517240121e-07,
      "loss": 0.77554929,
      "num_input_tokens_seen": 129987210,
      "step": 6045,
      "time_per_iteration": 2.7488033771514893
    },
    {
      "auxiliary_loss_clip": 0.01146414,
      "auxiliary_loss_mlp": 0.01025689,
      "balance_loss_clip": 1.04996872,
      "balance_loss_mlp": 1.01869464,
      "epoch": 0.7269885168039439,
      "flos": 33508138688640.0,
      "grad_norm": 1.9080242778828977,
      "language_loss": 0.83392531,
      "learning_rate": 7.320792257548545e-07,
      "loss": 0.85564631,
      "num_input_tokens_seen": 130008385,
      "step": 6046,
      "time_per_iteration": 2.845201253890991
    },
    {
      "auxiliary_loss_clip": 0.01145427,
      "auxiliary_loss_mlp": 0.01025994,
      "balance_loss_clip": 1.05129313,
      "balance_loss_mlp": 1.01756597,
      "epoch": 0.7271087596945831,
      "flos": 24313750548480.0,
      "grad_norm": 5.103017280902605,
      "language_loss": 0.76183808,
      "learning_rate": 7.314768921282704e-07,
      "loss": 0.78355229,
      "num_input_tokens_seen": 130029040,
      "step": 6047,
      "time_per_iteration": 2.7376983165740967
    },
    {
      "auxiliary_loss_clip": 0.0110987,
      "auxiliary_loss_mlp": 0.01029929,
      "balance_loss_clip": 1.04501462,
      "balance_loss_mlp": 1.02192128,
      "epoch": 0.7272290025852222,
      "flos": 23805147922560.0,
      "grad_norm": 3.0555269554804863,
      "language_loss": 0.73004168,
      "learning_rate": 7.30874750935633e-07,
      "loss": 0.75143969,
      "num_input_tokens_seen": 130048725,
      "step": 6048,
      "time_per_iteration": 2.8024959564208984
    },
    {
      "auxiliary_loss_clip": 0.01162238,
      "auxiliary_loss_mlp": 0.01025689,
      "balance_loss_clip": 1.05208206,
      "balance_loss_mlp": 1.01810765,
      "epoch": 0.7273492454758612,
      "flos": 16720367408640.0,
      "grad_norm": 1.8981740941381051,
      "language_loss": 0.79290628,
      "learning_rate": 7.30272802268286e-07,
      "loss": 0.81478554,
      "num_input_tokens_seen": 130065720,
      "step": 6049,
      "time_per_iteration": 2.6331145763397217
    },
    {
      "auxiliary_loss_clip": 0.01089573,
      "auxiliary_loss_mlp": 0.01024406,
      "balance_loss_clip": 1.04265046,
      "balance_loss_mlp": 1.01714027,
      "epoch": 0.7274694883665004,
      "flos": 28031330413440.0,
      "grad_norm": 1.809936629504744,
      "language_loss": 0.76477706,
      "learning_rate": 7.29671046217547e-07,
      "loss": 0.78591686,
      "num_input_tokens_seen": 130084830,
      "step": 6050,
      "time_per_iteration": 3.832097053527832
    },
    {
      "auxiliary_loss_clip": 0.01163148,
      "auxiliary_loss_mlp": 0.00761673,
      "balance_loss_clip": 1.05032182,
      "balance_loss_mlp": 1.00080979,
      "epoch": 0.7275897312571394,
      "flos": 30372706546560.0,
      "grad_norm": 2.41563586052319,
      "language_loss": 0.81864071,
      "learning_rate": 7.290694828746988e-07,
      "loss": 0.83788896,
      "num_input_tokens_seen": 130104495,
      "step": 6051,
      "time_per_iteration": 3.5643744468688965
    },
    {
      "auxiliary_loss_clip": 0.01166028,
      "auxiliary_loss_mlp": 0.01027647,
      "balance_loss_clip": 1.05470538,
      "balance_loss_mlp": 1.0199281,
      "epoch": 0.7277099741477785,
      "flos": 19204775498880.0,
      "grad_norm": 2.261313910656103,
      "language_loss": 0.85946894,
      "learning_rate": 7.284681123310004e-07,
      "loss": 0.88140571,
      "num_input_tokens_seen": 130123210,
      "step": 6052,
      "time_per_iteration": 2.706758975982666
    },
    {
      "auxiliary_loss_clip": 0.01165705,
      "auxiliary_loss_mlp": 0.01028638,
      "balance_loss_clip": 1.05058551,
      "balance_loss_mlp": 1.02060008,
      "epoch": 0.7278302170384175,
      "flos": 20667884186880.0,
      "grad_norm": 2.3374229838996707,
      "language_loss": 0.79923636,
      "learning_rate": 7.27866934677678e-07,
      "loss": 0.82117975,
      "num_input_tokens_seen": 130142880,
      "step": 6053,
      "time_per_iteration": 2.6585381031036377
    },
    {
      "auxiliary_loss_clip": 0.01151063,
      "auxiliary_loss_mlp": 0.01024068,
      "balance_loss_clip": 1.04923391,
      "balance_loss_mlp": 1.01704621,
      "epoch": 0.7279504599290567,
      "flos": 19093200877440.0,
      "grad_norm": 5.4973663330689835,
      "language_loss": 0.78189147,
      "learning_rate": 7.272659500059297e-07,
      "loss": 0.80364275,
      "num_input_tokens_seen": 130160220,
      "step": 6054,
      "time_per_iteration": 3.71710467338562
    },
    {
      "auxiliary_loss_clip": 0.01157839,
      "auxiliary_loss_mlp": 0.01029306,
      "balance_loss_clip": 1.05015779,
      "balance_loss_mlp": 1.02121162,
      "epoch": 0.7280707028196958,
      "flos": 19062174504960.0,
      "grad_norm": 5.845053415050293,
      "language_loss": 0.80528259,
      "learning_rate": 7.266651584069264e-07,
      "loss": 0.82715398,
      "num_input_tokens_seen": 130177885,
      "step": 6055,
      "time_per_iteration": 2.625210762023926
    },
    {
      "auxiliary_loss_clip": 0.01164289,
      "auxiliary_loss_mlp": 0.01031001,
      "balance_loss_clip": 1.05484915,
      "balance_loss_mlp": 1.02298999,
      "epoch": 0.7281909457103348,
      "flos": 37196308293120.0,
      "grad_norm": 4.371744915125488,
      "language_loss": 0.5714556,
      "learning_rate": 7.260645599718045e-07,
      "loss": 0.59340852,
      "num_input_tokens_seen": 130204240,
      "step": 6056,
      "time_per_iteration": 2.8418920040130615
    },
    {
      "auxiliary_loss_clip": 0.01108597,
      "auxiliary_loss_mlp": 0.00763087,
      "balance_loss_clip": 1.04762554,
      "balance_loss_mlp": 1.00086546,
      "epoch": 0.728311188600974,
      "flos": 20667094087680.0,
      "grad_norm": 2.9290502525504,
      "language_loss": 0.67802757,
      "learning_rate": 7.254641547916767e-07,
      "loss": 0.69674438,
      "num_input_tokens_seen": 130221735,
      "step": 6057,
      "time_per_iteration": 2.7980146408081055
    },
    {
      "auxiliary_loss_clip": 0.01164619,
      "auxiliary_loss_mlp": 0.00762503,
      "balance_loss_clip": 1.05422139,
      "balance_loss_mlp": 1.00074673,
      "epoch": 0.728431431491613,
      "flos": 28840685616000.0,
      "grad_norm": 1.8447431283663513,
      "language_loss": 0.6902833,
      "learning_rate": 7.248639429576226e-07,
      "loss": 0.70955455,
      "num_input_tokens_seen": 130241190,
      "step": 6058,
      "time_per_iteration": 2.7142128944396973
    },
    {
      "auxiliary_loss_clip": 0.0112872,
      "auxiliary_loss_mlp": 0.01027585,
      "balance_loss_clip": 1.04642558,
      "balance_loss_mlp": 1.01924658,
      "epoch": 0.7285516743822521,
      "flos": 25991856092160.0,
      "grad_norm": 3.113627453987989,
      "language_loss": 0.72314119,
      "learning_rate": 7.242639245606959e-07,
      "loss": 0.74470425,
      "num_input_tokens_seen": 130260980,
      "step": 6059,
      "time_per_iteration": 2.803819417953491
    },
    {
      "auxiliary_loss_clip": 0.01133451,
      "auxiliary_loss_mlp": 0.01026202,
      "balance_loss_clip": 1.04849219,
      "balance_loss_mlp": 1.01868606,
      "epoch": 0.7286719172728913,
      "flos": 16399721675520.0,
      "grad_norm": 2.4757088988499714,
      "language_loss": 0.82714856,
      "learning_rate": 7.236640996919168e-07,
      "loss": 0.84874517,
      "num_input_tokens_seen": 130280025,
      "step": 6060,
      "time_per_iteration": 2.6823699474334717
    },
    {
      "auxiliary_loss_clip": 0.01102964,
      "auxiliary_loss_mlp": 0.01026276,
      "balance_loss_clip": 1.04346943,
      "balance_loss_mlp": 1.0187484,
      "epoch": 0.7287921601635303,
      "flos": 22018161277440.0,
      "grad_norm": 1.6430926094741136,
      "language_loss": 0.70648468,
      "learning_rate": 7.230644684422782e-07,
      "loss": 0.72777712,
      "num_input_tokens_seen": 130300255,
      "step": 6061,
      "time_per_iteration": 2.7943241596221924
    },
    {
      "auxiliary_loss_clip": 0.01140474,
      "auxiliary_loss_mlp": 0.01030689,
      "balance_loss_clip": 1.04647148,
      "balance_loss_mlp": 1.02206397,
      "epoch": 0.7289124030541694,
      "flos": 24600927784320.0,
      "grad_norm": 1.874494975229405,
      "language_loss": 0.81734872,
      "learning_rate": 7.224650309027451e-07,
      "loss": 0.83906037,
      "num_input_tokens_seen": 130320005,
      "step": 6062,
      "time_per_iteration": 3.607860565185547
    },
    {
      "auxiliary_loss_clip": 0.01150547,
      "auxiliary_loss_mlp": 0.01030315,
      "balance_loss_clip": 1.05065811,
      "balance_loss_mlp": 1.02275729,
      "epoch": 0.7290326459448085,
      "flos": 21393638484480.0,
      "grad_norm": 2.3107041886016293,
      "language_loss": 0.69035488,
      "learning_rate": 7.218657871642506e-07,
      "loss": 0.71216351,
      "num_input_tokens_seen": 130338810,
      "step": 6063,
      "time_per_iteration": 2.720930814743042
    },
    {
      "auxiliary_loss_clip": 0.01152517,
      "auxiliary_loss_mlp": 0.01030082,
      "balance_loss_clip": 1.05011964,
      "balance_loss_mlp": 1.02165413,
      "epoch": 0.7291528888354476,
      "flos": 18587686821120.0,
      "grad_norm": 2.9144177391995516,
      "language_loss": 0.6232357,
      "learning_rate": 7.212667373177012e-07,
      "loss": 0.64506173,
      "num_input_tokens_seen": 130353805,
      "step": 6064,
      "time_per_iteration": 2.711069345474243
    },
    {
      "auxiliary_loss_clip": 0.01134519,
      "auxiliary_loss_mlp": 0.01026379,
      "balance_loss_clip": 1.0467217,
      "balance_loss_mlp": 1.01883292,
      "epoch": 0.7292731317260867,
      "flos": 18951066760320.0,
      "grad_norm": 2.0042619126949353,
      "language_loss": 0.75444382,
      "learning_rate": 7.206678814539704e-07,
      "loss": 0.77605283,
      "num_input_tokens_seen": 130372105,
      "step": 6065,
      "time_per_iteration": 2.7271318435668945
    },
    {
      "auxiliary_loss_clip": 0.01163342,
      "auxiliary_loss_mlp": 0.01027489,
      "balance_loss_clip": 1.05102754,
      "balance_loss_mlp": 1.01991296,
      "epoch": 0.7293933746167258,
      "flos": 21067569797760.0,
      "grad_norm": 1.6270841960270543,
      "language_loss": 0.73205388,
      "learning_rate": 7.20069219663904e-07,
      "loss": 0.75396216,
      "num_input_tokens_seen": 130391990,
      "step": 6066,
      "time_per_iteration": 2.706787586212158
    },
    {
      "auxiliary_loss_clip": 0.01174483,
      "auxiliary_loss_mlp": 0.01028955,
      "balance_loss_clip": 1.05083311,
      "balance_loss_mlp": 1.0209558,
      "epoch": 0.7295136175073649,
      "flos": 22453326547200.0,
      "grad_norm": 1.7079092785502918,
      "language_loss": 0.79506218,
      "learning_rate": 7.1947075203832e-07,
      "loss": 0.81709659,
      "num_input_tokens_seen": 130411970,
      "step": 6067,
      "time_per_iteration": 2.633532762527466
    },
    {
      "auxiliary_loss_clip": 0.01023393,
      "auxiliary_loss_mlp": 0.01000471,
      "balance_loss_clip": 1.0125041,
      "balance_loss_mlp": 0.99955863,
      "epoch": 0.7296338603980039,
      "flos": 56125506648960.0,
      "grad_norm": 0.8629968837866347,
      "language_loss": 0.60129374,
      "learning_rate": 7.188724786680049e-07,
      "loss": 0.62153238,
      "num_input_tokens_seen": 130472440,
      "step": 6068,
      "time_per_iteration": 3.3768136501312256
    },
    {
      "auxiliary_loss_clip": 0.011121,
      "auxiliary_loss_mlp": 0.0102598,
      "balance_loss_clip": 1.03993464,
      "balance_loss_mlp": 1.01821351,
      "epoch": 0.7297541032886431,
      "flos": 25228287751680.0,
      "grad_norm": 1.6037042328563678,
      "language_loss": 0.76064986,
      "learning_rate": 7.182743996437162e-07,
      "loss": 0.78203064,
      "num_input_tokens_seen": 130491975,
      "step": 6069,
      "time_per_iteration": 2.7843549251556396
    },
    {
      "auxiliary_loss_clip": 0.01168009,
      "auxiliary_loss_mlp": 0.01027362,
      "balance_loss_clip": 1.05153203,
      "balance_loss_mlp": 1.01889229,
      "epoch": 0.7298743461792822,
      "flos": 26467600752000.0,
      "grad_norm": 2.7435369427302763,
      "language_loss": 0.68844068,
      "learning_rate": 7.176765150561819e-07,
      "loss": 0.71039438,
      "num_input_tokens_seen": 130510580,
      "step": 6070,
      "time_per_iteration": 2.7336199283599854
    },
    {
      "auxiliary_loss_clip": 0.01124729,
      "auxiliary_loss_mlp": 0.01028585,
      "balance_loss_clip": 1.04566395,
      "balance_loss_mlp": 1.02021027,
      "epoch": 0.7299945890699212,
      "flos": 19569053278080.0,
      "grad_norm": 1.943597606085105,
      "language_loss": 0.80143672,
      "learning_rate": 7.170788249961002e-07,
      "loss": 0.82296985,
      "num_input_tokens_seen": 130529090,
      "step": 6071,
      "time_per_iteration": 2.7180190086364746
    },
    {
      "auxiliary_loss_clip": 0.01108137,
      "auxiliary_loss_mlp": 0.01030136,
      "balance_loss_clip": 1.04278982,
      "balance_loss_mlp": 1.02212548,
      "epoch": 0.7301148319605604,
      "flos": 22928963466240.0,
      "grad_norm": 1.7437518380595376,
      "language_loss": 0.87927055,
      "learning_rate": 7.164813295541418e-07,
      "loss": 0.90065324,
      "num_input_tokens_seen": 130548655,
      "step": 6072,
      "time_per_iteration": 2.7534096240997314
    },
    {
      "auxiliary_loss_clip": 0.01098099,
      "auxiliary_loss_mlp": 0.01027902,
      "balance_loss_clip": 1.04368782,
      "balance_loss_mlp": 1.02027559,
      "epoch": 0.7302350748511994,
      "flos": 25369703596800.0,
      "grad_norm": 2.0079440497174086,
      "language_loss": 0.70758718,
      "learning_rate": 7.15884028820944e-07,
      "loss": 0.72884715,
      "num_input_tokens_seen": 130567710,
      "step": 6073,
      "time_per_iteration": 2.817903757095337
    },
    {
      "auxiliary_loss_clip": 0.01163834,
      "auxiliary_loss_mlp": 0.01029506,
      "balance_loss_clip": 1.05071545,
      "balance_loss_mlp": 1.02239501,
      "epoch": 0.7303553177418385,
      "flos": 27819170732160.0,
      "grad_norm": 4.121877457917503,
      "language_loss": 0.60199928,
      "learning_rate": 7.152869228871185e-07,
      "loss": 0.62393266,
      "num_input_tokens_seen": 130590195,
      "step": 6074,
      "time_per_iteration": 2.6751866340637207
    },
    {
      "auxiliary_loss_clip": 0.01138902,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.04952693,
      "balance_loss_mlp": 1.02152383,
      "epoch": 0.7304755606324776,
      "flos": 24426510318720.0,
      "grad_norm": 2.362785767590253,
      "language_loss": 0.72132432,
      "learning_rate": 7.146900118432457e-07,
      "loss": 0.74301183,
      "num_input_tokens_seen": 130609940,
      "step": 6075,
      "time_per_iteration": 2.775803565979004
    },
    {
      "auxiliary_loss_clip": 0.01144219,
      "auxiliary_loss_mlp": 0.00762304,
      "balance_loss_clip": 1.04824424,
      "balance_loss_mlp": 1.00071597,
      "epoch": 0.7305958035231167,
      "flos": 23840483927040.0,
      "grad_norm": 3.049470159940036,
      "language_loss": 0.86024714,
      "learning_rate": 7.140932957798753e-07,
      "loss": 0.87931234,
      "num_input_tokens_seen": 130628380,
      "step": 6076,
      "time_per_iteration": 4.6366400718688965
    },
    {
      "auxiliary_loss_clip": 0.01149855,
      "auxiliary_loss_mlp": 0.01029113,
      "balance_loss_clip": 1.05014396,
      "balance_loss_mlp": 1.02117991,
      "epoch": 0.7307160464137558,
      "flos": 16726939597440.0,
      "grad_norm": 2.4234865572819326,
      "language_loss": 0.71356881,
      "learning_rate": 7.134967747875309e-07,
      "loss": 0.73535848,
      "num_input_tokens_seen": 130646590,
      "step": 6077,
      "time_per_iteration": 2.6193325519561768
    },
    {
      "auxiliary_loss_clip": 0.01132171,
      "auxiliary_loss_mlp": 0.01032924,
      "balance_loss_clip": 1.04555869,
      "balance_loss_mlp": 1.02565873,
      "epoch": 0.7308362893043949,
      "flos": 21798280172160.0,
      "grad_norm": 1.8671434017350157,
      "language_loss": 0.81923282,
      "learning_rate": 7.129004489567014e-07,
      "loss": 0.84088373,
      "num_input_tokens_seen": 130664070,
      "step": 6078,
      "time_per_iteration": 2.7588679790496826
    },
    {
      "auxiliary_loss_clip": 0.01128712,
      "auxiliary_loss_mlp": 0.01027625,
      "balance_loss_clip": 1.04442322,
      "balance_loss_mlp": 1.02003145,
      "epoch": 0.730956532195034,
      "flos": 10707377840640.0,
      "grad_norm": 2.759486651344767,
      "language_loss": 0.77901363,
      "learning_rate": 7.123043183778512e-07,
      "loss": 0.80057698,
      "num_input_tokens_seen": 130681400,
      "step": 6079,
      "time_per_iteration": 2.729884147644043
    },
    {
      "auxiliary_loss_clip": 0.0114728,
      "auxiliary_loss_mlp": 0.01028662,
      "balance_loss_clip": 1.05161798,
      "balance_loss_mlp": 1.0209434,
      "epoch": 0.731076775085673,
      "flos": 19791987039360.0,
      "grad_norm": 2.0460102477120343,
      "language_loss": 0.65427387,
      "learning_rate": 7.117083831414114e-07,
      "loss": 0.67603332,
      "num_input_tokens_seen": 130700675,
      "step": 6080,
      "time_per_iteration": 3.6691198348999023
    },
    {
      "auxiliary_loss_clip": 0.01173719,
      "auxiliary_loss_mlp": 0.01033305,
      "balance_loss_clip": 1.05166698,
      "balance_loss_mlp": 1.02527642,
      "epoch": 0.7311970179763122,
      "flos": 20447033414400.0,
      "grad_norm": 2.1796964059745827,
      "language_loss": 0.69927371,
      "learning_rate": 7.11112643337787e-07,
      "loss": 0.72134399,
      "num_input_tokens_seen": 130719720,
      "step": 6081,
      "time_per_iteration": 2.7945642471313477
    },
    {
      "auxiliary_loss_clip": 0.01116167,
      "auxiliary_loss_mlp": 0.0102862,
      "balance_loss_clip": 1.04586506,
      "balance_loss_mlp": 1.01992345,
      "epoch": 0.7313172608669513,
      "flos": 18513818501760.0,
      "grad_norm": 2.5059453820889486,
      "language_loss": 0.76882625,
      "learning_rate": 7.10517099057349e-07,
      "loss": 0.79027408,
      "num_input_tokens_seen": 130736670,
      "step": 6082,
      "time_per_iteration": 2.7367331981658936
    },
    {
      "auxiliary_loss_clip": 0.01148885,
      "auxiliary_loss_mlp": 0.01028813,
      "balance_loss_clip": 1.05010152,
      "balance_loss_mlp": 1.02005148,
      "epoch": 0.7314375037575903,
      "flos": 16180738410240.0,
      "grad_norm": 2.769385543172359,
      "language_loss": 0.6124292,
      "learning_rate": 7.099217503904411e-07,
      "loss": 0.63420618,
      "num_input_tokens_seen": 130754525,
      "step": 6083,
      "time_per_iteration": 2.8454997539520264
    },
    {
      "auxiliary_loss_clip": 0.01141564,
      "auxiliary_loss_mlp": 0.01025684,
      "balance_loss_clip": 1.0497905,
      "balance_loss_mlp": 1.01840079,
      "epoch": 0.7315577466482295,
      "flos": 17967940536960.0,
      "grad_norm": 1.8338679577092851,
      "language_loss": 0.90021205,
      "learning_rate": 7.093265974273788e-07,
      "loss": 0.92188454,
      "num_input_tokens_seen": 130772420,
      "step": 6084,
      "time_per_iteration": 2.6940722465515137
    },
    {
      "auxiliary_loss_clip": 0.01160575,
      "auxiliary_loss_mlp": 0.01025925,
      "balance_loss_clip": 1.04795265,
      "balance_loss_mlp": 1.01875722,
      "epoch": 0.7316779895388685,
      "flos": 18405440190720.0,
      "grad_norm": 1.8465725764925849,
      "language_loss": 0.72095954,
      "learning_rate": 7.087316402584447e-07,
      "loss": 0.74282449,
      "num_input_tokens_seen": 130791245,
      "step": 6085,
      "time_per_iteration": 2.6968812942504883
    },
    {
      "auxiliary_loss_clip": 0.01135623,
      "auxiliary_loss_mlp": 0.01027307,
      "balance_loss_clip": 1.04499769,
      "balance_loss_mlp": 1.01964152,
      "epoch": 0.7317982324295076,
      "flos": 17928294900480.0,
      "grad_norm": 1.9530800856649386,
      "language_loss": 0.86688095,
      "learning_rate": 7.081368789738953e-07,
      "loss": 0.88851023,
      "num_input_tokens_seen": 130808445,
      "step": 6086,
      "time_per_iteration": 2.6837728023529053
    },
    {
      "auxiliary_loss_clip": 0.01132052,
      "auxiliary_loss_mlp": 0.01020585,
      "balance_loss_clip": 1.04540658,
      "balance_loss_mlp": 1.01300037,
      "epoch": 0.7319184753201466,
      "flos": 27229840289280.0,
      "grad_norm": 1.9406434370606882,
      "language_loss": 0.77162004,
      "learning_rate": 7.075423136639537e-07,
      "loss": 0.79314637,
      "num_input_tokens_seen": 130827700,
      "step": 6087,
      "time_per_iteration": 2.778350353240967
    },
    {
      "auxiliary_loss_clip": 0.01133042,
      "auxiliary_loss_mlp": 0.01032552,
      "balance_loss_clip": 1.0441637,
      "balance_loss_mlp": 1.02422523,
      "epoch": 0.7320387182107858,
      "flos": 37448544574080.0,
      "grad_norm": 2.015923048174141,
      "language_loss": 0.74856317,
      "learning_rate": 7.069479444188149e-07,
      "loss": 0.77021909,
      "num_input_tokens_seen": 130848290,
      "step": 6088,
      "time_per_iteration": 3.8111705780029297
    },
    {
      "auxiliary_loss_clip": 0.01177687,
      "auxiliary_loss_mlp": 0.01026179,
      "balance_loss_clip": 1.05421078,
      "balance_loss_mlp": 1.01855564,
      "epoch": 0.7321589611014249,
      "flos": 17859023521920.0,
      "grad_norm": 1.7485341611218665,
      "language_loss": 0.82217765,
      "learning_rate": 7.063537713286453e-07,
      "loss": 0.84421635,
      "num_input_tokens_seen": 130865970,
      "step": 6089,
      "time_per_iteration": 2.6551969051361084
    },
    {
      "auxiliary_loss_clip": 0.01145667,
      "auxiliary_loss_mlp": 0.01027615,
      "balance_loss_clip": 1.04943562,
      "balance_loss_mlp": 1.01958668,
      "epoch": 0.7322792039920639,
      "flos": 26100593539200.0,
      "grad_norm": 2.0169618302680887,
      "language_loss": 0.8110286,
      "learning_rate": 7.057597944835803e-07,
      "loss": 0.83276141,
      "num_input_tokens_seen": 130885245,
      "step": 6090,
      "time_per_iteration": 2.747502326965332
    },
    {
      "auxiliary_loss_clip": 0.01111202,
      "auxiliary_loss_mlp": 0.01035123,
      "balance_loss_clip": 1.04528952,
      "balance_loss_mlp": 1.02744603,
      "epoch": 0.7323994468827031,
      "flos": 25369093065600.0,
      "grad_norm": 1.7136965103345205,
      "language_loss": 0.74968028,
      "learning_rate": 7.051660139737253e-07,
      "loss": 0.77114356,
      "num_input_tokens_seen": 130903465,
      "step": 6091,
      "time_per_iteration": 2.7580223083496094
    },
    {
      "auxiliary_loss_clip": 0.01177815,
      "auxiliary_loss_mlp": 0.0103169,
      "balance_loss_clip": 1.05414879,
      "balance_loss_mlp": 1.02409005,
      "epoch": 0.7325196897733421,
      "flos": 26907075653760.0,
      "grad_norm": 1.905547665222413,
      "language_loss": 0.77227503,
      "learning_rate": 7.045724298891565e-07,
      "loss": 0.79437006,
      "num_input_tokens_seen": 130922935,
      "step": 6092,
      "time_per_iteration": 2.683521270751953
    },
    {
      "auxiliary_loss_clip": 0.01065668,
      "auxiliary_loss_mlp": 0.01028557,
      "balance_loss_clip": 1.03932595,
      "balance_loss_mlp": 1.02072477,
      "epoch": 0.7326399326639812,
      "flos": 25775781828480.0,
      "grad_norm": 2.1334539885227137,
      "language_loss": 0.69174147,
      "learning_rate": 7.039790423199192e-07,
      "loss": 0.71268368,
      "num_input_tokens_seen": 130942575,
      "step": 6093,
      "time_per_iteration": 3.0980799198150635
    },
    {
      "auxiliary_loss_clip": 0.01162753,
      "auxiliary_loss_mlp": 0.01030927,
      "balance_loss_clip": 1.0512588,
      "balance_loss_mlp": 1.02298212,
      "epoch": 0.7327601755546204,
      "flos": 21032269706880.0,
      "grad_norm": 7.165850558929932,
      "language_loss": 0.77757329,
      "learning_rate": 7.033858513560322e-07,
      "loss": 0.79951006,
      "num_input_tokens_seen": 130958870,
      "step": 6094,
      "time_per_iteration": 2.9806196689605713
    },
    {
      "auxiliary_loss_clip": 0.01146879,
      "auxiliary_loss_mlp": 0.01025275,
      "balance_loss_clip": 1.05034828,
      "balance_loss_mlp": 1.01756191,
      "epoch": 0.7328804184452594,
      "flos": 16289224462080.0,
      "grad_norm": 2.685106894195104,
      "language_loss": 0.7613852,
      "learning_rate": 7.027928570874794e-07,
      "loss": 0.78310674,
      "num_input_tokens_seen": 130977060,
      "step": 6095,
      "time_per_iteration": 2.652989387512207
    },
    {
      "auxiliary_loss_clip": 0.01109994,
      "auxiliary_loss_mlp": 0.00762027,
      "balance_loss_clip": 1.04571307,
      "balance_loss_mlp": 1.00073743,
      "epoch": 0.7330006613358985,
      "flos": 17858233422720.0,
      "grad_norm": 2.1322899357618805,
      "language_loss": 0.85677505,
      "learning_rate": 7.022000596042194e-07,
      "loss": 0.87549531,
      "num_input_tokens_seen": 130994160,
      "step": 6096,
      "time_per_iteration": 2.7937633991241455
    },
    {
      "auxiliary_loss_clip": 0.01114996,
      "auxiliary_loss_mlp": 0.01023084,
      "balance_loss_clip": 1.04378462,
      "balance_loss_mlp": 1.01556158,
      "epoch": 0.7331209042265376,
      "flos": 22492074343680.0,
      "grad_norm": 2.406869103923877,
      "language_loss": 0.81833661,
      "learning_rate": 7.016074589961784e-07,
      "loss": 0.83971739,
      "num_input_tokens_seen": 131012725,
      "step": 6097,
      "time_per_iteration": 2.8094992637634277
    },
    {
      "auxiliary_loss_clip": 0.01149522,
      "auxiliary_loss_mlp": 0.01025636,
      "balance_loss_clip": 1.05104053,
      "balance_loss_mlp": 1.01833177,
      "epoch": 0.7332411471171767,
      "flos": 33072757937280.0,
      "grad_norm": 3.556531041587697,
      "language_loss": 0.67281103,
      "learning_rate": 7.01015055353253e-07,
      "loss": 0.69456255,
      "num_input_tokens_seen": 131035150,
      "step": 6098,
      "time_per_iteration": 2.804572105407715
    },
    {
      "auxiliary_loss_clip": 0.01153496,
      "auxiliary_loss_mlp": 0.01028088,
      "balance_loss_clip": 1.05100346,
      "balance_loss_mlp": 1.02008009,
      "epoch": 0.7333613900078157,
      "flos": 22743017735040.0,
      "grad_norm": 6.731599434191949,
      "language_loss": 0.7841289,
      "learning_rate": 7.004228487653123e-07,
      "loss": 0.80594468,
      "num_input_tokens_seen": 131055955,
      "step": 6099,
      "time_per_iteration": 2.7353060245513916
    },
    {
      "auxiliary_loss_clip": 0.01157828,
      "auxiliary_loss_mlp": 0.00761636,
      "balance_loss_clip": 1.05144358,
      "balance_loss_mlp": 1.000651,
      "epoch": 0.7334816328984549,
      "flos": 22346133384960.0,
      "grad_norm": 1.8653663028812217,
      "language_loss": 0.78789532,
      "learning_rate": 6.998308393221906e-07,
      "loss": 0.80709004,
      "num_input_tokens_seen": 131074360,
      "step": 6100,
      "time_per_iteration": 2.652428388595581
    },
    {
      "auxiliary_loss_clip": 0.01127633,
      "auxiliary_loss_mlp": 0.01030854,
      "balance_loss_clip": 1.04743361,
      "balance_loss_mlp": 1.02300966,
      "epoch": 0.733601875789094,
      "flos": 20736149984640.0,
      "grad_norm": 1.9695530917284065,
      "language_loss": 0.71301031,
      "learning_rate": 6.992390271136977e-07,
      "loss": 0.73459518,
      "num_input_tokens_seen": 131090070,
      "step": 6101,
      "time_per_iteration": 3.7610042095184326
    },
    {
      "auxiliary_loss_clip": 0.01134271,
      "auxiliary_loss_mlp": 0.00761821,
      "balance_loss_clip": 1.04682708,
      "balance_loss_mlp": 1.00070536,
      "epoch": 0.733722118679733,
      "flos": 22564362464640.0,
      "grad_norm": 1.823695496849745,
      "language_loss": 0.85613233,
      "learning_rate": 6.986474122296094e-07,
      "loss": 0.87509322,
      "num_input_tokens_seen": 131109185,
      "step": 6102,
      "time_per_iteration": 3.8487911224365234
    },
    {
      "auxiliary_loss_clip": 0.01139035,
      "auxiliary_loss_mlp": 0.01032082,
      "balance_loss_clip": 1.04752624,
      "balance_loss_mlp": 1.02389264,
      "epoch": 0.7338423615703722,
      "flos": 20084192179200.0,
      "grad_norm": 1.8894450276428891,
      "language_loss": 0.72558439,
      "learning_rate": 6.980559947596751e-07,
      "loss": 0.7472955,
      "num_input_tokens_seen": 131127725,
      "step": 6103,
      "time_per_iteration": 2.717949390411377
    },
    {
      "auxiliary_loss_clip": 0.01109206,
      "auxiliary_loss_mlp": 0.01028824,
      "balance_loss_clip": 1.04470634,
      "balance_loss_mlp": 1.02103961,
      "epoch": 0.7339626044610112,
      "flos": 21687675217920.0,
      "grad_norm": 1.9723940156206772,
      "language_loss": 0.76065874,
      "learning_rate": 6.974647747936109e-07,
      "loss": 0.78203905,
      "num_input_tokens_seen": 131146110,
      "step": 6104,
      "time_per_iteration": 2.7775216102600098
    },
    {
      "auxiliary_loss_clip": 0.01107981,
      "auxiliary_loss_mlp": 0.01025328,
      "balance_loss_clip": 1.04682219,
      "balance_loss_mlp": 1.01722789,
      "epoch": 0.7340828473516503,
      "flos": 15268248282240.0,
      "grad_norm": 2.3199207394524244,
      "language_loss": 0.82363796,
      "learning_rate": 6.968737524211039e-07,
      "loss": 0.84497106,
      "num_input_tokens_seen": 131162920,
      "step": 6105,
      "time_per_iteration": 2.7001771926879883
    },
    {
      "auxiliary_loss_clip": 0.01117827,
      "auxiliary_loss_mlp": 0.00762279,
      "balance_loss_clip": 1.04421759,
      "balance_loss_mlp": 1.00075722,
      "epoch": 0.7342030902422895,
      "flos": 22930112701440.0,
      "grad_norm": 1.9364468867359172,
      "language_loss": 0.80570185,
      "learning_rate": 6.962829277318132e-07,
      "loss": 0.824503,
      "num_input_tokens_seen": 131182515,
      "step": 6106,
      "time_per_iteration": 4.136912107467651
    },
    {
      "auxiliary_loss_clip": 0.01149925,
      "auxiliary_loss_mlp": 0.0102474,
      "balance_loss_clip": 1.05030429,
      "balance_loss_mlp": 1.01690805,
      "epoch": 0.7343233331329285,
      "flos": 25847890381440.0,
      "grad_norm": 1.8947109934964415,
      "language_loss": 0.83346015,
      "learning_rate": 6.956923008153652e-07,
      "loss": 0.85520679,
      "num_input_tokens_seen": 131202280,
      "step": 6107,
      "time_per_iteration": 2.8139936923980713
    },
    {
      "auxiliary_loss_clip": 0.0115518,
      "auxiliary_loss_mlp": 0.01029325,
      "balance_loss_clip": 1.0502398,
      "balance_loss_mlp": 1.02188683,
      "epoch": 0.7344435760235676,
      "flos": 18478985287680.0,
      "grad_norm": 2.1091332297543364,
      "language_loss": 0.83857334,
      "learning_rate": 6.951018717613593e-07,
      "loss": 0.86041844,
      "num_input_tokens_seen": 131221295,
      "step": 6108,
      "time_per_iteration": 2.663146734237671
    },
    {
      "auxiliary_loss_clip": 0.01145303,
      "auxiliary_loss_mlp": 0.01027818,
      "balance_loss_clip": 1.04988408,
      "balance_loss_mlp": 1.02036417,
      "epoch": 0.7345638189142067,
      "flos": 17640040256640.0,
      "grad_norm": 2.0736184600326806,
      "language_loss": 0.78622937,
      "learning_rate": 6.945116406593614e-07,
      "loss": 0.80796063,
      "num_input_tokens_seen": 131240150,
      "step": 6109,
      "time_per_iteration": 2.750995635986328
    },
    {
      "auxiliary_loss_clip": 0.01128843,
      "auxiliary_loss_mlp": 0.01030877,
      "balance_loss_clip": 1.04594779,
      "balance_loss_mlp": 1.02274096,
      "epoch": 0.7346840618048458,
      "flos": 20260225756800.0,
      "grad_norm": 2.2996637623851757,
      "language_loss": 0.74212289,
      "learning_rate": 6.939216075989089e-07,
      "loss": 0.76372004,
      "num_input_tokens_seen": 131258080,
      "step": 6110,
      "time_per_iteration": 2.7441978454589844
    },
    {
      "auxiliary_loss_clip": 0.01132051,
      "auxiliary_loss_mlp": 0.01029962,
      "balance_loss_clip": 1.05140424,
      "balance_loss_mlp": 1.02197552,
      "epoch": 0.7348043046954849,
      "flos": 29023183641600.0,
      "grad_norm": 1.8618712619562114,
      "language_loss": 0.66424048,
      "learning_rate": 6.933317726695109e-07,
      "loss": 0.68586057,
      "num_input_tokens_seen": 131279310,
      "step": 6111,
      "time_per_iteration": 2.844248056411743
    },
    {
      "auxiliary_loss_clip": 0.01161008,
      "auxiliary_loss_mlp": 0.0102836,
      "balance_loss_clip": 1.04797435,
      "balance_loss_mlp": 1.0199616,
      "epoch": 0.734924547586124,
      "flos": 17931203902080.0,
      "grad_norm": 3.379488100930561,
      "language_loss": 0.801507,
      "learning_rate": 6.92742135960644e-07,
      "loss": 0.82340074,
      "num_input_tokens_seen": 131297010,
      "step": 6112,
      "time_per_iteration": 2.634476661682129
    },
    {
      "auxiliary_loss_clip": 0.01030849,
      "auxiliary_loss_mlp": 0.0099969,
      "balance_loss_clip": 1.01106977,
      "balance_loss_mlp": 0.99881953,
      "epoch": 0.7350447904767631,
      "flos": 63588319850880.0,
      "grad_norm": 1.0749302566622625,
      "language_loss": 0.55574906,
      "learning_rate": 6.921526975617556e-07,
      "loss": 0.57605445,
      "num_input_tokens_seen": 131356470,
      "step": 6113,
      "time_per_iteration": 3.3264200687408447
    },
    {
      "auxiliary_loss_clip": 0.01149066,
      "auxiliary_loss_mlp": 0.01026927,
      "balance_loss_clip": 1.05086231,
      "balance_loss_mlp": 1.01866579,
      "epoch": 0.7351650333674021,
      "flos": 21580015178880.0,
      "grad_norm": 1.978846909900677,
      "language_loss": 0.75651109,
      "learning_rate": 6.915634575622631e-07,
      "loss": 0.77827108,
      "num_input_tokens_seen": 131374985,
      "step": 6114,
      "time_per_iteration": 3.6800806522369385
    },
    {
      "auxiliary_loss_clip": 0.01147643,
      "auxiliary_loss_mlp": 0.01029583,
      "balance_loss_clip": 1.04957938,
      "balance_loss_mlp": 1.0221231,
      "epoch": 0.7352852762580413,
      "flos": 18186349184640.0,
      "grad_norm": 2.001791697976034,
      "language_loss": 0.7095381,
      "learning_rate": 6.909744160515532e-07,
      "loss": 0.73131037,
      "num_input_tokens_seen": 131393125,
      "step": 6115,
      "time_per_iteration": 2.6463205814361572
    },
    {
      "auxiliary_loss_clip": 0.01147982,
      "auxiliary_loss_mlp": 0.01026334,
      "balance_loss_clip": 1.04778576,
      "balance_loss_mlp": 1.01768541,
      "epoch": 0.7354055191486804,
      "flos": 38910073063680.0,
      "grad_norm": 1.927519597217807,
      "language_loss": 0.69223642,
      "learning_rate": 6.903855731189849e-07,
      "loss": 0.7139796,
      "num_input_tokens_seen": 131415760,
      "step": 6116,
      "time_per_iteration": 2.8418455123901367
    },
    {
      "auxiliary_loss_clip": 0.01154232,
      "auxiliary_loss_mlp": 0.01026071,
      "balance_loss_clip": 1.04951429,
      "balance_loss_mlp": 1.01775014,
      "epoch": 0.7355257620393194,
      "flos": 16289978647680.0,
      "grad_norm": 2.306117813889957,
      "language_loss": 0.81463403,
      "learning_rate": 6.897969288538825e-07,
      "loss": 0.83643705,
      "num_input_tokens_seen": 131433705,
      "step": 6117,
      "time_per_iteration": 2.685767412185669
    },
    {
      "auxiliary_loss_clip": 0.01125107,
      "auxiliary_loss_mlp": 0.01022961,
      "balance_loss_clip": 1.04535306,
      "balance_loss_mlp": 1.01549232,
      "epoch": 0.7356460049299585,
      "flos": 18114240631680.0,
      "grad_norm": 1.6883036890742564,
      "language_loss": 0.81383455,
      "learning_rate": 6.892084833455452e-07,
      "loss": 0.83531523,
      "num_input_tokens_seen": 131453275,
      "step": 6118,
      "time_per_iteration": 2.686663866043091
    },
    {
      "auxiliary_loss_clip": 0.01177765,
      "auxiliary_loss_mlp": 0.01029598,
      "balance_loss_clip": 1.05418301,
      "balance_loss_mlp": 1.0220226,
      "epoch": 0.7357662478205976,
      "flos": 21325193118720.0,
      "grad_norm": 1.7285129373371535,
      "language_loss": 0.83832479,
      "learning_rate": 6.886202366832384e-07,
      "loss": 0.86039841,
      "num_input_tokens_seen": 131474960,
      "step": 6119,
      "time_per_iteration": 2.665717601776123
    },
    {
      "auxiliary_loss_clip": 0.01160378,
      "auxiliary_loss_mlp": 0.01027463,
      "balance_loss_clip": 1.05066991,
      "balance_loss_mlp": 1.01972628,
      "epoch": 0.7358864907112367,
      "flos": 14246841139200.0,
      "grad_norm": 1.8945974275057988,
      "language_loss": 0.74033803,
      "learning_rate": 6.880321889561987e-07,
      "loss": 0.76221645,
      "num_input_tokens_seen": 131492935,
      "step": 6120,
      "time_per_iteration": 2.6132168769836426
    },
    {
      "auxiliary_loss_clip": 0.01126104,
      "auxiliary_loss_mlp": 0.01025284,
      "balance_loss_clip": 1.04682589,
      "balance_loss_mlp": 1.01587892,
      "epoch": 0.7360067336018757,
      "flos": 22309684058880.0,
      "grad_norm": 2.268878344339526,
      "language_loss": 0.66022325,
      "learning_rate": 6.874443402536338e-07,
      "loss": 0.68173712,
      "num_input_tokens_seen": 131512025,
      "step": 6121,
      "time_per_iteration": 2.7853264808654785
    },
    {
      "auxiliary_loss_clip": 0.01119329,
      "auxiliary_loss_mlp": 0.00762549,
      "balance_loss_clip": 1.04292977,
      "balance_loss_mlp": 1.00072551,
      "epoch": 0.7361269764925149,
      "flos": 25554607833600.0,
      "grad_norm": 2.0678287260746444,
      "language_loss": 0.8084749,
      "learning_rate": 6.868566906647177e-07,
      "loss": 0.82729369,
      "num_input_tokens_seen": 131532975,
      "step": 6122,
      "time_per_iteration": 2.7654612064361572
    },
    {
      "auxiliary_loss_clip": 0.01094504,
      "auxiliary_loss_mlp": 0.01031387,
      "balance_loss_clip": 1.04249382,
      "balance_loss_mlp": 1.02312613,
      "epoch": 0.736247219383154,
      "flos": 20376505059840.0,
      "grad_norm": 2.0312037680734965,
      "language_loss": 0.83692265,
      "learning_rate": 6.862692402785984e-07,
      "loss": 0.85818154,
      "num_input_tokens_seen": 131553225,
      "step": 6123,
      "time_per_iteration": 2.9007978439331055
    },
    {
      "auxiliary_loss_clip": 0.01052198,
      "auxiliary_loss_mlp": 0.01002555,
      "balance_loss_clip": 1.0108403,
      "balance_loss_mlp": 1.00167298,
      "epoch": 0.736367462273793,
      "flos": 70339525735680.0,
      "grad_norm": 0.6892229066164762,
      "language_loss": 0.49490485,
      "learning_rate": 6.856819891843899e-07,
      "loss": 0.51545238,
      "num_input_tokens_seen": 131617930,
      "step": 6124,
      "time_per_iteration": 3.328535556793213
    },
    {
      "auxiliary_loss_clip": 0.01177705,
      "auxiliary_loss_mlp": 0.01025158,
      "balance_loss_clip": 1.05308461,
      "balance_loss_mlp": 1.01718855,
      "epoch": 0.7364877051644322,
      "flos": 22412711243520.0,
      "grad_norm": 2.085599037677463,
      "language_loss": 0.72154629,
      "learning_rate": 6.8509493747118e-07,
      "loss": 0.74357498,
      "num_input_tokens_seen": 131636740,
      "step": 6125,
      "time_per_iteration": 2.6076788902282715
    },
    {
      "auxiliary_loss_clip": 0.01120031,
      "auxiliary_loss_mlp": 0.01028341,
      "balance_loss_clip": 1.04480314,
      "balance_loss_mlp": 1.02058661,
      "epoch": 0.7366079480550712,
      "flos": 12130266274560.0,
      "grad_norm": 2.245045956905089,
      "language_loss": 0.88332516,
      "learning_rate": 6.845080852280221e-07,
      "loss": 0.90480888,
      "num_input_tokens_seen": 131653810,
      "step": 6126,
      "time_per_iteration": 2.8439900875091553
    },
    {
      "auxiliary_loss_clip": 0.0111726,
      "auxiliary_loss_mlp": 0.01030766,
      "balance_loss_clip": 1.04394579,
      "balance_loss_mlp": 1.02301764,
      "epoch": 0.7367281909457103,
      "flos": 15049336844160.0,
      "grad_norm": 1.6128562319026432,
      "language_loss": 0.74495125,
      "learning_rate": 6.839214325439409e-07,
      "loss": 0.76643145,
      "num_input_tokens_seen": 131671505,
      "step": 6127,
      "time_per_iteration": 3.603170156478882
    },
    {
      "auxiliary_loss_clip": 0.01042258,
      "auxiliary_loss_mlp": 0.00761661,
      "balance_loss_clip": 1.03214049,
      "balance_loss_mlp": 1.00068378,
      "epoch": 0.7368484338363495,
      "flos": 23510752053120.0,
      "grad_norm": 1.8522538886204538,
      "language_loss": 0.72191763,
      "learning_rate": 6.833349795079327e-07,
      "loss": 0.73995674,
      "num_input_tokens_seen": 131690615,
      "step": 6128,
      "time_per_iteration": 4.106255531311035
    },
    {
      "auxiliary_loss_clip": 0.01142034,
      "auxiliary_loss_mlp": 0.01029079,
      "balance_loss_clip": 1.04695606,
      "balance_loss_mlp": 1.02138126,
      "epoch": 0.7369686767269885,
      "flos": 27417833095680.0,
      "grad_norm": 2.577725370870297,
      "language_loss": 0.68797779,
      "learning_rate": 6.827487262089613e-07,
      "loss": 0.7096889,
      "num_input_tokens_seen": 131711120,
      "step": 6129,
      "time_per_iteration": 3.9721951484680176
    },
    {
      "auxiliary_loss_clip": 0.01065973,
      "auxiliary_loss_mlp": 0.01001044,
      "balance_loss_clip": 1.01053441,
      "balance_loss_mlp": 1.00013208,
      "epoch": 0.7370889196176276,
      "flos": 70293343824000.0,
      "grad_norm": 1.0057434965462082,
      "language_loss": 0.567505,
      "learning_rate": 6.821626727359606e-07,
      "loss": 0.58817518,
      "num_input_tokens_seen": 131776680,
      "step": 6130,
      "time_per_iteration": 3.2787115573883057
    },
    {
      "auxiliary_loss_clip": 0.01132989,
      "auxiliary_loss_mlp": 0.01031442,
      "balance_loss_clip": 1.04797125,
      "balance_loss_mlp": 1.02303815,
      "epoch": 0.7372091625082667,
      "flos": 18040839189120.0,
      "grad_norm": 3.0254376379393997,
      "language_loss": 0.77401334,
      "learning_rate": 6.815768191778348e-07,
      "loss": 0.79565763,
      "num_input_tokens_seen": 131794760,
      "step": 6131,
      "time_per_iteration": 2.7320733070373535
    },
    {
      "auxiliary_loss_clip": 0.01130334,
      "auxiliary_loss_mlp": 0.01028954,
      "balance_loss_clip": 1.04916954,
      "balance_loss_mlp": 1.01967382,
      "epoch": 0.7373294053989058,
      "flos": 33726331854720.0,
      "grad_norm": 2.2726592903461116,
      "language_loss": 0.72893715,
      "learning_rate": 6.809911656234569e-07,
      "loss": 0.75053,
      "num_input_tokens_seen": 131816735,
      "step": 6132,
      "time_per_iteration": 3.8267858028411865
    },
    {
      "auxiliary_loss_clip": 0.01130828,
      "auxiliary_loss_mlp": 0.00762106,
      "balance_loss_clip": 1.04925358,
      "balance_loss_mlp": 1.00063086,
      "epoch": 0.7374496482895448,
      "flos": 21506326427520.0,
      "grad_norm": 2.157256139450834,
      "language_loss": 0.78299862,
      "learning_rate": 6.804057121616707e-07,
      "loss": 0.80192792,
      "num_input_tokens_seen": 131834940,
      "step": 6133,
      "time_per_iteration": 2.705198049545288
    },
    {
      "auxiliary_loss_clip": 0.01165068,
      "auxiliary_loss_mlp": 0.01028559,
      "balance_loss_clip": 1.05077028,
      "balance_loss_mlp": 1.02010119,
      "epoch": 0.737569891180184,
      "flos": 24936908624640.0,
      "grad_norm": 2.1351290653586936,
      "language_loss": 0.72068304,
      "learning_rate": 6.798204588812888e-07,
      "loss": 0.74261934,
      "num_input_tokens_seen": 131854355,
      "step": 6134,
      "time_per_iteration": 2.7260637283325195
    },
    {
      "auxiliary_loss_clip": 0.01175527,
      "auxiliary_loss_mlp": 0.01026201,
      "balance_loss_clip": 1.05249405,
      "balance_loss_mlp": 1.01820183,
      "epoch": 0.7376901340708231,
      "flos": 20664544222080.0,
      "grad_norm": 1.6837320300165628,
      "language_loss": 0.75644171,
      "learning_rate": 6.792354058710937e-07,
      "loss": 0.77845895,
      "num_input_tokens_seen": 131871825,
      "step": 6135,
      "time_per_iteration": 2.6378676891326904
    },
    {
      "auxiliary_loss_clip": 0.01142127,
      "auxiliary_loss_mlp": 0.01028915,
      "balance_loss_clip": 1.04620576,
      "balance_loss_mlp": 1.02176869,
      "epoch": 0.7378103769614621,
      "flos": 23805794367360.0,
      "grad_norm": 1.9524545731704448,
      "language_loss": 0.6547606,
      "learning_rate": 6.786505532198374e-07,
      "loss": 0.67647099,
      "num_input_tokens_seen": 131890770,
      "step": 6136,
      "time_per_iteration": 2.696848154067993
    },
    {
      "auxiliary_loss_clip": 0.01164377,
      "auxiliary_loss_mlp": 0.0102352,
      "balance_loss_clip": 1.05226469,
      "balance_loss_mlp": 1.01543164,
      "epoch": 0.7379306198521013,
      "flos": 22237216369920.0,
      "grad_norm": 1.7784419514349767,
      "language_loss": 0.85196757,
      "learning_rate": 6.780659010162411e-07,
      "loss": 0.87384653,
      "num_input_tokens_seen": 131909720,
      "step": 6137,
      "time_per_iteration": 2.719486713409424
    },
    {
      "auxiliary_loss_clip": 0.01088286,
      "auxiliary_loss_mlp": 0.01029078,
      "balance_loss_clip": 1.04364252,
      "balance_loss_mlp": 1.02173519,
      "epoch": 0.7380508627427403,
      "flos": 14903108576640.0,
      "grad_norm": 3.964468977246311,
      "language_loss": 0.83297789,
      "learning_rate": 6.774814493489975e-07,
      "loss": 0.85415149,
      "num_input_tokens_seen": 131927395,
      "step": 6138,
      "time_per_iteration": 2.9408602714538574
    },
    {
      "auxiliary_loss_clip": 0.01131418,
      "auxiliary_loss_mlp": 0.01023556,
      "balance_loss_clip": 1.04541028,
      "balance_loss_mlp": 1.01628458,
      "epoch": 0.7381711056333794,
      "flos": 21685843624320.0,
      "grad_norm": 1.9832052452587758,
      "language_loss": 0.66493052,
      "learning_rate": 6.768971983067655e-07,
      "loss": 0.68648028,
      "num_input_tokens_seen": 131947725,
      "step": 6139,
      "time_per_iteration": 3.3729286193847656
    },
    {
      "auxiliary_loss_clip": 0.01031267,
      "auxiliary_loss_mlp": 0.007522,
      "balance_loss_clip": 1.0195123,
      "balance_loss_mlp": 1.0001204,
      "epoch": 0.7382913485240186,
      "flos": 52404263596800.0,
      "grad_norm": 1.0081742460545537,
      "language_loss": 0.67719269,
      "learning_rate": 6.763131479781772e-07,
      "loss": 0.69502735,
      "num_input_tokens_seen": 131997485,
      "step": 6140,
      "time_per_iteration": 4.013961553573608
    },
    {
      "auxiliary_loss_clip": 0.01083459,
      "auxiliary_loss_mlp": 0.01026503,
      "balance_loss_clip": 1.04313028,
      "balance_loss_mlp": 1.01895714,
      "epoch": 0.7384115914146576,
      "flos": 21798818876160.0,
      "grad_norm": 2.2042794316781973,
      "language_loss": 0.76558387,
      "learning_rate": 6.757292984518316e-07,
      "loss": 0.78668344,
      "num_input_tokens_seen": 132016885,
      "step": 6141,
      "time_per_iteration": 2.8074135780334473
    },
    {
      "auxiliary_loss_clip": 0.01054804,
      "auxiliary_loss_mlp": 0.01003503,
      "balance_loss_clip": 1.01017404,
      "balance_loss_mlp": 1.00253165,
      "epoch": 0.7385318343052967,
      "flos": 61494331662720.0,
      "grad_norm": 0.7478811896556277,
      "language_loss": 0.56303179,
      "learning_rate": 6.751456498162981e-07,
      "loss": 0.58361489,
      "num_input_tokens_seen": 132075920,
      "step": 6142,
      "time_per_iteration": 3.148869514465332
    },
    {
      "auxiliary_loss_clip": 0.011213,
      "auxiliary_loss_mlp": 0.01026099,
      "balance_loss_clip": 1.04485977,
      "balance_loss_mlp": 1.01866674,
      "epoch": 0.7386520771959358,
      "flos": 17013757697280.0,
      "grad_norm": 1.9589940393167493,
      "language_loss": 0.85632551,
      "learning_rate": 6.745622021601174e-07,
      "loss": 0.87779951,
      "num_input_tokens_seen": 132092945,
      "step": 6143,
      "time_per_iteration": 2.672351598739624
    },
    {
      "auxiliary_loss_clip": 0.01146297,
      "auxiliary_loss_mlp": 0.0102366,
      "balance_loss_clip": 1.04741049,
      "balance_loss_mlp": 1.01553595,
      "epoch": 0.7387723200865749,
      "flos": 18770759464320.0,
      "grad_norm": 2.0872370316369,
      "language_loss": 0.70182747,
      "learning_rate": 6.739789555717954e-07,
      "loss": 0.72352701,
      "num_input_tokens_seen": 132109920,
      "step": 6144,
      "time_per_iteration": 2.7643914222717285
    },
    {
      "auxiliary_loss_clip": 0.01110033,
      "auxiliary_loss_mlp": 0.01028975,
      "balance_loss_clip": 1.04515278,
      "balance_loss_mlp": 1.02124703,
      "epoch": 0.738892562977214,
      "flos": 22525542840960.0,
      "grad_norm": 2.6815783336951227,
      "language_loss": 0.77030241,
      "learning_rate": 6.733959101398124e-07,
      "loss": 0.7916925,
      "num_input_tokens_seen": 132128050,
      "step": 6145,
      "time_per_iteration": 2.788562536239624
    },
    {
      "auxiliary_loss_clip": 0.01149712,
      "auxiliary_loss_mlp": 0.01024284,
      "balance_loss_clip": 1.04635835,
      "balance_loss_mlp": 1.01651144,
      "epoch": 0.7390128058678531,
      "flos": 21501478091520.0,
      "grad_norm": 2.25827907265923,
      "language_loss": 0.81512505,
      "learning_rate": 6.728130659526143e-07,
      "loss": 0.83686501,
      "num_input_tokens_seen": 132145860,
      "step": 6146,
      "time_per_iteration": 2.679067373275757
    },
    {
      "auxiliary_loss_clip": 0.01119927,
      "auxiliary_loss_mlp": 0.01028055,
      "balance_loss_clip": 1.04550207,
      "balance_loss_mlp": 1.02000225,
      "epoch": 0.7391330487584922,
      "flos": 25776176878080.0,
      "grad_norm": 2.283417237796957,
      "language_loss": 0.71492296,
      "learning_rate": 6.7223042309862e-07,
      "loss": 0.73640275,
      "num_input_tokens_seen": 132166060,
      "step": 6147,
      "time_per_iteration": 2.8243143558502197
    },
    {
      "auxiliary_loss_clip": 0.0113876,
      "auxiliary_loss_mlp": 0.01029901,
      "balance_loss_clip": 1.04891753,
      "balance_loss_mlp": 1.0227778,
      "epoch": 0.7392532916491312,
      "flos": 28366736636160.0,
      "grad_norm": 2.704841427349435,
      "language_loss": 0.73867702,
      "learning_rate": 6.716479816662144e-07,
      "loss": 0.76036358,
      "num_input_tokens_seen": 132187790,
      "step": 6148,
      "time_per_iteration": 2.7130343914031982
    },
    {
      "auxiliary_loss_clip": 0.01132529,
      "auxiliary_loss_mlp": 0.0102382,
      "balance_loss_clip": 1.04796696,
      "balance_loss_mlp": 1.01619995,
      "epoch": 0.7393735345397703,
      "flos": 23585877348480.0,
      "grad_norm": 2.0132952985421855,
      "language_loss": 0.73735613,
      "learning_rate": 6.710657417437531e-07,
      "loss": 0.7589196,
      "num_input_tokens_seen": 132207495,
      "step": 6149,
      "time_per_iteration": 2.7597858905792236
    },
    {
      "auxiliary_loss_clip": 0.01129328,
      "auxiliary_loss_mlp": 0.0102349,
      "balance_loss_clip": 1.04620504,
      "balance_loss_mlp": 1.01537824,
      "epoch": 0.7394937774304094,
      "flos": 19974772373760.0,
      "grad_norm": 2.5237183455120036,
      "language_loss": 0.80323166,
      "learning_rate": 6.704837034195628e-07,
      "loss": 0.82475984,
      "num_input_tokens_seen": 132225960,
      "step": 6150,
      "time_per_iteration": 2.6951687335968018
    },
    {
      "auxiliary_loss_clip": 0.01119261,
      "auxiliary_loss_mlp": 0.01032957,
      "balance_loss_clip": 1.0428009,
      "balance_loss_mlp": 1.02473176,
      "epoch": 0.7396140203210485,
      "flos": 23478037741440.0,
      "grad_norm": 2.0104325874571876,
      "language_loss": 0.84943116,
      "learning_rate": 6.699018667819376e-07,
      "loss": 0.87095332,
      "num_input_tokens_seen": 132245360,
      "step": 6151,
      "time_per_iteration": 2.765444755554199
    },
    {
      "auxiliary_loss_clip": 0.01107043,
      "auxiliary_loss_mlp": 0.01030008,
      "balance_loss_clip": 1.03941131,
      "balance_loss_mlp": 1.02158546,
      "epoch": 0.7397342632116876,
      "flos": 25555433846400.0,
      "grad_norm": 1.6288693830435428,
      "language_loss": 0.73013473,
      "learning_rate": 6.693202319191415e-07,
      "loss": 0.75150532,
      "num_input_tokens_seen": 132267095,
      "step": 6152,
      "time_per_iteration": 2.7794370651245117
    },
    {
      "auxiliary_loss_clip": 0.01114641,
      "auxiliary_loss_mlp": 0.01031362,
      "balance_loss_clip": 1.04902327,
      "balance_loss_mlp": 1.02306533,
      "epoch": 0.7398545061023267,
      "flos": 24755021130240.0,
      "grad_norm": 1.9269426320285035,
      "language_loss": 0.74999803,
      "learning_rate": 6.687387989194084e-07,
      "loss": 0.77145803,
      "num_input_tokens_seen": 132286610,
      "step": 6153,
      "time_per_iteration": 3.773995876312256
    },
    {
      "auxiliary_loss_clip": 0.01159006,
      "auxiliary_loss_mlp": 0.01029094,
      "balance_loss_clip": 1.04780078,
      "balance_loss_mlp": 1.02102923,
      "epoch": 0.7399747489929658,
      "flos": 16508602776960.0,
      "grad_norm": 2.6494856456647438,
      "language_loss": 0.7964375,
      "learning_rate": 6.681575678709404e-07,
      "loss": 0.81831855,
      "num_input_tokens_seen": 132305300,
      "step": 6154,
      "time_per_iteration": 3.586608409881592
    },
    {
      "auxiliary_loss_clip": 0.01173523,
      "auxiliary_loss_mlp": 0.01022548,
      "balance_loss_clip": 1.05058694,
      "balance_loss_mlp": 1.01473999,
      "epoch": 0.7400949918836048,
      "flos": 24097065753600.0,
      "grad_norm": 8.136072666419777,
      "language_loss": 0.708103,
      "learning_rate": 6.67576538861911e-07,
      "loss": 0.73006368,
      "num_input_tokens_seen": 132323875,
      "step": 6155,
      "time_per_iteration": 2.6006975173950195
    },
    {
      "auxiliary_loss_clip": 0.01105111,
      "auxiliary_loss_mlp": 0.01027016,
      "balance_loss_clip": 1.04260087,
      "balance_loss_mlp": 1.01924086,
      "epoch": 0.740215234774244,
      "flos": 21802517976960.0,
      "grad_norm": 2.4341421095501614,
      "language_loss": 0.82366979,
      "learning_rate": 6.669957119804612e-07,
      "loss": 0.84499109,
      "num_input_tokens_seen": 132345510,
      "step": 6156,
      "time_per_iteration": 2.8371689319610596
    },
    {
      "auxiliary_loss_clip": 0.01133281,
      "auxiliary_loss_mlp": 0.01025825,
      "balance_loss_clip": 1.0449965,
      "balance_loss_mlp": 1.01770091,
      "epoch": 0.7403354776648831,
      "flos": 18733196816640.0,
      "grad_norm": 13.476730591402731,
      "language_loss": 0.72427797,
      "learning_rate": 6.66415087314702e-07,
      "loss": 0.74586904,
      "num_input_tokens_seen": 132360465,
      "step": 6157,
      "time_per_iteration": 2.6986958980560303
    },
    {
      "auxiliary_loss_clip": 0.0115744,
      "auxiliary_loss_mlp": 0.01029709,
      "balance_loss_clip": 1.05014646,
      "balance_loss_mlp": 1.02145934,
      "epoch": 0.7404557205555221,
      "flos": 16909581277440.0,
      "grad_norm": 2.6581359852081543,
      "language_loss": 0.72945809,
      "learning_rate": 6.65834664952714e-07,
      "loss": 0.7513296,
      "num_input_tokens_seen": 132377915,
      "step": 6158,
      "time_per_iteration": 3.575439691543579
    },
    {
      "auxiliary_loss_clip": 0.01112403,
      "auxiliary_loss_mlp": 0.01020996,
      "balance_loss_clip": 1.04458416,
      "balance_loss_mlp": 1.01343238,
      "epoch": 0.7405759634461613,
      "flos": 21214408596480.0,
      "grad_norm": 1.5853746180098185,
      "language_loss": 0.76270473,
      "learning_rate": 6.652544449825457e-07,
      "loss": 0.78403866,
      "num_input_tokens_seen": 132398170,
      "step": 6159,
      "time_per_iteration": 2.774670124053955
    },
    {
      "auxiliary_loss_clip": 0.01121469,
      "auxiliary_loss_mlp": 0.010278,
      "balance_loss_clip": 1.0426954,
      "balance_loss_mlp": 1.01927066,
      "epoch": 0.7406962063368003,
      "flos": 20480106862080.0,
      "grad_norm": 1.6301747974995868,
      "language_loss": 0.76645243,
      "learning_rate": 6.646744274922182e-07,
      "loss": 0.78794515,
      "num_input_tokens_seen": 132416615,
      "step": 6160,
      "time_per_iteration": 2.76821231842041
    },
    {
      "auxiliary_loss_clip": 0.01173764,
      "auxiliary_loss_mlp": 0.010274,
      "balance_loss_clip": 1.05110788,
      "balance_loss_mlp": 1.01964819,
      "epoch": 0.7408164492274394,
      "flos": 19791915212160.0,
      "grad_norm": 2.8877987373480463,
      "language_loss": 0.76027036,
      "learning_rate": 6.640946125697171e-07,
      "loss": 0.78228199,
      "num_input_tokens_seen": 132434145,
      "step": 6161,
      "time_per_iteration": 2.621497869491577
    },
    {
      "auxiliary_loss_clip": 0.01162442,
      "auxiliary_loss_mlp": 0.01025087,
      "balance_loss_clip": 1.04855776,
      "balance_loss_mlp": 1.01678443,
      "epoch": 0.7409366921180786,
      "flos": 29204855654400.0,
      "grad_norm": 7.242153190123752,
      "language_loss": 0.76226926,
      "learning_rate": 6.635150003030017e-07,
      "loss": 0.78414452,
      "num_input_tokens_seen": 132452670,
      "step": 6162,
      "time_per_iteration": 2.6613752841949463
    },
    {
      "auxiliary_loss_clip": 0.01124322,
      "auxiliary_loss_mlp": 0.01028152,
      "balance_loss_clip": 1.04667616,
      "balance_loss_mlp": 1.02026033,
      "epoch": 0.7410569350087176,
      "flos": 22930004960640.0,
      "grad_norm": 2.316646759571315,
      "language_loss": 0.86289787,
      "learning_rate": 6.629355907799981e-07,
      "loss": 0.88442266,
      "num_input_tokens_seen": 132472475,
      "step": 6163,
      "time_per_iteration": 2.7370595932006836
    },
    {
      "auxiliary_loss_clip": 0.01149428,
      "auxiliary_loss_mlp": 0.01031374,
      "balance_loss_clip": 1.04717684,
      "balance_loss_mlp": 1.02306533,
      "epoch": 0.7411771778993567,
      "flos": 30440397726720.0,
      "grad_norm": 1.9037633857406646,
      "language_loss": 0.69341028,
      "learning_rate": 6.623563840886015e-07,
      "loss": 0.71521831,
      "num_input_tokens_seen": 132493400,
      "step": 6164,
      "time_per_iteration": 2.708921432495117
    },
    {
      "auxiliary_loss_clip": 0.01161829,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.05110753,
      "balance_loss_mlp": 1.02136159,
      "epoch": 0.7412974207899958,
      "flos": 20522050968960.0,
      "grad_norm": 1.6482648230591308,
      "language_loss": 0.6972304,
      "learning_rate": 6.617773803166795e-07,
      "loss": 0.71914494,
      "num_input_tokens_seen": 132511725,
      "step": 6165,
      "time_per_iteration": 3.765838146209717
    },
    {
      "auxiliary_loss_clip": 0.01164981,
      "auxiliary_loss_mlp": 0.01027973,
      "balance_loss_clip": 1.04959559,
      "balance_loss_mlp": 1.01984239,
      "epoch": 0.7414176636806349,
      "flos": 22090700793600.0,
      "grad_norm": 2.053099470709531,
      "language_loss": 0.81917191,
      "learning_rate": 6.611985795520634e-07,
      "loss": 0.84110147,
      "num_input_tokens_seen": 132530270,
      "step": 6166,
      "time_per_iteration": 2.6886696815490723
    },
    {
      "auxiliary_loss_clip": 0.01167409,
      "auxiliary_loss_mlp": 0.01032418,
      "balance_loss_clip": 1.05169046,
      "balance_loss_mlp": 1.02406192,
      "epoch": 0.7415379065712739,
      "flos": 25155245445120.0,
      "grad_norm": 2.677993180994047,
      "language_loss": 0.77452606,
      "learning_rate": 6.606199818825588e-07,
      "loss": 0.79652429,
      "num_input_tokens_seen": 132550725,
      "step": 6167,
      "time_per_iteration": 2.6347670555114746
    },
    {
      "auxiliary_loss_clip": 0.01110243,
      "auxiliary_loss_mlp": 0.01025743,
      "balance_loss_clip": 1.04445624,
      "balance_loss_mlp": 1.01744556,
      "epoch": 0.7416581494619131,
      "flos": 16871731320960.0,
      "grad_norm": 4.394590062650407,
      "language_loss": 0.82008874,
      "learning_rate": 6.600415873959377e-07,
      "loss": 0.84144861,
      "num_input_tokens_seen": 132568600,
      "step": 6168,
      "time_per_iteration": 2.7369072437286377
    },
    {
      "auxiliary_loss_clip": 0.01132952,
      "auxiliary_loss_mlp": 0.01027314,
      "balance_loss_clip": 1.04582787,
      "balance_loss_mlp": 1.02026892,
      "epoch": 0.7417783923525522,
      "flos": 28438881102720.0,
      "grad_norm": 2.0634647901637275,
      "language_loss": 0.64673746,
      "learning_rate": 6.594633961799437e-07,
      "loss": 0.66834009,
      "num_input_tokens_seen": 132587640,
      "step": 6169,
      "time_per_iteration": 2.7252862453460693
    },
    {
      "auxiliary_loss_clip": 0.01143504,
      "auxiliary_loss_mlp": 0.01030078,
      "balance_loss_clip": 1.05014908,
      "balance_loss_mlp": 1.02231193,
      "epoch": 0.7418986352431912,
      "flos": 20084299920000.0,
      "grad_norm": 1.7871464580104803,
      "language_loss": 0.81683624,
      "learning_rate": 6.588854083222857e-07,
      "loss": 0.83857214,
      "num_input_tokens_seen": 132607075,
      "step": 6170,
      "time_per_iteration": 2.778822183609009
    },
    {
      "auxiliary_loss_clip": 0.01137254,
      "auxiliary_loss_mlp": 0.01029428,
      "balance_loss_clip": 1.05003679,
      "balance_loss_mlp": 1.02067232,
      "epoch": 0.7420188781338304,
      "flos": 18259571059200.0,
      "grad_norm": 2.411283238197045,
      "language_loss": 0.80728185,
      "learning_rate": 6.583076239106444e-07,
      "loss": 0.82894862,
      "num_input_tokens_seen": 132625580,
      "step": 6171,
      "time_per_iteration": 2.6596953868865967
    },
    {
      "auxiliary_loss_clip": 0.01161465,
      "auxiliary_loss_mlp": 0.01024393,
      "balance_loss_clip": 1.05208492,
      "balance_loss_mlp": 1.0167098,
      "epoch": 0.7421391210244694,
      "flos": 13771994319360.0,
      "grad_norm": 2.394168189626846,
      "language_loss": 0.75301933,
      "learning_rate": 6.577300430326707e-07,
      "loss": 0.77487791,
      "num_input_tokens_seen": 132640525,
      "step": 6172,
      "time_per_iteration": 2.6491479873657227
    },
    {
      "auxiliary_loss_clip": 0.01111362,
      "auxiliary_loss_mlp": 0.01027261,
      "balance_loss_clip": 1.04577041,
      "balance_loss_mlp": 1.01898789,
      "epoch": 0.7422593639151085,
      "flos": 15961683317760.0,
      "grad_norm": 2.560774949255724,
      "language_loss": 0.723369,
      "learning_rate": 6.571526657759821e-07,
      "loss": 0.74475527,
      "num_input_tokens_seen": 132656265,
      "step": 6173,
      "time_per_iteration": 2.7936196327209473
    },
    {
      "auxiliary_loss_clip": 0.01160829,
      "auxiliary_loss_mlp": 0.01025893,
      "balance_loss_clip": 1.04814422,
      "balance_loss_mlp": 1.01764989,
      "epoch": 0.7423796068057477,
      "flos": 30114400867200.0,
      "grad_norm": 1.8139308139413273,
      "language_loss": 0.70602232,
      "learning_rate": 6.565754922281663e-07,
      "loss": 0.72788954,
      "num_input_tokens_seen": 132678510,
      "step": 6174,
      "time_per_iteration": 2.658409357070923
    },
    {
      "auxiliary_loss_clip": 0.01142615,
      "auxiliary_loss_mlp": 0.01024826,
      "balance_loss_clip": 1.04651356,
      "balance_loss_mlp": 1.0167439,
      "epoch": 0.7424998496963867,
      "flos": 20521907314560.0,
      "grad_norm": 2.1391291364154448,
      "language_loss": 0.78487706,
      "learning_rate": 6.559985224767801e-07,
      "loss": 0.80655146,
      "num_input_tokens_seen": 132696385,
      "step": 6175,
      "time_per_iteration": 2.6869988441467285
    },
    {
      "auxiliary_loss_clip": 0.01099982,
      "auxiliary_loss_mlp": 0.01029858,
      "balance_loss_clip": 1.0424912,
      "balance_loss_mlp": 1.02210307,
      "epoch": 0.7426200925870258,
      "flos": 21871573873920.0,
      "grad_norm": 2.652993914122241,
      "language_loss": 0.75674868,
      "learning_rate": 6.55421756609349e-07,
      "loss": 0.77804708,
      "num_input_tokens_seen": 132714640,
      "step": 6176,
      "time_per_iteration": 2.757506847381592
    },
    {
      "auxiliary_loss_clip": 0.0113791,
      "auxiliary_loss_mlp": 0.01027833,
      "balance_loss_clip": 1.05094683,
      "balance_loss_mlp": 1.02014399,
      "epoch": 0.7427403354776649,
      "flos": 26432049265920.0,
      "grad_norm": 7.7215321920313755,
      "language_loss": 0.78819799,
      "learning_rate": 6.54845194713369e-07,
      "loss": 0.8098554,
      "num_input_tokens_seen": 132735590,
      "step": 6177,
      "time_per_iteration": 2.804875612258911
    },
    {
      "auxiliary_loss_clip": 0.01164634,
      "auxiliary_loss_mlp": 0.01032609,
      "balance_loss_clip": 1.05419755,
      "balance_loss_mlp": 1.02500987,
      "epoch": 0.742860578368304,
      "flos": 19898390102400.0,
      "grad_norm": 3.0012864848953695,
      "language_loss": 0.79999399,
      "learning_rate": 6.542688368763034e-07,
      "loss": 0.82196647,
      "num_input_tokens_seen": 132753995,
      "step": 6178,
      "time_per_iteration": 2.672675371170044
    },
    {
      "auxiliary_loss_clip": 0.01159884,
      "auxiliary_loss_mlp": 0.01032357,
      "balance_loss_clip": 1.05231345,
      "balance_loss_mlp": 1.02495718,
      "epoch": 0.742980821258943,
      "flos": 24827201510400.0,
      "grad_norm": 1.8192800617880258,
      "language_loss": 0.77438426,
      "learning_rate": 6.536926831855854e-07,
      "loss": 0.79630667,
      "num_input_tokens_seen": 132773160,
      "step": 6179,
      "time_per_iteration": 3.635931968688965
    },
    {
      "auxiliary_loss_clip": 0.01145718,
      "auxiliary_loss_mlp": 0.01024867,
      "balance_loss_clip": 1.04771614,
      "balance_loss_mlp": 1.0169214,
      "epoch": 0.7431010641495821,
      "flos": 25228646887680.0,
      "grad_norm": 2.7291058604036453,
      "language_loss": 0.73144454,
      "learning_rate": 6.531167337286165e-07,
      "loss": 0.7531504,
      "num_input_tokens_seen": 132793180,
      "step": 6180,
      "time_per_iteration": 4.091450452804565
    },
    {
      "auxiliary_loss_clip": 0.01160879,
      "auxiliary_loss_mlp": 0.01030881,
      "balance_loss_clip": 1.05292463,
      "balance_loss_mlp": 1.02319217,
      "epoch": 0.7432213070402213,
      "flos": 21762369550080.0,
      "grad_norm": 5.524328616410666,
      "language_loss": 0.80067831,
      "learning_rate": 6.52540988592768e-07,
      "loss": 0.82259595,
      "num_input_tokens_seen": 132814200,
      "step": 6181,
      "time_per_iteration": 2.6616055965423584
    },
    {
      "auxiliary_loss_clip": 0.01125207,
      "auxiliary_loss_mlp": 0.01024332,
      "balance_loss_clip": 1.04958236,
      "balance_loss_mlp": 1.01683998,
      "epoch": 0.7433415499308603,
      "flos": 14793832425600.0,
      "grad_norm": 2.2135874642568036,
      "language_loss": 0.83610761,
      "learning_rate": 6.519654478653814e-07,
      "loss": 0.85760301,
      "num_input_tokens_seen": 132832565,
      "step": 6182,
      "time_per_iteration": 2.730473756790161
    },
    {
      "auxiliary_loss_clip": 0.01025418,
      "auxiliary_loss_mlp": 0.01001565,
      "balance_loss_clip": 1.0236553,
      "balance_loss_mlp": 1.00068319,
      "epoch": 0.7434617928214994,
      "flos": 67155577297920.0,
      "grad_norm": 0.7615845295893796,
      "language_loss": 0.56029379,
      "learning_rate": 6.51390111633763e-07,
      "loss": 0.58056366,
      "num_input_tokens_seen": 132897840,
      "step": 6183,
      "time_per_iteration": 3.4116222858428955
    },
    {
      "auxiliary_loss_clip": 0.01161587,
      "auxiliary_loss_mlp": 0.01032188,
      "balance_loss_clip": 1.05142593,
      "balance_loss_mlp": 1.02427828,
      "epoch": 0.7435820357121385,
      "flos": 27377576928000.0,
      "grad_norm": 1.6954022801070812,
      "language_loss": 0.76154464,
      "learning_rate": 6.508149799851932e-07,
      "loss": 0.78348243,
      "num_input_tokens_seen": 132919505,
      "step": 6184,
      "time_per_iteration": 3.6283488273620605
    },
    {
      "auxiliary_loss_clip": 0.01130184,
      "auxiliary_loss_mlp": 0.01024175,
      "balance_loss_clip": 1.04517055,
      "balance_loss_mlp": 1.01744902,
      "epoch": 0.7437022786027776,
      "flos": 23987645948160.0,
      "grad_norm": 3.0726982497171442,
      "language_loss": 0.61914968,
      "learning_rate": 6.502400530069183e-07,
      "loss": 0.64069337,
      "num_input_tokens_seen": 132939390,
      "step": 6185,
      "time_per_iteration": 2.7357637882232666
    },
    {
      "auxiliary_loss_clip": 0.01159125,
      "auxiliary_loss_mlp": 0.01030335,
      "balance_loss_clip": 1.05040646,
      "balance_loss_mlp": 1.02156103,
      "epoch": 0.7438225214934167,
      "flos": 21866761451520.0,
      "grad_norm": 2.0994534675234418,
      "language_loss": 0.68558383,
      "learning_rate": 6.496653307861535e-07,
      "loss": 0.7074784,
      "num_input_tokens_seen": 132960060,
      "step": 6186,
      "time_per_iteration": 2.6171793937683105
    },
    {
      "auxiliary_loss_clip": 0.01137514,
      "auxiliary_loss_mlp": 0.01031933,
      "balance_loss_clip": 1.04890859,
      "balance_loss_mlp": 1.02413058,
      "epoch": 0.7439427643840558,
      "flos": 20230097224320.0,
      "grad_norm": 2.593579865716336,
      "language_loss": 0.65913832,
      "learning_rate": 6.490908134100857e-07,
      "loss": 0.6808328,
      "num_input_tokens_seen": 132978525,
      "step": 6187,
      "time_per_iteration": 2.7518043518066406
    },
    {
      "auxiliary_loss_clip": 0.01102894,
      "auxiliary_loss_mlp": 0.01030753,
      "balance_loss_clip": 1.04302967,
      "balance_loss_mlp": 1.02229476,
      "epoch": 0.7440630072746949,
      "flos": 20849915335680.0,
      "grad_norm": 1.974575265934767,
      "language_loss": 0.69637954,
      "learning_rate": 6.48516500965866e-07,
      "loss": 0.71771598,
      "num_input_tokens_seen": 132998460,
      "step": 6188,
      "time_per_iteration": 2.8303749561309814
    },
    {
      "auxiliary_loss_clip": 0.01142533,
      "auxiliary_loss_mlp": 0.01031256,
      "balance_loss_clip": 1.05023062,
      "balance_loss_mlp": 1.02315569,
      "epoch": 0.7441832501653339,
      "flos": 26503762769280.0,
      "grad_norm": 1.6710318104330273,
      "language_loss": 0.81799191,
      "learning_rate": 6.479423935406192e-07,
      "loss": 0.83972979,
      "num_input_tokens_seen": 133018445,
      "step": 6189,
      "time_per_iteration": 2.7077434062957764
    },
    {
      "auxiliary_loss_clip": 0.01057046,
      "auxiliary_loss_mlp": 0.01000729,
      "balance_loss_clip": 1.01181674,
      "balance_loss_mlp": 0.99977535,
      "epoch": 0.7443034930559731,
      "flos": 68602848088320.0,
      "grad_norm": 0.8083146021789376,
      "language_loss": 0.61983597,
      "learning_rate": 6.473684912214357e-07,
      "loss": 0.64041364,
      "num_input_tokens_seen": 133082005,
      "step": 6190,
      "time_per_iteration": 3.394434690475464
    },
    {
      "auxiliary_loss_clip": 0.01163553,
      "auxiliary_loss_mlp": 0.01027032,
      "balance_loss_clip": 1.05292916,
      "balance_loss_mlp": 1.01846099,
      "epoch": 0.7444237359466122,
      "flos": 18654982951680.0,
      "grad_norm": 5.348480000469675,
      "language_loss": 0.69843519,
      "learning_rate": 6.467947940953778e-07,
      "loss": 0.72034109,
      "num_input_tokens_seen": 133100530,
      "step": 6191,
      "time_per_iteration": 3.5741841793060303
    },
    {
      "auxiliary_loss_clip": 0.01112891,
      "auxiliary_loss_mlp": 0.01023615,
      "balance_loss_clip": 1.04508829,
      "balance_loss_mlp": 1.01603925,
      "epoch": 0.7445439788372512,
      "flos": 22817604326400.0,
      "grad_norm": 1.88051068695547,
      "language_loss": 0.72557247,
      "learning_rate": 6.462213022494732e-07,
      "loss": 0.74693757,
      "num_input_tokens_seen": 133119775,
      "step": 6192,
      "time_per_iteration": 2.852785110473633
    },
    {
      "auxiliary_loss_clip": 0.01038209,
      "auxiliary_loss_mlp": 0.00999046,
      "balance_loss_clip": 1.01961672,
      "balance_loss_mlp": 0.99808067,
      "epoch": 0.7446642217278904,
      "flos": 67045690615680.0,
      "grad_norm": 0.7714242336384004,
      "language_loss": 0.60845703,
      "learning_rate": 6.456480157707201e-07,
      "loss": 0.6288296,
      "num_input_tokens_seen": 133184550,
      "step": 6193,
      "time_per_iteration": 3.239701509475708
    },
    {
      "auxiliary_loss_clip": 0.01164675,
      "auxiliary_loss_mlp": 0.01029312,
      "balance_loss_clip": 1.05083907,
      "balance_loss_mlp": 1.02106881,
      "epoch": 0.7447844646185294,
      "flos": 17417465631360.0,
      "grad_norm": 2.2940143154750445,
      "language_loss": 0.85345775,
      "learning_rate": 6.450749347460866e-07,
      "loss": 0.87539762,
      "num_input_tokens_seen": 133201525,
      "step": 6194,
      "time_per_iteration": 2.6361358165740967
    },
    {
      "auxiliary_loss_clip": 0.01159758,
      "auxiliary_loss_mlp": 0.01028472,
      "balance_loss_clip": 1.04999268,
      "balance_loss_mlp": 1.02038336,
      "epoch": 0.7449047075091685,
      "flos": 26615876094720.0,
      "grad_norm": 1.7640264325982868,
      "language_loss": 0.79205954,
      "learning_rate": 6.445020592625083e-07,
      "loss": 0.81394184,
      "num_input_tokens_seen": 133222175,
      "step": 6195,
      "time_per_iteration": 2.7363431453704834
    },
    {
      "auxiliary_loss_clip": 0.01121155,
      "auxiliary_loss_mlp": 0.01021239,
      "balance_loss_clip": 1.04071367,
      "balance_loss_mlp": 1.01309133,
      "epoch": 0.7450249503998077,
      "flos": 14170458867840.0,
      "grad_norm": 2.6960382470360438,
      "language_loss": 0.80483484,
      "learning_rate": 6.4392938940689e-07,
      "loss": 0.82625884,
      "num_input_tokens_seen": 133237590,
      "step": 6196,
      "time_per_iteration": 2.671463966369629
    },
    {
      "auxiliary_loss_clip": 0.01125436,
      "auxiliary_loss_mlp": 0.01025158,
      "balance_loss_clip": 1.04624677,
      "balance_loss_mlp": 1.01728451,
      "epoch": 0.7451451932904467,
      "flos": 19606687752960.0,
      "grad_norm": 2.1330162417933507,
      "language_loss": 0.71254385,
      "learning_rate": 6.433569252661049e-07,
      "loss": 0.73404974,
      "num_input_tokens_seen": 133255590,
      "step": 6197,
      "time_per_iteration": 2.81387996673584
    },
    {
      "auxiliary_loss_clip": 0.01161973,
      "auxiliary_loss_mlp": 0.01031355,
      "balance_loss_clip": 1.04977202,
      "balance_loss_mlp": 1.02399945,
      "epoch": 0.7452654361810858,
      "flos": 12495405980160.0,
      "grad_norm": 1.8174045941325798,
      "language_loss": 0.71401322,
      "learning_rate": 6.427846669269952e-07,
      "loss": 0.73594648,
      "num_input_tokens_seen": 133273210,
      "step": 6198,
      "time_per_iteration": 2.684385061264038
    },
    {
      "auxiliary_loss_clip": 0.01151522,
      "auxiliary_loss_mlp": 0.01031061,
      "balance_loss_clip": 1.05323577,
      "balance_loss_mlp": 1.023983,
      "epoch": 0.7453856790717249,
      "flos": 22127329687680.0,
      "grad_norm": 2.3028104590648097,
      "language_loss": 0.82645369,
      "learning_rate": 6.422126144763729e-07,
      "loss": 0.84827948,
      "num_input_tokens_seen": 133292600,
      "step": 6199,
      "time_per_iteration": 2.7171967029571533
    },
    {
      "auxiliary_loss_clip": 0.01164334,
      "auxiliary_loss_mlp": 0.01024957,
      "balance_loss_clip": 1.05142355,
      "balance_loss_mlp": 1.01682651,
      "epoch": 0.745505921962364,
      "flos": 20010682995840.0,
      "grad_norm": 2.6711060380190923,
      "language_loss": 0.77122784,
      "learning_rate": 6.416407680010174e-07,
      "loss": 0.79312074,
      "num_input_tokens_seen": 133306960,
      "step": 6200,
      "time_per_iteration": 2.54848575592041
    },
    {
      "auxiliary_loss_clip": 0.01085994,
      "auxiliary_loss_mlp": 0.01032348,
      "balance_loss_clip": 1.03713143,
      "balance_loss_mlp": 1.02456951,
      "epoch": 0.745626164853003,
      "flos": 24677884673280.0,
      "grad_norm": 2.254374423410843,
      "language_loss": 0.81488043,
      "learning_rate": 6.410691275876774e-07,
      "loss": 0.83606386,
      "num_input_tokens_seen": 133326380,
      "step": 6201,
      "time_per_iteration": 2.848529577255249
    },
    {
      "auxiliary_loss_clip": 0.01143541,
      "auxiliary_loss_mlp": 0.01032443,
      "balance_loss_clip": 1.04896414,
      "balance_loss_mlp": 1.0249567,
      "epoch": 0.7457464077436422,
      "flos": 14538830797440.0,
      "grad_norm": 3.266223145365345,
      "language_loss": 0.76873749,
      "learning_rate": 6.404976933230704e-07,
      "loss": 0.7904973,
      "num_input_tokens_seen": 133342900,
      "step": 6202,
      "time_per_iteration": 2.7276651859283447
    },
    {
      "auxiliary_loss_clip": 0.01151052,
      "auxiliary_loss_mlp": 0.01028968,
      "balance_loss_clip": 1.05332208,
      "balance_loss_mlp": 1.02109385,
      "epoch": 0.7458666506342813,
      "flos": 34021194600960.0,
      "grad_norm": 1.8302014431942955,
      "language_loss": 0.73159921,
      "learning_rate": 6.399264652938813e-07,
      "loss": 0.75339943,
      "num_input_tokens_seen": 133363805,
      "step": 6203,
      "time_per_iteration": 2.7937467098236084
    },
    {
      "auxiliary_loss_clip": 0.01162145,
      "auxiliary_loss_mlp": 0.00762141,
      "balance_loss_clip": 1.04969764,
      "balance_loss_mlp": 1.00061202,
      "epoch": 0.7459868935249203,
      "flos": 24279025075200.0,
      "grad_norm": 2.115670119282808,
      "language_loss": 0.74446601,
      "learning_rate": 6.393554435867679e-07,
      "loss": 0.76370889,
      "num_input_tokens_seen": 133384655,
      "step": 6204,
      "time_per_iteration": 2.7291717529296875
    },
    {
      "auxiliary_loss_clip": 0.0113909,
      "auxiliary_loss_mlp": 0.01034462,
      "balance_loss_clip": 1.04671764,
      "balance_loss_mlp": 1.02657664,
      "epoch": 0.7461071364155595,
      "flos": 21908777385600.0,
      "grad_norm": 2.2401123900789566,
      "language_loss": 0.84064007,
      "learning_rate": 6.387846282883502e-07,
      "loss": 0.86237562,
      "num_input_tokens_seen": 133401185,
      "step": 6205,
      "time_per_iteration": 3.7106878757476807
    },
    {
      "auxiliary_loss_clip": 0.01117702,
      "auxiliary_loss_mlp": 0.01029419,
      "balance_loss_clip": 1.04661715,
      "balance_loss_mlp": 1.0216645,
      "epoch": 0.7462273793061985,
      "flos": 22889712879360.0,
      "grad_norm": 1.9946011679688338,
      "language_loss": 0.77056426,
      "learning_rate": 6.38214019485223e-07,
      "loss": 0.79203546,
      "num_input_tokens_seen": 133420010,
      "step": 6206,
      "time_per_iteration": 3.746394395828247
    },
    {
      "auxiliary_loss_clip": 0.01147276,
      "auxiliary_loss_mlp": 0.01026948,
      "balance_loss_clip": 1.04884744,
      "balance_loss_mlp": 1.01884151,
      "epoch": 0.7463476221968376,
      "flos": 19968451580160.0,
      "grad_norm": 1.8048098346797985,
      "language_loss": 0.71839428,
      "learning_rate": 6.376436172639461e-07,
      "loss": 0.7401365,
      "num_input_tokens_seen": 133437855,
      "step": 6207,
      "time_per_iteration": 2.752572774887085
    },
    {
      "auxiliary_loss_clip": 0.01163522,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.05191112,
      "balance_loss_mlp": 1.02379084,
      "epoch": 0.7464678650874768,
      "flos": 16836610798080.0,
      "grad_norm": 3.3811320023526186,
      "language_loss": 0.65423024,
      "learning_rate": 6.370734217110487e-07,
      "loss": 0.67618895,
      "num_input_tokens_seen": 133456600,
      "step": 6208,
      "time_per_iteration": 2.64365291595459
    },
    {
      "auxiliary_loss_clip": 0.01154566,
      "auxiliary_loss_mlp": 0.01030641,
      "balance_loss_clip": 1.0539031,
      "balance_loss_mlp": 1.02243352,
      "epoch": 0.7465881079781158,
      "flos": 48100869843840.0,
      "grad_norm": 1.4332457557890241,
      "language_loss": 0.64267176,
      "learning_rate": 6.36503432913031e-07,
      "loss": 0.66452384,
      "num_input_tokens_seen": 133479745,
      "step": 6209,
      "time_per_iteration": 3.873183250427246
    },
    {
      "auxiliary_loss_clip": 0.01137188,
      "auxiliary_loss_mlp": 0.0102773,
      "balance_loss_clip": 1.0488236,
      "balance_loss_mlp": 1.01918876,
      "epoch": 0.7467083508687549,
      "flos": 19677359761920.0,
      "grad_norm": 2.618123390269975,
      "language_loss": 0.68797982,
      "learning_rate": 6.359336509563569e-07,
      "loss": 0.70962906,
      "num_input_tokens_seen": 133495765,
      "step": 6210,
      "time_per_iteration": 2.7160415649414062
    },
    {
      "auxiliary_loss_clip": 0.01159265,
      "auxiliary_loss_mlp": 0.01031272,
      "balance_loss_clip": 1.04929614,
      "balance_loss_mlp": 1.02353525,
      "epoch": 0.7468285937593939,
      "flos": 17895436934400.0,
      "grad_norm": 2.242857241221728,
      "language_loss": 0.80685693,
      "learning_rate": 6.353640759274641e-07,
      "loss": 0.82876229,
      "num_input_tokens_seen": 133514655,
      "step": 6211,
      "time_per_iteration": 2.6369056701660156
    },
    {
      "auxiliary_loss_clip": 0.01162316,
      "auxiliary_loss_mlp": 0.01028028,
      "balance_loss_clip": 1.04937005,
      "balance_loss_mlp": 1.01999903,
      "epoch": 0.7469488366500331,
      "flos": 23141446369920.0,
      "grad_norm": 3.8113162873850417,
      "language_loss": 0.74822384,
      "learning_rate": 6.347947079127556e-07,
      "loss": 0.7701273,
      "num_input_tokens_seen": 133532555,
      "step": 6212,
      "time_per_iteration": 2.6944985389709473
    },
    {
      "auxiliary_loss_clip": 0.01147566,
      "auxiliary_loss_mlp": 0.01032695,
      "balance_loss_clip": 1.05287826,
      "balance_loss_mlp": 1.02534604,
      "epoch": 0.7470690795406721,
      "flos": 16690849407360.0,
      "grad_norm": 2.046095291107244,
      "language_loss": 0.76638937,
      "learning_rate": 6.342255469986053e-07,
      "loss": 0.78819191,
      "num_input_tokens_seen": 133551300,
      "step": 6213,
      "time_per_iteration": 2.657559871673584
    },
    {
      "auxiliary_loss_clip": 0.0114353,
      "auxiliary_loss_mlp": 0.01027672,
      "balance_loss_clip": 1.04776132,
      "balance_loss_mlp": 1.02000129,
      "epoch": 0.7471893224313112,
      "flos": 25192700352000.0,
      "grad_norm": 1.8041510875678686,
      "language_loss": 0.76662868,
      "learning_rate": 6.336565932713533e-07,
      "loss": 0.78834069,
      "num_input_tokens_seen": 133570725,
      "step": 6214,
      "time_per_iteration": 2.7662408351898193
    },
    {
      "auxiliary_loss_clip": 0.01135285,
      "auxiliary_loss_mlp": 0.01026653,
      "balance_loss_clip": 1.05028784,
      "balance_loss_mlp": 1.01870811,
      "epoch": 0.7473095653219504,
      "flos": 22526225199360.0,
      "grad_norm": 2.002674835949827,
      "language_loss": 0.77744281,
      "learning_rate": 6.330878468173088e-07,
      "loss": 0.79906219,
      "num_input_tokens_seen": 133590790,
      "step": 6215,
      "time_per_iteration": 2.749222755432129
    },
    {
      "auxiliary_loss_clip": 0.0113924,
      "auxiliary_loss_mlp": 0.01029025,
      "balance_loss_clip": 1.04564357,
      "balance_loss_mlp": 1.02120757,
      "epoch": 0.7474298082125894,
      "flos": 18113989236480.0,
      "grad_norm": 1.797677126180352,
      "language_loss": 0.73077631,
      "learning_rate": 6.32519307722752e-07,
      "loss": 0.75245893,
      "num_input_tokens_seen": 133608685,
      "step": 6216,
      "time_per_iteration": 2.760885715484619
    },
    {
      "auxiliary_loss_clip": 0.0099765,
      "auxiliary_loss_mlp": 0.00999314,
      "balance_loss_clip": 1.01494646,
      "balance_loss_mlp": 0.99844342,
      "epoch": 0.7475500511032285,
      "flos": 62086535193600.0,
      "grad_norm": 0.8388068718966204,
      "language_loss": 0.54891986,
      "learning_rate": 6.31950976073929e-07,
      "loss": 0.5688895,
      "num_input_tokens_seen": 133662775,
      "step": 6217,
      "time_per_iteration": 4.560904264450073
    },
    {
      "auxiliary_loss_clip": 0.01157184,
      "auxiliary_loss_mlp": 0.01028591,
      "balance_loss_clip": 1.04970837,
      "balance_loss_mlp": 1.02050292,
      "epoch": 0.7476702939938676,
      "flos": 17785586165760.0,
      "grad_norm": 2.720079227050622,
      "language_loss": 0.80909181,
      "learning_rate": 6.31382851957055e-07,
      "loss": 0.83094954,
      "num_input_tokens_seen": 133679595,
      "step": 6218,
      "time_per_iteration": 2.911752223968506
    },
    {
      "auxiliary_loss_clip": 0.01148738,
      "auxiliary_loss_mlp": 0.01030397,
      "balance_loss_clip": 1.05154479,
      "balance_loss_mlp": 1.02305102,
      "epoch": 0.7477905368845067,
      "flos": 27927944092800.0,
      "grad_norm": 3.176731629533473,
      "language_loss": 0.71346718,
      "learning_rate": 6.308149354583143e-07,
      "loss": 0.73525858,
      "num_input_tokens_seen": 133699000,
      "step": 6219,
      "time_per_iteration": 2.8878297805786133
    },
    {
      "auxiliary_loss_clip": 0.01179773,
      "auxiliary_loss_mlp": 0.010339,
      "balance_loss_clip": 1.05347848,
      "balance_loss_mlp": 1.02529275,
      "epoch": 0.7479107797751458,
      "flos": 26870374932480.0,
      "grad_norm": 2.9746861871328725,
      "language_loss": 0.82032025,
      "learning_rate": 6.302472266638586e-07,
      "loss": 0.84245694,
      "num_input_tokens_seen": 133719540,
      "step": 6220,
      "time_per_iteration": 2.651564836502075
    },
    {
      "auxiliary_loss_clip": 0.01109224,
      "auxiliary_loss_mlp": 0.01029622,
      "balance_loss_clip": 1.04709792,
      "balance_loss_mlp": 1.02046406,
      "epoch": 0.7480310226657849,
      "flos": 33943375785600.0,
      "grad_norm": 2.3344266102124975,
      "language_loss": 0.70133036,
      "learning_rate": 6.296797256598101e-07,
      "loss": 0.72271883,
      "num_input_tokens_seen": 133741020,
      "step": 6221,
      "time_per_iteration": 2.8099348545074463
    },
    {
      "auxiliary_loss_clip": 0.01105047,
      "auxiliary_loss_mlp": 0.01030748,
      "balance_loss_clip": 1.0411191,
      "balance_loss_mlp": 1.02344012,
      "epoch": 0.748151265556424,
      "flos": 24826555065600.0,
      "grad_norm": 1.8684415139191257,
      "language_loss": 0.81198412,
      "learning_rate": 6.291124325322576e-07,
      "loss": 0.83334208,
      "num_input_tokens_seen": 133761145,
      "step": 6222,
      "time_per_iteration": 2.807760715484619
    },
    {
      "auxiliary_loss_clip": 0.01161885,
      "auxiliary_loss_mlp": 0.01028878,
      "balance_loss_clip": 1.05083799,
      "balance_loss_mlp": 1.02087283,
      "epoch": 0.748271508447063,
      "flos": 38399351535360.0,
      "grad_norm": 1.752987897945187,
      "language_loss": 0.62519884,
      "learning_rate": 6.285453473672595e-07,
      "loss": 0.64710641,
      "num_input_tokens_seen": 133783715,
      "step": 6223,
      "time_per_iteration": 2.8110265731811523
    },
    {
      "auxiliary_loss_clip": 0.01128863,
      "auxiliary_loss_mlp": 0.01025805,
      "balance_loss_clip": 1.04325449,
      "balance_loss_mlp": 1.01826203,
      "epoch": 0.7483917513377022,
      "flos": 21541842000000.0,
      "grad_norm": 1.8893436567322548,
      "language_loss": 0.75859398,
      "learning_rate": 6.279784702508415e-07,
      "loss": 0.78014064,
      "num_input_tokens_seen": 133804465,
      "step": 6224,
      "time_per_iteration": 2.7906112670898438
    },
    {
      "auxiliary_loss_clip": 0.01055641,
      "auxiliary_loss_mlp": 0.01000366,
      "balance_loss_clip": 1.0096159,
      "balance_loss_mlp": 0.999412,
      "epoch": 0.7485119942283412,
      "flos": 62314532772480.0,
      "grad_norm": 0.7783942629656883,
      "language_loss": 0.58489096,
      "learning_rate": 6.274118012689979e-07,
      "loss": 0.60545099,
      "num_input_tokens_seen": 133866365,
      "step": 6225,
      "time_per_iteration": 3.2989795207977295
    },
    {
      "auxiliary_loss_clip": 0.01122788,
      "auxiliary_loss_mlp": 0.01029694,
      "balance_loss_clip": 1.043504,
      "balance_loss_mlp": 1.02096224,
      "epoch": 0.7486322371189803,
      "flos": 29937613104000.0,
      "grad_norm": 2.0707605693197877,
      "language_loss": 0.68542159,
      "learning_rate": 6.268453405076943e-07,
      "loss": 0.70694637,
      "num_input_tokens_seen": 133888760,
      "step": 6226,
      "time_per_iteration": 2.83892822265625
    },
    {
      "auxiliary_loss_clip": 0.01175569,
      "auxiliary_loss_mlp": 0.01025797,
      "balance_loss_clip": 1.05263984,
      "balance_loss_mlp": 1.0179379,
      "epoch": 0.7487524800096195,
      "flos": 18949414734720.0,
      "grad_norm": 2.016406194299528,
      "language_loss": 0.82731891,
      "learning_rate": 6.262790880528592e-07,
      "loss": 0.84933257,
      "num_input_tokens_seen": 133906380,
      "step": 6227,
      "time_per_iteration": 2.5928947925567627
    },
    {
      "auxiliary_loss_clip": 0.01170346,
      "auxiliary_loss_mlp": 0.00762451,
      "balance_loss_clip": 1.05302799,
      "balance_loss_mlp": 1.00057721,
      "epoch": 0.7488727229002585,
      "flos": 18697393935360.0,
      "grad_norm": 4.035089722260279,
      "language_loss": 0.8003239,
      "learning_rate": 6.257130439903951e-07,
      "loss": 0.81965184,
      "num_input_tokens_seen": 133922875,
      "step": 6228,
      "time_per_iteration": 2.6386945247650146
    },
    {
      "auxiliary_loss_clip": 0.01116793,
      "auxiliary_loss_mlp": 0.01034321,
      "balance_loss_clip": 1.04777622,
      "balance_loss_mlp": 1.02643502,
      "epoch": 0.7489929657908976,
      "flos": 23623368168960.0,
      "grad_norm": 2.9473854878156382,
      "language_loss": 0.81345141,
      "learning_rate": 6.251472084061695e-07,
      "loss": 0.83496261,
      "num_input_tokens_seen": 133941795,
      "step": 6229,
      "time_per_iteration": 2.7119364738464355
    },
    {
      "auxiliary_loss_clip": 0.01125173,
      "auxiliary_loss_mlp": 0.01026806,
      "balance_loss_clip": 1.04478383,
      "balance_loss_mlp": 1.01937664,
      "epoch": 0.7491132086815367,
      "flos": 20551533056640.0,
      "grad_norm": 1.9430085722503845,
      "language_loss": 0.89302242,
      "learning_rate": 6.245815813860191e-07,
      "loss": 0.91454226,
      "num_input_tokens_seen": 133957305,
      "step": 6230,
      "time_per_iteration": 3.051032781600952
    },
    {
      "auxiliary_loss_clip": 0.01129236,
      "auxiliary_loss_mlp": 0.01030686,
      "balance_loss_clip": 1.04248977,
      "balance_loss_mlp": 1.02249002,
      "epoch": 0.7492334515721758,
      "flos": 23003011353600.0,
      "grad_norm": 2.153587682786618,
      "language_loss": 0.71021622,
      "learning_rate": 6.240161630157495e-07,
      "loss": 0.73181546,
      "num_input_tokens_seen": 133976660,
      "step": 6231,
      "time_per_iteration": 3.6850764751434326
    },
    {
      "auxiliary_loss_clip": 0.01179107,
      "auxiliary_loss_mlp": 0.01027097,
      "balance_loss_clip": 1.05275524,
      "balance_loss_mlp": 1.01892507,
      "epoch": 0.7493536944628149,
      "flos": 16398823835520.0,
      "grad_norm": 2.6739651048094535,
      "language_loss": 0.70548427,
      "learning_rate": 6.23450953381133e-07,
      "loss": 0.72754633,
      "num_input_tokens_seen": 133994750,
      "step": 6232,
      "time_per_iteration": 2.570115566253662
    },
    {
      "auxiliary_loss_clip": 0.01147697,
      "auxiliary_loss_mlp": 0.01031045,
      "balance_loss_clip": 1.04741478,
      "balance_loss_mlp": 1.02348089,
      "epoch": 0.749473937353454,
      "flos": 15338561155200.0,
      "grad_norm": 3.636210251062607,
      "language_loss": 0.68148571,
      "learning_rate": 6.228859525679131e-07,
      "loss": 0.70327312,
      "num_input_tokens_seen": 134009165,
      "step": 6233,
      "time_per_iteration": 2.7444913387298584
    },
    {
      "auxiliary_loss_clip": 0.01124641,
      "auxiliary_loss_mlp": 0.01023183,
      "balance_loss_clip": 1.04274189,
      "balance_loss_mlp": 1.01557803,
      "epoch": 0.7495941802440931,
      "flos": 18951138587520.0,
      "grad_norm": 3.700190824063216,
      "language_loss": 0.80243814,
      "learning_rate": 6.223211606617986e-07,
      "loss": 0.82391644,
      "num_input_tokens_seen": 134027585,
      "step": 6234,
      "time_per_iteration": 2.694706439971924
    },
    {
      "auxiliary_loss_clip": 0.01136307,
      "auxiliary_loss_mlp": 0.01033302,
      "balance_loss_clip": 1.04765999,
      "balance_loss_mlp": 1.02620339,
      "epoch": 0.7497144231347321,
      "flos": 22492469393280.0,
      "grad_norm": 1.78430960267406,
      "language_loss": 0.8400014,
      "learning_rate": 6.217565777484701e-07,
      "loss": 0.86169755,
      "num_input_tokens_seen": 134046680,
      "step": 6235,
      "time_per_iteration": 3.759284496307373
    },
    {
      "auxiliary_loss_clip": 0.01163654,
      "auxiliary_loss_mlp": 0.01028582,
      "balance_loss_clip": 1.05150127,
      "balance_loss_mlp": 1.02106619,
      "epoch": 0.7498346660253713,
      "flos": 24243509502720.0,
      "grad_norm": 1.9404535057134835,
      "language_loss": 0.80910993,
      "learning_rate": 6.211922039135722e-07,
      "loss": 0.8310324,
      "num_input_tokens_seen": 134066825,
      "step": 6236,
      "time_per_iteration": 2.7201030254364014
    },
    {
      "auxiliary_loss_clip": 0.01125145,
      "auxiliary_loss_mlp": 0.01030894,
      "balance_loss_clip": 1.04657543,
      "balance_loss_mlp": 1.02336848,
      "epoch": 0.7499549089160104,
      "flos": 24387080163840.0,
      "grad_norm": 1.989501389976169,
      "language_loss": 0.81204218,
      "learning_rate": 6.206280392427201e-07,
      "loss": 0.83360255,
      "num_input_tokens_seen": 134086410,
      "step": 6237,
      "time_per_iteration": 2.7169077396392822
    },
    {
      "auxiliary_loss_clip": 0.0114204,
      "auxiliary_loss_mlp": 0.01025498,
      "balance_loss_clip": 1.04751468,
      "balance_loss_mlp": 1.01745105,
      "epoch": 0.7500751518066494,
      "flos": 34057320704640.0,
      "grad_norm": 1.7917932876984626,
      "language_loss": 0.739012,
      "learning_rate": 6.200640838214983e-07,
      "loss": 0.76068735,
      "num_input_tokens_seen": 134109185,
      "step": 6238,
      "time_per_iteration": 2.780864953994751
    },
    {
      "auxiliary_loss_clip": 0.01163344,
      "auxiliary_loss_mlp": 0.01025426,
      "balance_loss_clip": 1.05236423,
      "balance_loss_mlp": 1.01799631,
      "epoch": 0.7501953946972886,
      "flos": 18843586289280.0,
      "grad_norm": 2.0589489783260713,
      "language_loss": 0.67407346,
      "learning_rate": 6.195003377354578e-07,
      "loss": 0.69596112,
      "num_input_tokens_seen": 134128455,
      "step": 6239,
      "time_per_iteration": 2.608626365661621
    },
    {
      "auxiliary_loss_clip": 0.01108061,
      "auxiliary_loss_mlp": 0.01031054,
      "balance_loss_clip": 1.04495311,
      "balance_loss_mlp": 1.02307892,
      "epoch": 0.7503156375879276,
      "flos": 20257675891200.0,
      "grad_norm": 2.9421053344261763,
      "language_loss": 0.73320913,
      "learning_rate": 6.189368010701183e-07,
      "loss": 0.75460029,
      "num_input_tokens_seen": 134145515,
      "step": 6240,
      "time_per_iteration": 2.7730348110198975
    },
    {
      "auxiliary_loss_clip": 0.01164446,
      "auxiliary_loss_mlp": 0.01026049,
      "balance_loss_clip": 1.05087638,
      "balance_loss_mlp": 1.01828849,
      "epoch": 0.7504358804785667,
      "flos": 13480040574720.0,
      "grad_norm": 2.83530060020167,
      "language_loss": 0.77185798,
      "learning_rate": 6.183734739109683e-07,
      "loss": 0.79376292,
      "num_input_tokens_seen": 134163335,
      "step": 6241,
      "time_per_iteration": 2.6410460472106934
    },
    {
      "auxiliary_loss_clip": 0.01153361,
      "auxiliary_loss_mlp": 0.01030692,
      "balance_loss_clip": 1.05247796,
      "balance_loss_mlp": 1.02191234,
      "epoch": 0.7505561233692057,
      "flos": 29461042431360.0,
      "grad_norm": 2.4956580524763736,
      "language_loss": 0.69006491,
      "learning_rate": 6.178103563434629e-07,
      "loss": 0.71190536,
      "num_input_tokens_seen": 134182335,
      "step": 6242,
      "time_per_iteration": 3.7306525707244873
    },
    {
      "auxiliary_loss_clip": 0.01136764,
      "auxiliary_loss_mlp": 0.01026693,
      "balance_loss_clip": 1.04460216,
      "balance_loss_mlp": 1.01902795,
      "epoch": 0.7506763662598449,
      "flos": 20302457172480.0,
      "grad_norm": 1.6044498547422363,
      "language_loss": 0.83828461,
      "learning_rate": 6.172474484530283e-07,
      "loss": 0.85991913,
      "num_input_tokens_seen": 134201070,
      "step": 6243,
      "time_per_iteration": 2.744269847869873
    },
    {
      "auxiliary_loss_clip": 0.01122841,
      "auxiliary_loss_mlp": 0.01025844,
      "balance_loss_clip": 1.04434371,
      "balance_loss_mlp": 1.01751125,
      "epoch": 0.750796609150484,
      "flos": 37230961939200.0,
      "grad_norm": 1.961555487320284,
      "language_loss": 0.76011169,
      "learning_rate": 6.166847503250563e-07,
      "loss": 0.78159851,
      "num_input_tokens_seen": 134223310,
      "step": 6244,
      "time_per_iteration": 2.8050637245178223
    },
    {
      "auxiliary_loss_clip": 0.01138335,
      "auxiliary_loss_mlp": 0.01024494,
      "balance_loss_clip": 1.04972625,
      "balance_loss_mlp": 1.0170548,
      "epoch": 0.750916852041123,
      "flos": 19609417186560.0,
      "grad_norm": 2.8469000558452313,
      "language_loss": 0.78866982,
      "learning_rate": 6.161222620449078e-07,
      "loss": 0.81029809,
      "num_input_tokens_seen": 134242085,
      "step": 6245,
      "time_per_iteration": 2.7729151248931885
    },
    {
      "auxiliary_loss_clip": 0.01129923,
      "auxiliary_loss_mlp": 0.01025679,
      "balance_loss_clip": 1.0470618,
      "balance_loss_mlp": 1.01767123,
      "epoch": 0.7510370949317622,
      "flos": 25112690807040.0,
      "grad_norm": 2.5893861392176456,
      "language_loss": 0.80138493,
      "learning_rate": 6.155599836979117e-07,
      "loss": 0.82294101,
      "num_input_tokens_seen": 134260770,
      "step": 6246,
      "time_per_iteration": 2.7340569496154785
    },
    {
      "auxiliary_loss_clip": 0.0116153,
      "auxiliary_loss_mlp": 0.01029303,
      "balance_loss_clip": 1.05259442,
      "balance_loss_mlp": 1.02115488,
      "epoch": 0.7511573378224012,
      "flos": 19062282245760.0,
      "grad_norm": 6.54170184010422,
      "language_loss": 0.8160153,
      "learning_rate": 6.149979153693649e-07,
      "loss": 0.83792365,
      "num_input_tokens_seen": 134278025,
      "step": 6247,
      "time_per_iteration": 2.7192509174346924
    },
    {
      "auxiliary_loss_clip": 0.01170397,
      "auxiliary_loss_mlp": 0.01032934,
      "balance_loss_clip": 1.05393815,
      "balance_loss_mlp": 1.02469015,
      "epoch": 0.7512775807130403,
      "flos": 19937676602880.0,
      "grad_norm": 6.868695056584409,
      "language_loss": 0.76751238,
      "learning_rate": 6.144360571445343e-07,
      "loss": 0.78954566,
      "num_input_tokens_seen": 134297170,
      "step": 6248,
      "time_per_iteration": 2.6369457244873047
    },
    {
      "auxiliary_loss_clip": 0.01148532,
      "auxiliary_loss_mlp": 0.01027232,
      "balance_loss_clip": 1.04939389,
      "balance_loss_mlp": 1.01930785,
      "epoch": 0.7513978236036795,
      "flos": 20739920912640.0,
      "grad_norm": 1.8155561536761544,
      "language_loss": 0.80295658,
      "learning_rate": 6.138744091086509e-07,
      "loss": 0.82471418,
      "num_input_tokens_seen": 134316755,
      "step": 6249,
      "time_per_iteration": 2.7076268196105957
    },
    {
      "auxiliary_loss_clip": 0.01135487,
      "auxiliary_loss_mlp": 0.01029099,
      "balance_loss_clip": 1.05173135,
      "balance_loss_mlp": 1.02068329,
      "epoch": 0.7515180664943185,
      "flos": 27563163523200.0,
      "grad_norm": 2.3805107914267105,
      "language_loss": 0.72983307,
      "learning_rate": 6.133129713469183e-07,
      "loss": 0.75147891,
      "num_input_tokens_seen": 134335960,
      "step": 6250,
      "time_per_iteration": 2.8993096351623535
    },
    {
      "auxiliary_loss_clip": 0.01144131,
      "auxiliary_loss_mlp": 0.01033115,
      "balance_loss_clip": 1.04679883,
      "balance_loss_mlp": 1.02463973,
      "epoch": 0.7516383093849576,
      "flos": 33803181002880.0,
      "grad_norm": 1.8658964913189253,
      "language_loss": 0.64481765,
      "learning_rate": 6.127517439445053e-07,
      "loss": 0.66659009,
      "num_input_tokens_seen": 134356805,
      "step": 6251,
      "time_per_iteration": 2.7364912033081055
    },
    {
      "auxiliary_loss_clip": 0.0116053,
      "auxiliary_loss_mlp": 0.01029762,
      "balance_loss_clip": 1.05101466,
      "balance_loss_mlp": 1.02262759,
      "epoch": 0.7517585522755967,
      "flos": 29746172592000.0,
      "grad_norm": 2.7733229709011415,
      "language_loss": 0.82025421,
      "learning_rate": 6.121907269865498e-07,
      "loss": 0.84215713,
      "num_input_tokens_seen": 134376295,
      "step": 6252,
      "time_per_iteration": 2.7282731533050537
    },
    {
      "auxiliary_loss_clip": 0.01000889,
      "auxiliary_loss_mlp": 0.01001641,
      "balance_loss_clip": 1.01054347,
      "balance_loss_mlp": 1.00063372,
      "epoch": 0.7518787951662358,
      "flos": 69807974319360.0,
      "grad_norm": 0.9283405046063257,
      "language_loss": 0.67169213,
      "learning_rate": 6.116299205581577e-07,
      "loss": 0.69171739,
      "num_input_tokens_seen": 134431125,
      "step": 6253,
      "time_per_iteration": 3.394313097000122
    },
    {
      "auxiliary_loss_clip": 0.01081103,
      "auxiliary_loss_mlp": 0.01029088,
      "balance_loss_clip": 1.04063547,
      "balance_loss_mlp": 1.02034974,
      "epoch": 0.7519990380568748,
      "flos": 34203225749760.0,
      "grad_norm": 2.9129609903393257,
      "language_loss": 0.68959773,
      "learning_rate": 6.110693247444018e-07,
      "loss": 0.71069962,
      "num_input_tokens_seen": 134452960,
      "step": 6254,
      "time_per_iteration": 3.3863186836242676
    },
    {
      "auxiliary_loss_clip": 0.01137116,
      "auxiliary_loss_mlp": 0.01023794,
      "balance_loss_clip": 1.04498851,
      "balance_loss_mlp": 1.0161525,
      "epoch": 0.752119280947514,
      "flos": 21725704742400.0,
      "grad_norm": 1.6597769165049547,
      "language_loss": 0.82320762,
      "learning_rate": 6.105089396303258e-07,
      "loss": 0.84481668,
      "num_input_tokens_seen": 134471350,
      "step": 6255,
      "time_per_iteration": 3.156846046447754
    },
    {
      "auxiliary_loss_clip": 0.01165257,
      "auxiliary_loss_mlp": 0.01026322,
      "balance_loss_clip": 1.05264688,
      "balance_loss_mlp": 1.01819777,
      "epoch": 0.7522395238381531,
      "flos": 32742774668160.0,
      "grad_norm": 1.9179658017857215,
      "language_loss": 0.75797939,
      "learning_rate": 6.099487653009383e-07,
      "loss": 0.77989519,
      "num_input_tokens_seen": 134490695,
      "step": 6256,
      "time_per_iteration": 2.724843978881836
    },
    {
      "auxiliary_loss_clip": 0.01130935,
      "auxiliary_loss_mlp": 0.00761985,
      "balance_loss_clip": 1.04404783,
      "balance_loss_mlp": 1.00063872,
      "epoch": 0.7523597667287921,
      "flos": 23476026579840.0,
      "grad_norm": 2.0440137517486523,
      "language_loss": 0.83178759,
      "learning_rate": 6.093888018412192e-07,
      "loss": 0.85071683,
      "num_input_tokens_seen": 134506885,
      "step": 6257,
      "time_per_iteration": 4.787440538406372
    },
    {
      "auxiliary_loss_clip": 0.01055981,
      "auxiliary_loss_mlp": 0.00751864,
      "balance_loss_clip": 1.008991,
      "balance_loss_mlp": 1.00005865,
      "epoch": 0.7524800096194313,
      "flos": 67346730501120.0,
      "grad_norm": 0.7016517535754254,
      "language_loss": 0.54597497,
      "learning_rate": 6.088290493361125e-07,
      "loss": 0.56405342,
      "num_input_tokens_seen": 134571770,
      "step": 6258,
      "time_per_iteration": 3.429882764816284
    },
    {
      "auxiliary_loss_clip": 0.01153025,
      "auxiliary_loss_mlp": 0.01031069,
      "balance_loss_clip": 1.05022502,
      "balance_loss_mlp": 1.02273023,
      "epoch": 0.7526002525100703,
      "flos": 13006055681280.0,
      "grad_norm": 2.981008862269382,
      "language_loss": 0.70973742,
      "learning_rate": 6.082695078705322e-07,
      "loss": 0.73157829,
      "num_input_tokens_seen": 134589250,
      "step": 6259,
      "time_per_iteration": 2.6631762981414795
    },
    {
      "auxiliary_loss_clip": 0.01126247,
      "auxiliary_loss_mlp": 0.01033085,
      "balance_loss_clip": 1.04596007,
      "balance_loss_mlp": 1.0245676,
      "epoch": 0.7527204954007094,
      "flos": 21397229844480.0,
      "grad_norm": 2.329763985856576,
      "language_loss": 0.69401431,
      "learning_rate": 6.077101775293618e-07,
      "loss": 0.71560764,
      "num_input_tokens_seen": 134608075,
      "step": 6260,
      "time_per_iteration": 2.816392183303833
    },
    {
      "auxiliary_loss_clip": 0.01107059,
      "auxiliary_loss_mlp": 0.01031028,
      "balance_loss_clip": 1.04351163,
      "balance_loss_mlp": 1.02304089,
      "epoch": 0.7528407382913486,
      "flos": 18947188091520.0,
      "grad_norm": 2.9181675578598183,
      "language_loss": 0.82631052,
      "learning_rate": 6.071510583974504e-07,
      "loss": 0.84769142,
      "num_input_tokens_seen": 134623260,
      "step": 6261,
      "time_per_iteration": 3.874345541000366
    },
    {
      "auxiliary_loss_clip": 0.01162248,
      "auxiliary_loss_mlp": 0.01029469,
      "balance_loss_clip": 1.05183256,
      "balance_loss_mlp": 1.02125549,
      "epoch": 0.7529609811819876,
      "flos": 15231798956160.0,
      "grad_norm": 2.6415512749660297,
      "language_loss": 0.72155428,
      "learning_rate": 6.065921505596161e-07,
      "loss": 0.74347138,
      "num_input_tokens_seen": 134641540,
      "step": 6262,
      "time_per_iteration": 2.7102720737457275
    },
    {
      "auxiliary_loss_clip": 0.01089216,
      "auxiliary_loss_mlp": 0.01023883,
      "balance_loss_clip": 1.03983235,
      "balance_loss_mlp": 1.01587248,
      "epoch": 0.7530812240726267,
      "flos": 19354487385600.0,
      "grad_norm": 1.7149891040816803,
      "language_loss": 0.77286267,
      "learning_rate": 6.060334541006445e-07,
      "loss": 0.79399371,
      "num_input_tokens_seen": 134660035,
      "step": 6263,
      "time_per_iteration": 2.833235263824463
    },
    {
      "auxiliary_loss_clip": 0.01122306,
      "auxiliary_loss_mlp": 0.01028258,
      "balance_loss_clip": 1.04531431,
      "balance_loss_mlp": 1.02020502,
      "epoch": 0.7532014669632658,
      "flos": 27748247328000.0,
      "grad_norm": 1.735270700253295,
      "language_loss": 0.69167757,
      "learning_rate": 6.05474969105289e-07,
      "loss": 0.71318328,
      "num_input_tokens_seen": 134683025,
      "step": 6264,
      "time_per_iteration": 2.787452220916748
    },
    {
      "auxiliary_loss_clip": 0.01166503,
      "auxiliary_loss_mlp": 0.0103075,
      "balance_loss_clip": 1.05139112,
      "balance_loss_mlp": 1.02276862,
      "epoch": 0.7533217098539049,
      "flos": 14137421333760.0,
      "grad_norm": 1.9956012835696721,
      "language_loss": 0.73795164,
      "learning_rate": 6.049166956582725e-07,
      "loss": 0.75992405,
      "num_input_tokens_seen": 134701290,
      "step": 6265,
      "time_per_iteration": 2.6656758785247803
    },
    {
      "auxiliary_loss_clip": 0.01145362,
      "auxiliary_loss_mlp": 0.01030593,
      "balance_loss_clip": 1.04714036,
      "balance_loss_mlp": 1.0229758,
      "epoch": 0.753441952744544,
      "flos": 26429068437120.0,
      "grad_norm": 2.7209352625652152,
      "language_loss": 0.87773556,
      "learning_rate": 6.043586338442841e-07,
      "loss": 0.89949512,
      "num_input_tokens_seen": 134720345,
      "step": 6266,
      "time_per_iteration": 2.6882169246673584
    },
    {
      "auxiliary_loss_clip": 0.01138405,
      "auxiliary_loss_mlp": 0.01028538,
      "balance_loss_clip": 1.04588687,
      "balance_loss_mlp": 1.02137625,
      "epoch": 0.7535621956351831,
      "flos": 23878621192320.0,
      "grad_norm": 1.4308051376127444,
      "language_loss": 0.73415959,
      "learning_rate": 6.038007837479815e-07,
      "loss": 0.75582898,
      "num_input_tokens_seen": 134741450,
      "step": 6267,
      "time_per_iteration": 2.9051780700683594
    },
    {
      "auxiliary_loss_clip": 0.01159273,
      "auxiliary_loss_mlp": 0.01025275,
      "balance_loss_clip": 1.05092812,
      "balance_loss_mlp": 1.01789021,
      "epoch": 0.7536824385258222,
      "flos": 21795873960960.0,
      "grad_norm": 4.217782575600305,
      "language_loss": 0.64143074,
      "learning_rate": 6.032431454539897e-07,
      "loss": 0.66327626,
      "num_input_tokens_seen": 134760295,
      "step": 6268,
      "time_per_iteration": 3.7405683994293213
    },
    {
      "auxiliary_loss_clip": 0.01159086,
      "auxiliary_loss_mlp": 0.01027645,
      "balance_loss_clip": 1.05294919,
      "balance_loss_mlp": 1.02059948,
      "epoch": 0.7538026814164612,
      "flos": 28911644933760.0,
      "grad_norm": 1.9894604320515645,
      "language_loss": 0.81841272,
      "learning_rate": 6.026857190469014e-07,
      "loss": 0.84028006,
      "num_input_tokens_seen": 134782050,
      "step": 6269,
      "time_per_iteration": 2.690981864929199
    },
    {
      "auxiliary_loss_clip": 0.01132608,
      "auxiliary_loss_mlp": 0.01025006,
      "balance_loss_clip": 1.04975653,
      "balance_loss_mlp": 1.01709366,
      "epoch": 0.7539229243071004,
      "flos": 21104701482240.0,
      "grad_norm": 2.3527086242299995,
      "language_loss": 0.74056309,
      "learning_rate": 6.0212850461128e-07,
      "loss": 0.76213926,
      "num_input_tokens_seen": 134801170,
      "step": 6270,
      "time_per_iteration": 2.858858823776245
    },
    {
      "auxiliary_loss_clip": 0.01159627,
      "auxiliary_loss_mlp": 0.01029559,
      "balance_loss_clip": 1.04959273,
      "balance_loss_mlp": 1.02095234,
      "epoch": 0.7540431671977395,
      "flos": 15158469340800.0,
      "grad_norm": 5.981311100439345,
      "language_loss": 0.74676645,
      "learning_rate": 6.015715022316516e-07,
      "loss": 0.76865828,
      "num_input_tokens_seen": 134819150,
      "step": 6271,
      "time_per_iteration": 2.6292707920074463
    },
    {
      "auxiliary_loss_clip": 0.01177648,
      "auxiliary_loss_mlp": 0.01029907,
      "balance_loss_clip": 1.05141044,
      "balance_loss_mlp": 1.02118063,
      "epoch": 0.7541634100883785,
      "flos": 18770579896320.0,
      "grad_norm": 2.544775784724025,
      "language_loss": 0.78358155,
      "learning_rate": 6.010147119925154e-07,
      "loss": 0.80565703,
      "num_input_tokens_seen": 134836905,
      "step": 6272,
      "time_per_iteration": 2.6525044441223145
    },
    {
      "auxiliary_loss_clip": 0.01107895,
      "auxiliary_loss_mlp": 0.01023275,
      "balance_loss_clip": 1.0465529,
      "balance_loss_mlp": 1.01567578,
      "epoch": 0.7542836529790176,
      "flos": 20594770053120.0,
      "grad_norm": 16.014276879600697,
      "language_loss": 0.66607624,
      "learning_rate": 6.004581339783348e-07,
      "loss": 0.68738794,
      "num_input_tokens_seen": 134855225,
      "step": 6273,
      "time_per_iteration": 2.7886886596679688
    },
    {
      "auxiliary_loss_clip": 0.01135795,
      "auxiliary_loss_mlp": 0.01033349,
      "balance_loss_clip": 1.04651868,
      "balance_loss_mlp": 1.02482545,
      "epoch": 0.7544038958696567,
      "flos": 19095104298240.0,
      "grad_norm": 2.593966347340249,
      "language_loss": 0.68658012,
      "learning_rate": 5.999017682735425e-07,
      "loss": 0.70827156,
      "num_input_tokens_seen": 134871615,
      "step": 6274,
      "time_per_iteration": 2.743131160736084
    },
    {
      "auxiliary_loss_clip": 0.01097699,
      "auxiliary_loss_mlp": 0.01028859,
      "balance_loss_clip": 1.03997207,
      "balance_loss_mlp": 1.02087188,
      "epoch": 0.7545241387602958,
      "flos": 31723306859520.0,
      "grad_norm": 4.822504872648613,
      "language_loss": 0.66829836,
      "learning_rate": 5.993456149625387e-07,
      "loss": 0.68956393,
      "num_input_tokens_seen": 134892765,
      "step": 6275,
      "time_per_iteration": 2.868215322494507
    },
    {
      "auxiliary_loss_clip": 0.01133228,
      "auxiliary_loss_mlp": 0.01029985,
      "balance_loss_clip": 1.04499602,
      "balance_loss_mlp": 1.02239108,
      "epoch": 0.7546443816509348,
      "flos": 20296495514880.0,
      "grad_norm": 4.183412986734651,
      "language_loss": 0.82569933,
      "learning_rate": 5.987896741296909e-07,
      "loss": 0.84733152,
      "num_input_tokens_seen": 134910505,
      "step": 6276,
      "time_per_iteration": 2.6990318298339844
    },
    {
      "auxiliary_loss_clip": 0.0110867,
      "auxiliary_loss_mlp": 0.01034875,
      "balance_loss_clip": 1.04767418,
      "balance_loss_mlp": 1.02729297,
      "epoch": 0.754764624541574,
      "flos": 23696159080320.0,
      "grad_norm": 1.9861484381506904,
      "language_loss": 0.79128414,
      "learning_rate": 5.982339458593361e-07,
      "loss": 0.81271958,
      "num_input_tokens_seen": 134930445,
      "step": 6277,
      "time_per_iteration": 2.8443779945373535
    },
    {
      "auxiliary_loss_clip": 0.0113864,
      "auxiliary_loss_mlp": 0.01019917,
      "balance_loss_clip": 1.0487076,
      "balance_loss_mlp": 1.01266909,
      "epoch": 0.7548848674322131,
      "flos": 25337204766720.0,
      "grad_norm": 1.7095762310430207,
      "language_loss": 0.8403734,
      "learning_rate": 5.976784302357767e-07,
      "loss": 0.86195898,
      "num_input_tokens_seen": 134951010,
      "step": 6278,
      "time_per_iteration": 2.665675640106201
    },
    {
      "auxiliary_loss_clip": 0.01140159,
      "auxiliary_loss_mlp": 0.01027017,
      "balance_loss_clip": 1.05097044,
      "balance_loss_mlp": 1.01948011,
      "epoch": 0.7550051103228521,
      "flos": 19573147428480.0,
      "grad_norm": 2.1769618298703777,
      "language_loss": 0.73687613,
      "learning_rate": 5.971231273432855e-07,
      "loss": 0.7585479,
      "num_input_tokens_seen": 134970495,
      "step": 6279,
      "time_per_iteration": 2.7158639430999756
    },
    {
      "auxiliary_loss_clip": 0.01033826,
      "auxiliary_loss_mlp": 0.01004771,
      "balance_loss_clip": 1.00991285,
      "balance_loss_mlp": 1.00380576,
      "epoch": 0.7551253532134913,
      "flos": 64150068648960.0,
      "grad_norm": 0.8078846816513601,
      "language_loss": 0.5448842,
      "learning_rate": 5.965680372661e-07,
      "loss": 0.56527019,
      "num_input_tokens_seen": 135028060,
      "step": 6280,
      "time_per_iteration": 3.173922061920166
    },
    {
      "auxiliary_loss_clip": 0.01107155,
      "auxiliary_loss_mlp": 0.01021829,
      "balance_loss_clip": 1.04880595,
      "balance_loss_mlp": 1.01513588,
      "epoch": 0.7552455961041303,
      "flos": 26067986968320.0,
      "grad_norm": 2.1616402974956372,
      "language_loss": 0.55961245,
      "learning_rate": 5.960131600884266e-07,
      "loss": 0.58090228,
      "num_input_tokens_seen": 135047330,
      "step": 6281,
      "time_per_iteration": 2.796992778778076
    },
    {
      "auxiliary_loss_clip": 0.01133781,
      "auxiliary_loss_mlp": 0.01025887,
      "balance_loss_clip": 1.04783988,
      "balance_loss_mlp": 1.01892257,
      "epoch": 0.7553658389947694,
      "flos": 24498223822080.0,
      "grad_norm": 2.286811444805938,
      "language_loss": 0.76038617,
      "learning_rate": 5.954584958944413e-07,
      "loss": 0.7819829,
      "num_input_tokens_seen": 135065995,
      "step": 6282,
      "time_per_iteration": 2.7444822788238525
    },
    {
      "auxiliary_loss_clip": 0.01146381,
      "auxiliary_loss_mlp": 0.01025932,
      "balance_loss_clip": 1.05062044,
      "balance_loss_mlp": 1.01849341,
      "epoch": 0.7554860818854086,
      "flos": 21799465320960.0,
      "grad_norm": 2.2348546317258915,
      "language_loss": 0.82313979,
      "learning_rate": 5.949040447682854e-07,
      "loss": 0.84486294,
      "num_input_tokens_seen": 135085820,
      "step": 6283,
      "time_per_iteration": 4.5989954471588135
    },
    {
      "auxiliary_loss_clip": 0.01122954,
      "auxiliary_loss_mlp": 0.01024717,
      "balance_loss_clip": 1.04894805,
      "balance_loss_mlp": 1.01657522,
      "epoch": 0.7556063247760476,
      "flos": 16362123114240.0,
      "grad_norm": 4.377399941660616,
      "language_loss": 0.68719566,
      "learning_rate": 5.943498067940686e-07,
      "loss": 0.70867234,
      "num_input_tokens_seen": 135102845,
      "step": 6284,
      "time_per_iteration": 2.730184316635132
    },
    {
      "auxiliary_loss_clip": 0.01147061,
      "auxiliary_loss_mlp": 0.01032714,
      "balance_loss_clip": 1.05013072,
      "balance_loss_mlp": 1.02550745,
      "epoch": 0.7557265676666867,
      "flos": 27235155502080.0,
      "grad_norm": 2.291615961040734,
      "language_loss": 0.81730473,
      "learning_rate": 5.937957820558686e-07,
      "loss": 0.83910245,
      "num_input_tokens_seen": 135122190,
      "step": 6285,
      "time_per_iteration": 2.7857062816619873
    },
    {
      "auxiliary_loss_clip": 0.01044259,
      "auxiliary_loss_mlp": 0.01001898,
      "balance_loss_clip": 1.01049733,
      "balance_loss_mlp": 1.00097394,
      "epoch": 0.7558468105573258,
      "flos": 62189131415040.0,
      "grad_norm": 0.8486814320396773,
      "language_loss": 0.65212226,
      "learning_rate": 5.932419706377296e-07,
      "loss": 0.67258388,
      "num_input_tokens_seen": 135180495,
      "step": 6286,
      "time_per_iteration": 3.1938109397888184
    },
    {
      "auxiliary_loss_clip": 0.01159519,
      "auxiliary_loss_mlp": 0.01024108,
      "balance_loss_clip": 1.05304122,
      "balance_loss_mlp": 1.01607919,
      "epoch": 0.7559670534479649,
      "flos": 33249078823680.0,
      "grad_norm": 1.7587476622785834,
      "language_loss": 0.74163443,
      "learning_rate": 5.92688372623666e-07,
      "loss": 0.76347071,
      "num_input_tokens_seen": 135199200,
      "step": 6287,
      "time_per_iteration": 3.7748630046844482
    },
    {
      "auxiliary_loss_clip": 0.01146838,
      "auxiliary_loss_mlp": 0.01028034,
      "balance_loss_clip": 1.04799008,
      "balance_loss_mlp": 1.01895666,
      "epoch": 0.7560872963386039,
      "flos": 14064379027200.0,
      "grad_norm": 2.156674263515872,
      "language_loss": 0.74408078,
      "learning_rate": 5.921349880976574e-07,
      "loss": 0.76582944,
      "num_input_tokens_seen": 135217035,
      "step": 6288,
      "time_per_iteration": 2.692918062210083
    },
    {
      "auxiliary_loss_clip": 0.01163365,
      "auxiliary_loss_mlp": 0.01026603,
      "balance_loss_clip": 1.04957545,
      "balance_loss_mlp": 1.01879466,
      "epoch": 0.7562075392292431,
      "flos": 20412307941120.0,
      "grad_norm": 1.8683355789843372,
      "language_loss": 0.81479406,
      "learning_rate": 5.915818171436515e-07,
      "loss": 0.83669376,
      "num_input_tokens_seen": 135236370,
      "step": 6289,
      "time_per_iteration": 2.62117075920105
    },
    {
      "auxiliary_loss_clip": 0.01105192,
      "auxiliary_loss_mlp": 0.01025451,
      "balance_loss_clip": 1.04245877,
      "balance_loss_mlp": 1.01739848,
      "epoch": 0.7563277821198822,
      "flos": 20376792368640.0,
      "grad_norm": 1.6828578223783512,
      "language_loss": 0.7476536,
      "learning_rate": 5.910288598455642e-07,
      "loss": 0.76896,
      "num_input_tokens_seen": 135255720,
      "step": 6290,
      "time_per_iteration": 2.799091339111328
    },
    {
      "auxiliary_loss_clip": 0.01164104,
      "auxiliary_loss_mlp": 0.01024945,
      "balance_loss_clip": 1.05344617,
      "balance_loss_mlp": 1.01675487,
      "epoch": 0.7564480250105212,
      "flos": 18588261438720.0,
      "grad_norm": 2.585946858586236,
      "language_loss": 0.74998581,
      "learning_rate": 5.90476116287278e-07,
      "loss": 0.77187634,
      "num_input_tokens_seen": 135273320,
      "step": 6291,
      "time_per_iteration": 2.5860517024993896
    },
    {
      "auxiliary_loss_clip": 0.01140741,
      "auxiliary_loss_mlp": 0.01024134,
      "balance_loss_clip": 1.04739046,
      "balance_loss_mlp": 1.01674843,
      "epoch": 0.7565682679011604,
      "flos": 21215521918080.0,
      "grad_norm": 1.9521026765483454,
      "language_loss": 0.6802299,
      "learning_rate": 5.899235865526456e-07,
      "loss": 0.70187867,
      "num_input_tokens_seen": 135292615,
      "step": 6292,
      "time_per_iteration": 2.730410099029541
    },
    {
      "auxiliary_loss_clip": 0.01148699,
      "auxiliary_loss_mlp": 0.01025208,
      "balance_loss_clip": 1.05283511,
      "balance_loss_mlp": 1.01786458,
      "epoch": 0.7566885107917994,
      "flos": 20449008662400.0,
      "grad_norm": 2.108496313569736,
      "language_loss": 0.8212055,
      "learning_rate": 5.893712707254825e-07,
      "loss": 0.84294456,
      "num_input_tokens_seen": 135310075,
      "step": 6293,
      "time_per_iteration": 2.6660892963409424
    },
    {
      "auxiliary_loss_clip": 0.01156298,
      "auxiliary_loss_mlp": 0.00762728,
      "balance_loss_clip": 1.04963076,
      "balance_loss_mlp": 1.00060487,
      "epoch": 0.7568087536824385,
      "flos": 19025832919680.0,
      "grad_norm": 3.616173865952878,
      "language_loss": 0.66003108,
      "learning_rate": 5.888191688895769e-07,
      "loss": 0.67922127,
      "num_input_tokens_seen": 135327335,
      "step": 6294,
      "time_per_iteration": 2.7083287239074707
    },
    {
      "auxiliary_loss_clip": 0.01146812,
      "auxiliary_loss_mlp": 0.01026962,
      "balance_loss_clip": 1.0466162,
      "balance_loss_mlp": 1.01871896,
      "epoch": 0.7569289965730777,
      "flos": 15225442248960.0,
      "grad_norm": 2.4892724415713525,
      "language_loss": 0.61961484,
      "learning_rate": 5.882672811286813e-07,
      "loss": 0.64135259,
      "num_input_tokens_seen": 135343615,
      "step": 6295,
      "time_per_iteration": 3.618645191192627
    },
    {
      "auxiliary_loss_clip": 0.01164396,
      "auxiliary_loss_mlp": 0.01029633,
      "balance_loss_clip": 1.05183721,
      "balance_loss_mlp": 1.02112103,
      "epoch": 0.7570492394637167,
      "flos": 20769367086720.0,
      "grad_norm": 2.244684900121997,
      "language_loss": 0.69335854,
      "learning_rate": 5.877156075265166e-07,
      "loss": 0.71529877,
      "num_input_tokens_seen": 135359880,
      "step": 6296,
      "time_per_iteration": 2.6304969787597656
    },
    {
      "auxiliary_loss_clip": 0.01099767,
      "auxiliary_loss_mlp": 0.01031011,
      "balance_loss_clip": 1.04216731,
      "balance_loss_mlp": 1.02358449,
      "epoch": 0.7571694823543558,
      "flos": 15664091137920.0,
      "grad_norm": 2.5598339985790486,
      "language_loss": 0.69406402,
      "learning_rate": 5.871641481667715e-07,
      "loss": 0.71537179,
      "num_input_tokens_seen": 135374325,
      "step": 6297,
      "time_per_iteration": 2.725959300994873
    },
    {
      "auxiliary_loss_clip": 0.01163573,
      "auxiliary_loss_mlp": 0.01029767,
      "balance_loss_clip": 1.05088484,
      "balance_loss_mlp": 1.0213151,
      "epoch": 0.7572897252449949,
      "flos": 25409241492480.0,
      "grad_norm": 1.6162511801578856,
      "language_loss": 0.84624636,
      "learning_rate": 5.866129031331011e-07,
      "loss": 0.8681798,
      "num_input_tokens_seen": 135393980,
      "step": 6298,
      "time_per_iteration": 2.6830146312713623
    },
    {
      "auxiliary_loss_clip": 0.01151025,
      "auxiliary_loss_mlp": 0.0102637,
      "balance_loss_clip": 1.04781651,
      "balance_loss_mlp": 1.01839519,
      "epoch": 0.757409968135634,
      "flos": 24279348297600.0,
      "grad_norm": 1.9904598198129775,
      "language_loss": 0.83506233,
      "learning_rate": 5.8606187250913e-07,
      "loss": 0.85683632,
      "num_input_tokens_seen": 135412030,
      "step": 6299,
      "time_per_iteration": 2.734898328781128
    },
    {
      "auxiliary_loss_clip": 0.01129623,
      "auxiliary_loss_mlp": 0.01030548,
      "balance_loss_clip": 1.05320597,
      "balance_loss_mlp": 1.02271569,
      "epoch": 0.757530211026273,
      "flos": 24133766474880.0,
      "grad_norm": 1.9646265925200697,
      "language_loss": 0.84218216,
      "learning_rate": 5.855110563784482e-07,
      "loss": 0.86378396,
      "num_input_tokens_seen": 135430565,
      "step": 6300,
      "time_per_iteration": 2.6996543407440186
    },
    {
      "auxiliary_loss_clip": 0.01121212,
      "auxiliary_loss_mlp": 0.01025245,
      "balance_loss_clip": 1.04182374,
      "balance_loss_mlp": 1.01796401,
      "epoch": 0.7576504539169122,
      "flos": 23951807153280.0,
      "grad_norm": 1.8573136908320178,
      "language_loss": 0.64427799,
      "learning_rate": 5.849604548246156e-07,
      "loss": 0.66574258,
      "num_input_tokens_seen": 135451675,
      "step": 6301,
      "time_per_iteration": 2.798739433288574
    },
    {
      "auxiliary_loss_clip": 0.01151073,
      "auxiliary_loss_mlp": 0.01025806,
      "balance_loss_clip": 1.04850316,
      "balance_loss_mlp": 1.01752734,
      "epoch": 0.7577706968075513,
      "flos": 21251360712960.0,
      "grad_norm": 3.571092453677707,
      "language_loss": 0.80533922,
      "learning_rate": 5.844100679311565e-07,
      "loss": 0.82710803,
      "num_input_tokens_seen": 135470635,
      "step": 6302,
      "time_per_iteration": 2.6620192527770996
    },
    {
      "auxiliary_loss_clip": 0.01161771,
      "auxiliary_loss_mlp": 0.01028806,
      "balance_loss_clip": 1.05118525,
      "balance_loss_mlp": 1.02059805,
      "epoch": 0.7578909396981903,
      "flos": 18296595002880.0,
      "grad_norm": 2.0873611361006237,
      "language_loss": 0.76695025,
      "learning_rate": 5.838598957815637e-07,
      "loss": 0.78885597,
      "num_input_tokens_seen": 135487865,
      "step": 6303,
      "time_per_iteration": 2.680634021759033
    },
    {
      "auxiliary_loss_clip": 0.01143043,
      "auxiliary_loss_mlp": 0.01024532,
      "balance_loss_clip": 1.04813051,
      "balance_loss_mlp": 1.01656902,
      "epoch": 0.7580111825888295,
      "flos": 25373869574400.0,
      "grad_norm": 1.6826047455017943,
      "language_loss": 0.85324323,
      "learning_rate": 5.833099384592996e-07,
      "loss": 0.874919,
      "num_input_tokens_seen": 135508440,
      "step": 6304,
      "time_per_iteration": 2.8476040363311768
    },
    {
      "auxiliary_loss_clip": 0.01163763,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.04988563,
      "balance_loss_mlp": 1.02117491,
      "epoch": 0.7581314254794685,
      "flos": 23768662682880.0,
      "grad_norm": 2.44333563948049,
      "language_loss": 0.7168144,
      "learning_rate": 5.827601960477913e-07,
      "loss": 0.73874164,
      "num_input_tokens_seen": 135526365,
      "step": 6305,
      "time_per_iteration": 2.6552863121032715
    },
    {
      "auxiliary_loss_clip": 0.01147627,
      "auxiliary_loss_mlp": 0.00761438,
      "balance_loss_clip": 1.04779375,
      "balance_loss_mlp": 1.00055456,
      "epoch": 0.7582516683701076,
      "flos": 22054610603520.0,
      "grad_norm": 1.7677485006268139,
      "language_loss": 0.70851624,
      "learning_rate": 5.822106686304344e-07,
      "loss": 0.72760689,
      "num_input_tokens_seen": 135545655,
      "step": 6306,
      "time_per_iteration": 2.670142412185669
    },
    {
      "auxiliary_loss_clip": 0.01133615,
      "auxiliary_loss_mlp": 0.01029425,
      "balance_loss_clip": 1.0481596,
      "balance_loss_mlp": 1.02166128,
      "epoch": 0.7583719112607467,
      "flos": 31649725848960.0,
      "grad_norm": 1.788986846235978,
      "language_loss": 0.58034837,
      "learning_rate": 5.816613562905919e-07,
      "loss": 0.60197878,
      "num_input_tokens_seen": 135566840,
      "step": 6307,
      "time_per_iteration": 2.81264328956604
    },
    {
      "auxiliary_loss_clip": 0.01136961,
      "auxiliary_loss_mlp": 0.01027944,
      "balance_loss_clip": 1.04706657,
      "balance_loss_mlp": 1.02012968,
      "epoch": 0.7584921541513858,
      "flos": 33068376478080.0,
      "grad_norm": 1.8063161729374377,
      "language_loss": 0.70112479,
      "learning_rate": 5.811122591115933e-07,
      "loss": 0.72277379,
      "num_input_tokens_seen": 135587825,
      "step": 6308,
      "time_per_iteration": 3.674799919128418
    },
    {
      "auxiliary_loss_clip": 0.01108356,
      "auxiliary_loss_mlp": 0.01025929,
      "balance_loss_clip": 1.0433259,
      "balance_loss_mlp": 1.01791239,
      "epoch": 0.7586123970420249,
      "flos": 23326350606720.0,
      "grad_norm": 3.0202517176440224,
      "language_loss": 0.71589541,
      "learning_rate": 5.805633771767376e-07,
      "loss": 0.73723823,
      "num_input_tokens_seen": 135605220,
      "step": 6309,
      "time_per_iteration": 3.642169713973999
    },
    {
      "auxiliary_loss_clip": 0.01136647,
      "auxiliary_loss_mlp": 0.00761407,
      "balance_loss_clip": 1.04943442,
      "balance_loss_mlp": 1.00059366,
      "epoch": 0.7587326399326639,
      "flos": 18334229477760.0,
      "grad_norm": 1.8803958746317304,
      "language_loss": 0.77534014,
      "learning_rate": 5.800147105692888e-07,
      "loss": 0.7943207,
      "num_input_tokens_seen": 135624795,
      "step": 6310,
      "time_per_iteration": 2.7643795013427734
    },
    {
      "auxiliary_loss_clip": 0.0114374,
      "auxiliary_loss_mlp": 0.01027664,
      "balance_loss_clip": 1.04953587,
      "balance_loss_mlp": 1.02003431,
      "epoch": 0.7588528828233031,
      "flos": 17275080119040.0,
      "grad_norm": 1.6412114197411638,
      "language_loss": 0.79415154,
      "learning_rate": 5.794662593724795e-07,
      "loss": 0.81586552,
      "num_input_tokens_seen": 135643800,
      "step": 6311,
      "time_per_iteration": 2.7159676551818848
    },
    {
      "auxiliary_loss_clip": 0.01164056,
      "auxiliary_loss_mlp": 0.01030067,
      "balance_loss_clip": 1.0534023,
      "balance_loss_mlp": 1.02169824,
      "epoch": 0.7589731257139422,
      "flos": 17713621267200.0,
      "grad_norm": 8.090809353266533,
      "language_loss": 0.75386041,
      "learning_rate": 5.789180236695091e-07,
      "loss": 0.77580166,
      "num_input_tokens_seen": 135660655,
      "step": 6312,
      "time_per_iteration": 2.5880424976348877
    },
    {
      "auxiliary_loss_clip": 0.01138992,
      "auxiliary_loss_mlp": 0.01027343,
      "balance_loss_clip": 1.0460465,
      "balance_loss_mlp": 1.0197314,
      "epoch": 0.7590933686045812,
      "flos": 15961072786560.0,
      "grad_norm": 2.337456472625016,
      "language_loss": 0.85438955,
      "learning_rate": 5.78370003543544e-07,
      "loss": 0.87605286,
      "num_input_tokens_seen": 135679410,
      "step": 6313,
      "time_per_iteration": 3.644684314727783
    },
    {
      "auxiliary_loss_clip": 0.0113465,
      "auxiliary_loss_mlp": 0.01031297,
      "balance_loss_clip": 1.05013847,
      "balance_loss_mlp": 1.02312529,
      "epoch": 0.7592136114952204,
      "flos": 21068072588160.0,
      "grad_norm": 2.0178549978185987,
      "language_loss": 0.84148496,
      "learning_rate": 5.778221990777203e-07,
      "loss": 0.8631444,
      "num_input_tokens_seen": 135697150,
      "step": 6314,
      "time_per_iteration": 2.6931769847869873
    },
    {
      "auxiliary_loss_clip": 0.01145829,
      "auxiliary_loss_mlp": 0.01033865,
      "balance_loss_clip": 1.04752231,
      "balance_loss_mlp": 1.02581215,
      "epoch": 0.7593338543858594,
      "flos": 25297666871040.0,
      "grad_norm": 2.2362217085664637,
      "language_loss": 0.8288936,
      "learning_rate": 5.772746103551372e-07,
      "loss": 0.85069048,
      "num_input_tokens_seen": 135712545,
      "step": 6315,
      "time_per_iteration": 2.7503089904785156
    },
    {
      "auxiliary_loss_clip": 0.01072634,
      "auxiliary_loss_mlp": 0.0102888,
      "balance_loss_clip": 1.04090524,
      "balance_loss_mlp": 1.02086902,
      "epoch": 0.7594540972764985,
      "flos": 31832367528960.0,
      "grad_norm": 2.0780882277665267,
      "language_loss": 0.71703166,
      "learning_rate": 5.767272374588648e-07,
      "loss": 0.73804677,
      "num_input_tokens_seen": 135733950,
      "step": 6316,
      "time_per_iteration": 2.9429690837860107
    },
    {
      "auxiliary_loss_clip": 0.01092842,
      "auxiliary_loss_mlp": 0.01034777,
      "balance_loss_clip": 1.04322159,
      "balance_loss_mlp": 1.02625334,
      "epoch": 0.7595743401671377,
      "flos": 37597250880000.0,
      "grad_norm": 2.2164339231482386,
      "language_loss": 0.7850455,
      "learning_rate": 5.76180080471939e-07,
      "loss": 0.80632162,
      "num_input_tokens_seen": 135757120,
      "step": 6317,
      "time_per_iteration": 2.917818546295166
    },
    {
      "auxiliary_loss_clip": 0.01164601,
      "auxiliary_loss_mlp": 0.01028459,
      "balance_loss_clip": 1.05227995,
      "balance_loss_mlp": 1.01919675,
      "epoch": 0.7596945830577767,
      "flos": 18287724343680.0,
      "grad_norm": 1.9738566973080962,
      "language_loss": 0.72426409,
      "learning_rate": 5.756331394773631e-07,
      "loss": 0.74619472,
      "num_input_tokens_seen": 135773335,
      "step": 6318,
      "time_per_iteration": 2.707504987716675
    },
    {
      "auxiliary_loss_clip": 0.01154001,
      "auxiliary_loss_mlp": 0.01032772,
      "balance_loss_clip": 1.04795671,
      "balance_loss_mlp": 1.02489853,
      "epoch": 0.7598148259484158,
      "flos": 22233122219520.0,
      "grad_norm": 1.9778308878663446,
      "language_loss": 0.76153731,
      "learning_rate": 5.750864145581071e-07,
      "loss": 0.78340507,
      "num_input_tokens_seen": 135792555,
      "step": 6319,
      "time_per_iteration": 2.672657012939453
    },
    {
      "auxiliary_loss_clip": 0.01120573,
      "auxiliary_loss_mlp": 0.01025838,
      "balance_loss_clip": 1.03999424,
      "balance_loss_mlp": 1.01843786,
      "epoch": 0.7599350688390549,
      "flos": 27161718145920.0,
      "grad_norm": 2.471309966236405,
      "language_loss": 0.86238337,
      "learning_rate": 5.745399057971085e-07,
      "loss": 0.88384748,
      "num_input_tokens_seen": 135813690,
      "step": 6320,
      "time_per_iteration": 3.74704647064209
    },
    {
      "auxiliary_loss_clip": 0.01178299,
      "auxiliary_loss_mlp": 0.01030293,
      "balance_loss_clip": 1.05393076,
      "balance_loss_mlp": 1.02219319,
      "epoch": 0.760055311729694,
      "flos": 15560704817280.0,
      "grad_norm": 2.1713464697322955,
      "language_loss": 0.75395095,
      "learning_rate": 5.739936132772738e-07,
      "loss": 0.77603686,
      "num_input_tokens_seen": 135832255,
      "step": 6321,
      "time_per_iteration": 2.603825807571411
    },
    {
      "auxiliary_loss_clip": 0.01097752,
      "auxiliary_loss_mlp": 0.01032606,
      "balance_loss_clip": 1.04205263,
      "balance_loss_mlp": 1.02380204,
      "epoch": 0.760175554620333,
      "flos": 25155496840320.0,
      "grad_norm": 2.092508598504786,
      "language_loss": 0.74437767,
      "learning_rate": 5.734475370814733e-07,
      "loss": 0.76568127,
      "num_input_tokens_seen": 135851935,
      "step": 6322,
      "time_per_iteration": 2.8027877807617188
    },
    {
      "auxiliary_loss_clip": 0.01161792,
      "auxiliary_loss_mlp": 0.01028524,
      "balance_loss_clip": 1.04796481,
      "balance_loss_mlp": 1.02042365,
      "epoch": 0.7602957975109722,
      "flos": 24353791234560.0,
      "grad_norm": 1.540742313855119,
      "language_loss": 0.78645581,
      "learning_rate": 5.729016772925483e-07,
      "loss": 0.80835891,
      "num_input_tokens_seen": 135873510,
      "step": 6323,
      "time_per_iteration": 2.675870656967163
    },
    {
      "auxiliary_loss_clip": 0.01108145,
      "auxiliary_loss_mlp": 0.01031216,
      "balance_loss_clip": 1.0440402,
      "balance_loss_mlp": 1.02276373,
      "epoch": 0.7604160404016113,
      "flos": 25192664438400.0,
      "grad_norm": 1.8040540770151368,
      "language_loss": 0.70787859,
      "learning_rate": 5.723560339933038e-07,
      "loss": 0.72927213,
      "num_input_tokens_seen": 135893845,
      "step": 6324,
      "time_per_iteration": 2.7587194442749023
    },
    {
      "auxiliary_loss_clip": 0.01131189,
      "auxiliary_loss_mlp": 0.01026421,
      "balance_loss_clip": 1.04586124,
      "balance_loss_mlp": 1.01826715,
      "epoch": 0.7605362832922503,
      "flos": 29861841363840.0,
      "grad_norm": 3.93751566595438,
      "language_loss": 0.65629965,
      "learning_rate": 5.71810607266513e-07,
      "loss": 0.67787576,
      "num_input_tokens_seen": 135912430,
      "step": 6325,
      "time_per_iteration": 2.808436870574951
    },
    {
      "auxiliary_loss_clip": 0.01103358,
      "auxiliary_loss_mlp": 0.01027158,
      "balance_loss_clip": 1.04358494,
      "balance_loss_mlp": 1.01951993,
      "epoch": 0.7606565261828895,
      "flos": 13917935278080.0,
      "grad_norm": 2.3068646703936246,
      "language_loss": 0.60423112,
      "learning_rate": 5.712653971949184e-07,
      "loss": 0.62553632,
      "num_input_tokens_seen": 135930550,
      "step": 6326,
      "time_per_iteration": 2.7906370162963867
    },
    {
      "auxiliary_loss_clip": 0.01124297,
      "auxiliary_loss_mlp": 0.01028351,
      "balance_loss_clip": 1.04855275,
      "balance_loss_mlp": 1.0203464,
      "epoch": 0.7607767690735285,
      "flos": 18551273408640.0,
      "grad_norm": 3.0531960834588205,
      "language_loss": 0.75869,
      "learning_rate": 5.707204038612268e-07,
      "loss": 0.78021657,
      "num_input_tokens_seen": 135947980,
      "step": 6327,
      "time_per_iteration": 2.7111880779266357
    },
    {
      "auxiliary_loss_clip": 0.01133545,
      "auxiliary_loss_mlp": 0.01039247,
      "balance_loss_clip": 1.05110371,
      "balance_loss_mlp": 1.03012168,
      "epoch": 0.7608970119641676,
      "flos": 20922993555840.0,
      "grad_norm": 2.110059108789008,
      "language_loss": 0.74057496,
      "learning_rate": 5.701756273481138e-07,
      "loss": 0.76230288,
      "num_input_tokens_seen": 135965400,
      "step": 6328,
      "time_per_iteration": 2.658849000930786
    },
    {
      "auxiliary_loss_clip": 0.01175834,
      "auxiliary_loss_mlp": 0.01028291,
      "balance_loss_clip": 1.05322218,
      "balance_loss_mlp": 1.02129364,
      "epoch": 0.7610172548548068,
      "flos": 23807302738560.0,
      "grad_norm": 2.1679297683618746,
      "language_loss": 0.74137521,
      "learning_rate": 5.696310677382212e-07,
      "loss": 0.76341641,
      "num_input_tokens_seen": 135986795,
      "step": 6329,
      "time_per_iteration": 2.7351748943328857
    },
    {
      "auxiliary_loss_clip": 0.0105029,
      "auxiliary_loss_mlp": 0.01001291,
      "balance_loss_clip": 1.00839877,
      "balance_loss_mlp": 1.00034285,
      "epoch": 0.7611374977454458,
      "flos": 66496580426880.0,
      "grad_norm": 0.8677260705262094,
      "language_loss": 0.61653656,
      "learning_rate": 5.690867251141576e-07,
      "loss": 0.63705236,
      "num_input_tokens_seen": 136053450,
      "step": 6330,
      "time_per_iteration": 3.321324110031128
    },
    {
      "auxiliary_loss_clip": 0.01085866,
      "auxiliary_loss_mlp": 0.01028622,
      "balance_loss_clip": 1.04048371,
      "balance_loss_mlp": 1.01945448,
      "epoch": 0.7612577406360849,
      "flos": 15633136592640.0,
      "grad_norm": 2.4280918032916228,
      "language_loss": 0.91729701,
      "learning_rate": 5.685425995585013e-07,
      "loss": 0.93844187,
      "num_input_tokens_seen": 136071375,
      "step": 6331,
      "time_per_iteration": 2.9946155548095703
    },
    {
      "auxiliary_loss_clip": 0.01035883,
      "auxiliary_loss_mlp": 0.01003236,
      "balance_loss_clip": 1.00944138,
      "balance_loss_mlp": 1.00222301,
      "epoch": 0.761377983526724,
      "flos": 60526253237760.0,
      "grad_norm": 0.7586564255396143,
      "language_loss": 0.58939606,
      "learning_rate": 5.679986911537935e-07,
      "loss": 0.60978723,
      "num_input_tokens_seen": 136138905,
      "step": 6332,
      "time_per_iteration": 3.678074598312378
    },
    {
      "auxiliary_loss_clip": 0.01091478,
      "auxiliary_loss_mlp": 0.01030924,
      "balance_loss_clip": 1.04220343,
      "balance_loss_mlp": 1.02299702,
      "epoch": 0.7614982264173631,
      "flos": 35772522019200.0,
      "grad_norm": 2.7969023679088174,
      "language_loss": 0.67442727,
      "learning_rate": 5.674549999825462e-07,
      "loss": 0.69565129,
      "num_input_tokens_seen": 136161720,
      "step": 6333,
      "time_per_iteration": 2.9530513286590576
    },
    {
      "auxiliary_loss_clip": 0.01033033,
      "auxiliary_loss_mlp": 0.01000958,
      "balance_loss_clip": 1.00854445,
      "balance_loss_mlp": 0.99993271,
      "epoch": 0.7616184693080021,
      "flos": 67925502345600.0,
      "grad_norm": 0.913706349774798,
      "language_loss": 0.71289438,
      "learning_rate": 5.669115261272363e-07,
      "loss": 0.73323429,
      "num_input_tokens_seen": 136222040,
      "step": 6334,
      "time_per_iteration": 4.443989038467407
    },
    {
      "auxiliary_loss_clip": 0.01131924,
      "auxiliary_loss_mlp": 0.010279,
      "balance_loss_clip": 1.04578483,
      "balance_loss_mlp": 1.01903653,
      "epoch": 0.7617387121986413,
      "flos": 20521979141760.0,
      "grad_norm": 2.3677815508589357,
      "language_loss": 0.72829187,
      "learning_rate": 5.663682696703081e-07,
      "loss": 0.74989009,
      "num_input_tokens_seen": 136240305,
      "step": 6335,
      "time_per_iteration": 3.9705350399017334
    },
    {
      "auxiliary_loss_clip": 0.01127289,
      "auxiliary_loss_mlp": 0.01031298,
      "balance_loss_clip": 1.04550004,
      "balance_loss_mlp": 1.02419043,
      "epoch": 0.7618589550892804,
      "flos": 18624495283200.0,
      "grad_norm": 1.794190166090454,
      "language_loss": 0.82201546,
      "learning_rate": 5.658252306941746e-07,
      "loss": 0.84360135,
      "num_input_tokens_seen": 136259625,
      "step": 6336,
      "time_per_iteration": 2.791944742202759
    },
    {
      "auxiliary_loss_clip": 0.01166518,
      "auxiliary_loss_mlp": 0.00763168,
      "balance_loss_clip": 1.05131876,
      "balance_loss_mlp": 1.00055182,
      "epoch": 0.7619791979799194,
      "flos": 17453735389440.0,
      "grad_norm": 2.091648253215974,
      "language_loss": 0.75912726,
      "learning_rate": 5.65282409281212e-07,
      "loss": 0.77842414,
      "num_input_tokens_seen": 136277090,
      "step": 6337,
      "time_per_iteration": 2.6184189319610596
    },
    {
      "auxiliary_loss_clip": 0.0115012,
      "auxiliary_loss_mlp": 0.00762355,
      "balance_loss_clip": 1.05171144,
      "balance_loss_mlp": 1.00055885,
      "epoch": 0.7620994408705585,
      "flos": 14137421333760.0,
      "grad_norm": 2.1582389060429517,
      "language_loss": 0.70198119,
      "learning_rate": 5.64739805513768e-07,
      "loss": 0.72110593,
      "num_input_tokens_seen": 136294635,
      "step": 6338,
      "time_per_iteration": 2.661137342453003
    },
    {
      "auxiliary_loss_clip": 0.01040869,
      "auxiliary_loss_mlp": 0.01002715,
      "balance_loss_clip": 1.00831723,
      "balance_loss_mlp": 1.00163603,
      "epoch": 0.7622196837611976,
      "flos": 70708792527360.0,
      "grad_norm": 0.7880681898585165,
      "language_loss": 0.55595946,
      "learning_rate": 5.641974194741541e-07,
      "loss": 0.57639527,
      "num_input_tokens_seen": 136350320,
      "step": 6339,
      "time_per_iteration": 4.31404447555542
    },
    {
      "auxiliary_loss_clip": 0.01041084,
      "auxiliary_loss_mlp": 0.01000506,
      "balance_loss_clip": 1.00874674,
      "balance_loss_mlp": 0.99952281,
      "epoch": 0.7623399266518367,
      "flos": 60684150447360.0,
      "grad_norm": 0.7817142510032603,
      "language_loss": 0.63621354,
      "learning_rate": 5.636552512446502e-07,
      "loss": 0.65662944,
      "num_input_tokens_seen": 136411375,
      "step": 6340,
      "time_per_iteration": 3.1022815704345703
    },
    {
      "auxiliary_loss_clip": 0.01156664,
      "auxiliary_loss_mlp": 0.01028928,
      "balance_loss_clip": 1.051525,
      "balance_loss_mlp": 1.02107835,
      "epoch": 0.7624601695424758,
      "flos": 26468893641600.0,
      "grad_norm": 1.9130090336492231,
      "language_loss": 0.78006792,
      "learning_rate": 5.631133009075027e-07,
      "loss": 0.80192387,
      "num_input_tokens_seen": 136430560,
      "step": 6341,
      "time_per_iteration": 2.7227461338043213
    },
    {
      "auxiliary_loss_clip": 0.01134736,
      "auxiliary_loss_mlp": 0.0102423,
      "balance_loss_clip": 1.04979348,
      "balance_loss_mlp": 1.0169971,
      "epoch": 0.7625804124331149,
      "flos": 19135755515520.0,
      "grad_norm": 2.0335042580722127,
      "language_loss": 0.68990999,
      "learning_rate": 5.625715685449242e-07,
      "loss": 0.71149969,
      "num_input_tokens_seen": 136448665,
      "step": 6342,
      "time_per_iteration": 2.6942861080169678
    },
    {
      "auxiliary_loss_clip": 0.01106327,
      "auxiliary_loss_mlp": 0.0102903,
      "balance_loss_clip": 1.04074931,
      "balance_loss_mlp": 1.02101958,
      "epoch": 0.762700655323754,
      "flos": 26213101914240.0,
      "grad_norm": 2.454029927155402,
      "language_loss": 0.71896005,
      "learning_rate": 5.620300542390966e-07,
      "loss": 0.74031365,
      "num_input_tokens_seen": 136469710,
      "step": 6343,
      "time_per_iteration": 2.8672852516174316
    },
    {
      "auxiliary_loss_clip": 0.01148956,
      "auxiliary_loss_mlp": 0.01024563,
      "balance_loss_clip": 1.04913485,
      "balance_loss_mlp": 1.01739562,
      "epoch": 0.762820898214393,
      "flos": 22382582711040.0,
      "grad_norm": 3.5103033874471863,
      "language_loss": 0.85158861,
      "learning_rate": 5.614887580721659e-07,
      "loss": 0.8733238,
      "num_input_tokens_seen": 136489855,
      "step": 6344,
      "time_per_iteration": 2.7352254390716553
    },
    {
      "auxiliary_loss_clip": 0.01163147,
      "auxiliary_loss_mlp": 0.01032249,
      "balance_loss_clip": 1.05292082,
      "balance_loss_mlp": 1.02409458,
      "epoch": 0.7629411411050322,
      "flos": 15700504550400.0,
      "grad_norm": 2.65985757745682,
      "language_loss": 0.740592,
      "learning_rate": 5.609476801262481e-07,
      "loss": 0.762546,
      "num_input_tokens_seen": 136504715,
      "step": 6345,
      "time_per_iteration": 2.6948091983795166
    },
    {
      "auxiliary_loss_clip": 0.01170729,
      "auxiliary_loss_mlp": 0.01034954,
      "balance_loss_clip": 1.05336142,
      "balance_loss_mlp": 1.02726173,
      "epoch": 0.7630613839956712,
      "flos": 13770342293760.0,
      "grad_norm": 2.2181880551595885,
      "language_loss": 0.64208531,
      "learning_rate": 5.604068204834223e-07,
      "loss": 0.66414213,
      "num_input_tokens_seen": 136521610,
      "step": 6346,
      "time_per_iteration": 3.578254461288452
    },
    {
      "auxiliary_loss_clip": 0.01147193,
      "auxiliary_loss_mlp": 0.0102957,
      "balance_loss_clip": 1.04751575,
      "balance_loss_mlp": 1.02179134,
      "epoch": 0.7631816268863103,
      "flos": 14569569861120.0,
      "grad_norm": 2.9406600112512735,
      "language_loss": 0.76984513,
      "learning_rate": 5.598661792257367e-07,
      "loss": 0.7916128,
      "num_input_tokens_seen": 136538655,
      "step": 6347,
      "time_per_iteration": 2.654320478439331
    },
    {
      "auxiliary_loss_clip": 0.01123984,
      "auxiliary_loss_mlp": 0.01023917,
      "balance_loss_clip": 1.04563999,
      "balance_loss_mlp": 1.01636159,
      "epoch": 0.7633018697769495,
      "flos": 19062210418560.0,
      "grad_norm": 2.019204327237246,
      "language_loss": 0.76224113,
      "learning_rate": 5.593257564352071e-07,
      "loss": 0.78372014,
      "num_input_tokens_seen": 136557095,
      "step": 6348,
      "time_per_iteration": 2.7644922733306885
    },
    {
      "auxiliary_loss_clip": 0.01157995,
      "auxiliary_loss_mlp": 0.01025526,
      "balance_loss_clip": 1.04963517,
      "balance_loss_mlp": 1.01817334,
      "epoch": 0.7634221126675885,
      "flos": 22052958577920.0,
      "grad_norm": 1.6219811782109361,
      "language_loss": 0.75708795,
      "learning_rate": 5.58785552193815e-07,
      "loss": 0.77892315,
      "num_input_tokens_seen": 136577340,
      "step": 6349,
      "time_per_iteration": 2.7498254776000977
    },
    {
      "auxiliary_loss_clip": 0.0115729,
      "auxiliary_loss_mlp": 0.01028483,
      "balance_loss_clip": 1.04841077,
      "balance_loss_mlp": 1.02105021,
      "epoch": 0.7635423555582276,
      "flos": 29382720825600.0,
      "grad_norm": 2.370641136743002,
      "language_loss": 0.75854641,
      "learning_rate": 5.582455665835086e-07,
      "loss": 0.78040409,
      "num_input_tokens_seen": 136597635,
      "step": 6350,
      "time_per_iteration": 2.712167978286743
    },
    {
      "auxiliary_loss_clip": 0.01108989,
      "auxiliary_loss_mlp": 0.01036499,
      "balance_loss_clip": 1.04465747,
      "balance_loss_mlp": 1.02765346,
      "epoch": 0.7636625984488667,
      "flos": 17784903807360.0,
      "grad_norm": 3.084061108931543,
      "language_loss": 0.72747278,
      "learning_rate": 5.577057996862036e-07,
      "loss": 0.74892765,
      "num_input_tokens_seen": 136615260,
      "step": 6351,
      "time_per_iteration": 2.700289487838745
    },
    {
      "auxiliary_loss_clip": 0.01159225,
      "auxiliary_loss_mlp": 0.0102429,
      "balance_loss_clip": 1.05022025,
      "balance_loss_mlp": 1.01706254,
      "epoch": 0.7637828413395058,
      "flos": 23734583654400.0,
      "grad_norm": 1.7352453900031535,
      "language_loss": 0.76370484,
      "learning_rate": 5.571662515837814e-07,
      "loss": 0.78553998,
      "num_input_tokens_seen": 136637220,
      "step": 6352,
      "time_per_iteration": 2.733119249343872
    },
    {
      "auxiliary_loss_clip": 0.01148497,
      "auxiliary_loss_mlp": 0.01026128,
      "balance_loss_clip": 1.04859293,
      "balance_loss_mlp": 1.0192287,
      "epoch": 0.7639030842301449,
      "flos": 36283279461120.0,
      "grad_norm": 1.761716450398269,
      "language_loss": 0.83871162,
      "learning_rate": 5.566269223580926e-07,
      "loss": 0.8604579,
      "num_input_tokens_seen": 136658930,
      "step": 6353,
      "time_per_iteration": 2.8074917793273926
    },
    {
      "auxiliary_loss_clip": 0.011596,
      "auxiliary_loss_mlp": 0.01028019,
      "balance_loss_clip": 1.05171299,
      "balance_loss_mlp": 1.02117038,
      "epoch": 0.764023327120784,
      "flos": 28878104609280.0,
      "grad_norm": 1.8902728962983635,
      "language_loss": 0.75618505,
      "learning_rate": 5.560878120909511e-07,
      "loss": 0.77806127,
      "num_input_tokens_seen": 136681530,
      "step": 6354,
      "time_per_iteration": 2.7242143154144287
    },
    {
      "auxiliary_loss_clip": 0.01041818,
      "auxiliary_loss_mlp": 0.00752018,
      "balance_loss_clip": 1.00823557,
      "balance_loss_mlp": 0.99999571,
      "epoch": 0.7641435700114231,
      "flos": 64789711067520.0,
      "grad_norm": 0.8453524490877309,
      "language_loss": 0.58550704,
      "learning_rate": 5.55548920864141e-07,
      "loss": 0.60344541,
      "num_input_tokens_seen": 136742185,
      "step": 6355,
      "time_per_iteration": 3.269174337387085
    },
    {
      "auxiliary_loss_clip": 0.01102883,
      "auxiliary_loss_mlp": 0.01021103,
      "balance_loss_clip": 1.0455153,
      "balance_loss_mlp": 1.01357484,
      "epoch": 0.7642638129020621,
      "flos": 16835784785280.0,
      "grad_norm": 1.7441617853465166,
      "language_loss": 0.7812767,
      "learning_rate": 5.550102487594113e-07,
      "loss": 0.80251658,
      "num_input_tokens_seen": 136760855,
      "step": 6356,
      "time_per_iteration": 2.7371113300323486
    },
    {
      "auxiliary_loss_clip": 0.01151698,
      "auxiliary_loss_mlp": 0.01030101,
      "balance_loss_clip": 1.04793429,
      "balance_loss_mlp": 1.02192914,
      "epoch": 0.7643840557927013,
      "flos": 30408940391040.0,
      "grad_norm": 1.5369117435395543,
      "language_loss": 0.7173205,
      "learning_rate": 5.54471795858477e-07,
      "loss": 0.73913848,
      "num_input_tokens_seen": 136780925,
      "step": 6357,
      "time_per_iteration": 2.8180148601531982
    },
    {
      "auxiliary_loss_clip": 0.01144941,
      "auxiliary_loss_mlp": 0.01025122,
      "balance_loss_clip": 1.04930222,
      "balance_loss_mlp": 1.01697457,
      "epoch": 0.7645042986833404,
      "flos": 16983234115200.0,
      "grad_norm": 2.313107298044799,
      "language_loss": 0.82857293,
      "learning_rate": 5.539335622430235e-07,
      "loss": 0.85027361,
      "num_input_tokens_seen": 136799545,
      "step": 6358,
      "time_per_iteration": 2.6685845851898193
    },
    {
      "auxiliary_loss_clip": 0.01173229,
      "auxiliary_loss_mlp": 0.01028335,
      "balance_loss_clip": 1.05113411,
      "balance_loss_mlp": 1.0208075,
      "epoch": 0.7646245415739794,
      "flos": 17311493531520.0,
      "grad_norm": 2.057842038490528,
      "language_loss": 0.74527466,
      "learning_rate": 5.533955479946975e-07,
      "loss": 0.76729035,
      "num_input_tokens_seen": 136818325,
      "step": 6359,
      "time_per_iteration": 2.6554441452026367
    },
    {
      "auxiliary_loss_clip": 0.010459,
      "auxiliary_loss_mlp": 0.00999147,
      "balance_loss_clip": 1.00830173,
      "balance_loss_mlp": 0.99820536,
      "epoch": 0.7647447844646186,
      "flos": 70402332666240.0,
      "grad_norm": 0.8558477425604886,
      "language_loss": 0.65708804,
      "learning_rate": 5.528577531951173e-07,
      "loss": 0.67753851,
      "num_input_tokens_seen": 136878730,
      "step": 6360,
      "time_per_iteration": 5.066188335418701
    },
    {
      "auxiliary_loss_clip": 0.01131757,
      "auxiliary_loss_mlp": 0.01026643,
      "balance_loss_clip": 1.04519892,
      "balance_loss_mlp": 1.01920414,
      "epoch": 0.7648650273552576,
      "flos": 17675914965120.0,
      "grad_norm": 2.048739387800681,
      "language_loss": 0.73826087,
      "learning_rate": 5.523201779258653e-07,
      "loss": 0.75984496,
      "num_input_tokens_seen": 136897705,
      "step": 6361,
      "time_per_iteration": 2.647812843322754
    },
    {
      "auxiliary_loss_clip": 0.01129483,
      "auxiliary_loss_mlp": 0.01030276,
      "balance_loss_clip": 1.043414,
      "balance_loss_mlp": 1.02188337,
      "epoch": 0.7649852702458967,
      "flos": 22162019247360.0,
      "grad_norm": 1.7570941417583328,
      "language_loss": 0.84216344,
      "learning_rate": 5.517828222684912e-07,
      "loss": 0.86376107,
      "num_input_tokens_seen": 136918360,
      "step": 6362,
      "time_per_iteration": 2.7174019813537598
    },
    {
      "auxiliary_loss_clip": 0.01041099,
      "auxiliary_loss_mlp": 0.01003382,
      "balance_loss_clip": 1.00800776,
      "balance_loss_mlp": 1.00244665,
      "epoch": 0.7651055131365359,
      "flos": 69848338227840.0,
      "grad_norm": 0.7762219781413485,
      "language_loss": 0.58964133,
      "learning_rate": 5.512456863045117e-07,
      "loss": 0.6100862,
      "num_input_tokens_seen": 136979050,
      "step": 6363,
      "time_per_iteration": 3.2340071201324463
    },
    {
      "auxiliary_loss_clip": 0.01139965,
      "auxiliary_loss_mlp": 0.01023283,
      "balance_loss_clip": 1.04655719,
      "balance_loss_mlp": 1.01503956,
      "epoch": 0.7652257560271749,
      "flos": 19464014931840.0,
      "grad_norm": 2.973749235390504,
      "language_loss": 0.742535,
      "learning_rate": 5.507087701154089e-07,
      "loss": 0.76416743,
      "num_input_tokens_seen": 136998970,
      "step": 6364,
      "time_per_iteration": 2.8962595462799072
    },
    {
      "auxiliary_loss_clip": 0.01123593,
      "auxiliary_loss_mlp": 0.0103028,
      "balance_loss_clip": 1.04639745,
      "balance_loss_mlp": 1.02296638,
      "epoch": 0.765345998917814,
      "flos": 15961108700160.0,
      "grad_norm": 2.4290906801382848,
      "language_loss": 0.75062281,
      "learning_rate": 5.50172073782634e-07,
      "loss": 0.77216154,
      "num_input_tokens_seen": 137016950,
      "step": 6365,
      "time_per_iteration": 3.6569857597351074
    },
    {
      "auxiliary_loss_clip": 0.01106814,
      "auxiliary_loss_mlp": 0.01026341,
      "balance_loss_clip": 1.04593372,
      "balance_loss_mlp": 1.01850867,
      "epoch": 0.7654662418084531,
      "flos": 23659853408640.0,
      "grad_norm": 1.8175150335374275,
      "language_loss": 0.87807947,
      "learning_rate": 5.496355973876023e-07,
      "loss": 0.89941108,
      "num_input_tokens_seen": 137036205,
      "step": 6366,
      "time_per_iteration": 2.759427309036255
    },
    {
      "auxiliary_loss_clip": 0.01123633,
      "auxiliary_loss_mlp": 0.01030257,
      "balance_loss_clip": 1.04603887,
      "balance_loss_mlp": 1.02179277,
      "epoch": 0.7655864846990922,
      "flos": 41463608878080.0,
      "grad_norm": 1.8550742546060799,
      "language_loss": 0.71368647,
      "learning_rate": 5.490993410116984e-07,
      "loss": 0.73522538,
      "num_input_tokens_seen": 137059195,
      "step": 6367,
      "time_per_iteration": 2.8986191749572754
    },
    {
      "auxiliary_loss_clip": 0.01129879,
      "auxiliary_loss_mlp": 0.01028984,
      "balance_loss_clip": 1.04813528,
      "balance_loss_mlp": 1.02128935,
      "epoch": 0.7657067275897312,
      "flos": 43142684088960.0,
      "grad_norm": 1.8154750742556363,
      "language_loss": 0.69688475,
      "learning_rate": 5.485633047362704e-07,
      "loss": 0.71847343,
      "num_input_tokens_seen": 137081200,
      "step": 6368,
      "time_per_iteration": 2.903862476348877
    },
    {
      "auxiliary_loss_clip": 0.01182785,
      "auxiliary_loss_mlp": 0.00762328,
      "balance_loss_clip": 1.05708623,
      "balance_loss_mlp": 1.00061917,
      "epoch": 0.7658269704803703,
      "flos": 17311780840320.0,
      "grad_norm": 3.185550254127103,
      "language_loss": 0.78832221,
      "learning_rate": 5.480274886426341e-07,
      "loss": 0.80777335,
      "num_input_tokens_seen": 137097840,
      "step": 6369,
      "time_per_iteration": 2.5953598022460938
    },
    {
      "auxiliary_loss_clip": 0.01145229,
      "auxiliary_loss_mlp": 0.01028391,
      "balance_loss_clip": 1.0481832,
      "balance_loss_mlp": 1.02101207,
      "epoch": 0.7659472133710095,
      "flos": 12568160977920.0,
      "grad_norm": 2.7647990625540544,
      "language_loss": 0.77851939,
      "learning_rate": 5.474918928120744e-07,
      "loss": 0.80025554,
      "num_input_tokens_seen": 137114335,
      "step": 6370,
      "time_per_iteration": 2.783806324005127
    },
    {
      "auxiliary_loss_clip": 0.01162032,
      "auxiliary_loss_mlp": 0.01030795,
      "balance_loss_clip": 1.05085242,
      "balance_loss_mlp": 1.023875,
      "epoch": 0.7660674562616485,
      "flos": 22707430335360.0,
      "grad_norm": 1.7528588791018729,
      "language_loss": 0.87527508,
      "learning_rate": 5.469565173258392e-07,
      "loss": 0.89720327,
      "num_input_tokens_seen": 137132850,
      "step": 6371,
      "time_per_iteration": 3.6575822830200195
    },
    {
      "auxiliary_loss_clip": 0.01142152,
      "auxiliary_loss_mlp": 0.01027121,
      "balance_loss_clip": 1.04961026,
      "balance_loss_mlp": 1.0186038,
      "epoch": 0.7661876991522876,
      "flos": 17056455989760.0,
      "grad_norm": 2.365234105388929,
      "language_loss": 0.6384781,
      "learning_rate": 5.464213622651454e-07,
      "loss": 0.66017085,
      "num_input_tokens_seen": 137150665,
      "step": 6372,
      "time_per_iteration": 2.7820515632629395
    },
    {
      "auxiliary_loss_clip": 0.01136237,
      "auxiliary_loss_mlp": 0.01019799,
      "balance_loss_clip": 1.04768169,
      "balance_loss_mlp": 1.01126385,
      "epoch": 0.7663079420429267,
      "flos": 20084228092800.0,
      "grad_norm": 2.229596097498973,
      "language_loss": 0.84441864,
      "learning_rate": 5.458864277111753e-07,
      "loss": 0.86597902,
      "num_input_tokens_seen": 137168500,
      "step": 6373,
      "time_per_iteration": 2.700939416885376
    },
    {
      "auxiliary_loss_clip": 0.01128463,
      "auxiliary_loss_mlp": 0.01029462,
      "balance_loss_clip": 1.04520917,
      "balance_loss_mlp": 1.02214837,
      "epoch": 0.7664281849335658,
      "flos": 12677473042560.0,
      "grad_norm": 2.249353195563891,
      "language_loss": 0.69068688,
      "learning_rate": 5.453517137450769e-07,
      "loss": 0.71226609,
      "num_input_tokens_seen": 137185075,
      "step": 6374,
      "time_per_iteration": 2.6979124546051025
    },
    {
      "auxiliary_loss_clip": 0.01143475,
      "auxiliary_loss_mlp": 0.01028264,
      "balance_loss_clip": 1.04682612,
      "balance_loss_mlp": 1.02037239,
      "epoch": 0.7665484278242048,
      "flos": 22345271458560.0,
      "grad_norm": 1.5966277283720085,
      "language_loss": 0.75940478,
      "learning_rate": 5.448172204479684e-07,
      "loss": 0.78112221,
      "num_input_tokens_seen": 137204355,
      "step": 6375,
      "time_per_iteration": 2.7749807834625244
    },
    {
      "auxiliary_loss_clip": 0.01145435,
      "auxiliary_loss_mlp": 0.01025909,
      "balance_loss_clip": 1.04858208,
      "balance_loss_mlp": 1.01788044,
      "epoch": 0.766668670714844,
      "flos": 23617909301760.0,
      "grad_norm": 9.51252516023126,
      "language_loss": 0.74789047,
      "learning_rate": 5.442829479009294e-07,
      "loss": 0.76960385,
      "num_input_tokens_seen": 137223135,
      "step": 6376,
      "time_per_iteration": 2.6433775424957275
    },
    {
      "auxiliary_loss_clip": 0.01132453,
      "auxiliary_loss_mlp": 0.01030657,
      "balance_loss_clip": 1.0486393,
      "balance_loss_mlp": 1.02199662,
      "epoch": 0.7667889136054831,
      "flos": 19427134642560.0,
      "grad_norm": 1.9523730637775845,
      "language_loss": 0.72334754,
      "learning_rate": 5.437488961850103e-07,
      "loss": 0.74497873,
      "num_input_tokens_seen": 137242935,
      "step": 6377,
      "time_per_iteration": 2.778609275817871
    },
    {
      "auxiliary_loss_clip": 0.01087609,
      "auxiliary_loss_mlp": 0.01025632,
      "balance_loss_clip": 1.03967571,
      "balance_loss_mlp": 1.017851,
      "epoch": 0.7669091564961221,
      "flos": 26866352609280.0,
      "grad_norm": 3.103643581820871,
      "language_loss": 0.75562459,
      "learning_rate": 5.432150653812258e-07,
      "loss": 0.776757,
      "num_input_tokens_seen": 137262970,
      "step": 6378,
      "time_per_iteration": 2.9375667572021484
    },
    {
      "auxiliary_loss_clip": 0.01161166,
      "auxiliary_loss_mlp": 0.01028799,
      "balance_loss_clip": 1.04871869,
      "balance_loss_mlp": 1.02115178,
      "epoch": 0.7670293993867613,
      "flos": 12385303816320.0,
      "grad_norm": 3.074082279075925,
      "language_loss": 0.83018738,
      "learning_rate": 5.42681455570557e-07,
      "loss": 0.85208702,
      "num_input_tokens_seen": 137279500,
      "step": 6379,
      "time_per_iteration": 3.1209232807159424
    },
    {
      "auxiliary_loss_clip": 0.01132818,
      "auxiliary_loss_mlp": 0.01029759,
      "balance_loss_clip": 1.04779017,
      "balance_loss_mlp": 1.02231431,
      "epoch": 0.7671496422774003,
      "flos": 21762944167680.0,
      "grad_norm": 2.038994209144066,
      "language_loss": 0.65023994,
      "learning_rate": 5.42148066833954e-07,
      "loss": 0.6718657,
      "num_input_tokens_seen": 137298745,
      "step": 6380,
      "time_per_iteration": 2.7205679416656494
    },
    {
      "auxiliary_loss_clip": 0.01124424,
      "auxiliary_loss_mlp": 0.01030926,
      "balance_loss_clip": 1.04802454,
      "balance_loss_mlp": 1.02322543,
      "epoch": 0.7672698851680394,
      "flos": 21069221823360.0,
      "grad_norm": 5.205179637869253,
      "language_loss": 0.75020361,
      "learning_rate": 5.416148992523289e-07,
      "loss": 0.77175713,
      "num_input_tokens_seen": 137317320,
      "step": 6381,
      "time_per_iteration": 2.813605785369873
    },
    {
      "auxiliary_loss_clip": 0.01123424,
      "auxiliary_loss_mlp": 0.01024399,
      "balance_loss_clip": 1.04843593,
      "balance_loss_mlp": 1.01698422,
      "epoch": 0.7673901280586786,
      "flos": 16976697840000.0,
      "grad_norm": 1.9702158911137149,
      "language_loss": 0.7870537,
      "learning_rate": 5.410819529065644e-07,
      "loss": 0.808532,
      "num_input_tokens_seen": 137335275,
      "step": 6382,
      "time_per_iteration": 2.8525807857513428
    },
    {
      "auxiliary_loss_clip": 0.01152029,
      "auxiliary_loss_mlp": 0.01029289,
      "balance_loss_clip": 1.04947925,
      "balance_loss_mlp": 1.02111149,
      "epoch": 0.7675103709493176,
      "flos": 29242669697280.0,
      "grad_norm": 3.607928838931537,
      "language_loss": 0.65355027,
      "learning_rate": 5.405492278775079e-07,
      "loss": 0.67536342,
      "num_input_tokens_seen": 137355055,
      "step": 6383,
      "time_per_iteration": 2.7045321464538574
    },
    {
      "auxiliary_loss_clip": 0.01166692,
      "auxiliary_loss_mlp": 0.01026951,
      "balance_loss_clip": 1.05256367,
      "balance_loss_mlp": 1.01817131,
      "epoch": 0.7676306138399567,
      "flos": 29023004073600.0,
      "grad_norm": 2.0209826882522015,
      "language_loss": 0.80146718,
      "learning_rate": 5.400167242459732e-07,
      "loss": 0.8234036,
      "num_input_tokens_seen": 137374015,
      "step": 6384,
      "time_per_iteration": 2.7473597526550293
    },
    {
      "auxiliary_loss_clip": 0.01162548,
      "auxiliary_loss_mlp": 0.0102332,
      "balance_loss_clip": 1.05010128,
      "balance_loss_mlp": 1.01571453,
      "epoch": 0.7677508567305958,
      "flos": 22565116650240.0,
      "grad_norm": 1.9573787312718836,
      "language_loss": 0.80891985,
      "learning_rate": 5.394844420927405e-07,
      "loss": 0.83077848,
      "num_input_tokens_seen": 137393625,
      "step": 6385,
      "time_per_iteration": 2.661816358566284
    },
    {
      "auxiliary_loss_clip": 0.01142168,
      "auxiliary_loss_mlp": 0.01032559,
      "balance_loss_clip": 1.04732537,
      "balance_loss_mlp": 1.02497721,
      "epoch": 0.7678710996212349,
      "flos": 25411432222080.0,
      "grad_norm": 2.1464636025971693,
      "language_loss": 0.73415536,
      "learning_rate": 5.389523814985562e-07,
      "loss": 0.75590265,
      "num_input_tokens_seen": 137413045,
      "step": 6386,
      "time_per_iteration": 4.666497468948364
    },
    {
      "auxiliary_loss_clip": 0.01117426,
      "auxiliary_loss_mlp": 0.01031119,
      "balance_loss_clip": 1.0445621,
      "balance_loss_mlp": 1.02291417,
      "epoch": 0.767991342511874,
      "flos": 26756825063040.0,
      "grad_norm": 2.085813201439706,
      "language_loss": 0.76309204,
      "learning_rate": 5.384205425441344e-07,
      "loss": 0.78457749,
      "num_input_tokens_seen": 137433955,
      "step": 6387,
      "time_per_iteration": 2.8097541332244873
    },
    {
      "auxiliary_loss_clip": 0.01138531,
      "auxiliary_loss_mlp": 0.00761805,
      "balance_loss_clip": 1.04551256,
      "balance_loss_mlp": 1.00057971,
      "epoch": 0.7681115854025131,
      "flos": 26359509749760.0,
      "grad_norm": 1.7373853848362675,
      "language_loss": 0.8448773,
      "learning_rate": 5.378889253101537e-07,
      "loss": 0.86388063,
      "num_input_tokens_seen": 137454510,
      "step": 6388,
      "time_per_iteration": 2.744626998901367
    },
    {
      "auxiliary_loss_clip": 0.01124714,
      "auxiliary_loss_mlp": 0.01023612,
      "balance_loss_clip": 1.04343939,
      "balance_loss_mlp": 1.01613176,
      "epoch": 0.7682318282931522,
      "flos": 23257043314560.0,
      "grad_norm": 1.8799665316008816,
      "language_loss": 0.81302971,
      "learning_rate": 5.373575298772617e-07,
      "loss": 0.83451295,
      "num_input_tokens_seen": 137473630,
      "step": 6389,
      "time_per_iteration": 2.79335618019104
    },
    {
      "auxiliary_loss_clip": 0.01041893,
      "auxiliary_loss_mlp": 0.00999833,
      "balance_loss_clip": 1.00806904,
      "balance_loss_mlp": 0.99889147,
      "epoch": 0.7683520711837912,
      "flos": 70072457137920.0,
      "grad_norm": 0.758522232184069,
      "language_loss": 0.61264896,
      "learning_rate": 5.368263563260689e-07,
      "loss": 0.6330663,
      "num_input_tokens_seen": 137538765,
      "step": 6390,
      "time_per_iteration": 3.3363945484161377
    },
    {
      "auxiliary_loss_clip": 0.01166054,
      "auxiliary_loss_mlp": 0.01028022,
      "balance_loss_clip": 1.05263853,
      "balance_loss_mlp": 1.02010059,
      "epoch": 0.7684723140744304,
      "flos": 18624890332800.0,
      "grad_norm": 1.6731344296360873,
      "language_loss": 0.6431818,
      "learning_rate": 5.362954047371537e-07,
      "loss": 0.66512251,
      "num_input_tokens_seen": 137557875,
      "step": 6391,
      "time_per_iteration": 3.5831284523010254
    },
    {
      "auxiliary_loss_clip": 0.01169853,
      "auxiliary_loss_mlp": 0.01025808,
      "balance_loss_clip": 1.05565548,
      "balance_loss_mlp": 1.01810694,
      "epoch": 0.7685925569650695,
      "flos": 27452989532160.0,
      "grad_norm": 1.767987168974722,
      "language_loss": 0.7210958,
      "learning_rate": 5.357646751910627e-07,
      "loss": 0.74305236,
      "num_input_tokens_seen": 137579055,
      "step": 6392,
      "time_per_iteration": 2.8153204917907715
    },
    {
      "auxiliary_loss_clip": 0.01164275,
      "auxiliary_loss_mlp": 0.01033745,
      "balance_loss_clip": 1.05121803,
      "balance_loss_mlp": 1.02543569,
      "epoch": 0.7687127998557085,
      "flos": 24535714642560.0,
      "grad_norm": 1.9677665403497733,
      "language_loss": 0.79860097,
      "learning_rate": 5.352341677683061e-07,
      "loss": 0.8205812,
      "num_input_tokens_seen": 137600355,
      "step": 6393,
      "time_per_iteration": 2.6708202362060547
    },
    {
      "auxiliary_loss_clip": 0.01114931,
      "auxiliary_loss_mlp": 0.0103275,
      "balance_loss_clip": 1.0456202,
      "balance_loss_mlp": 1.02532899,
      "epoch": 0.7688330427463477,
      "flos": 25155963717120.0,
      "grad_norm": 1.9125833070790252,
      "language_loss": 0.79350001,
      "learning_rate": 5.347038825493617e-07,
      "loss": 0.81497681,
      "num_input_tokens_seen": 137621885,
      "step": 6394,
      "time_per_iteration": 2.805546760559082
    },
    {
      "auxiliary_loss_clip": 0.01107563,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.04454732,
      "balance_loss_mlp": 1.02193582,
      "epoch": 0.7689532856369867,
      "flos": 21211284113280.0,
      "grad_norm": 2.2562478205617307,
      "language_loss": 0.68329656,
      "learning_rate": 5.341738196146732e-07,
      "loss": 0.70466554,
      "num_input_tokens_seen": 137640230,
      "step": 6395,
      "time_per_iteration": 2.7266461849212646
    },
    {
      "auxiliary_loss_clip": 0.01163937,
      "auxiliary_loss_mlp": 0.01026487,
      "balance_loss_clip": 1.05121517,
      "balance_loss_mlp": 1.01784182,
      "epoch": 0.7690735285276258,
      "flos": 25119083427840.0,
      "grad_norm": 8.346481853555815,
      "language_loss": 0.73147732,
      "learning_rate": 5.336439790446503e-07,
      "loss": 0.75338155,
      "num_input_tokens_seen": 137659330,
      "step": 6396,
      "time_per_iteration": 2.6990199089050293
    },
    {
      "auxiliary_loss_clip": 0.01174012,
      "auxiliary_loss_mlp": 0.01025812,
      "balance_loss_clip": 1.05186248,
      "balance_loss_mlp": 1.01804006,
      "epoch": 0.769193771418265,
      "flos": 54744020640000.0,
      "grad_norm": 2.5100983388563805,
      "language_loss": 0.62825251,
      "learning_rate": 5.331143609196711e-07,
      "loss": 0.65025079,
      "num_input_tokens_seen": 137683145,
      "step": 6397,
      "time_per_iteration": 3.908101797103882
    },
    {
      "auxiliary_loss_clip": 0.01150899,
      "auxiliary_loss_mlp": 0.00762326,
      "balance_loss_clip": 1.04659534,
      "balance_loss_mlp": 1.00049663,
      "epoch": 0.769314014308904,
      "flos": 37341890115840.0,
      "grad_norm": 1.9781865695347072,
      "language_loss": 0.77340138,
      "learning_rate": 5.325849653200758e-07,
      "loss": 0.79253364,
      "num_input_tokens_seen": 137707095,
      "step": 6398,
      "time_per_iteration": 2.7888681888580322
    },
    {
      "auxiliary_loss_clip": 0.01159911,
      "auxiliary_loss_mlp": 0.0102855,
      "balance_loss_clip": 1.05284095,
      "balance_loss_mlp": 1.02069402,
      "epoch": 0.7694342571995431,
      "flos": 20631686256000.0,
      "grad_norm": 2.865833068851492,
      "language_loss": 0.76734173,
      "learning_rate": 5.32055792326175e-07,
      "loss": 0.78922629,
      "num_input_tokens_seen": 137725520,
      "step": 6399,
      "time_per_iteration": 2.6890454292297363
    },
    {
      "auxiliary_loss_clip": 0.0114628,
      "auxiliary_loss_mlp": 0.01031448,
      "balance_loss_clip": 1.04924977,
      "balance_loss_mlp": 1.02319264,
      "epoch": 0.7695545000901821,
      "flos": 24207706621440.0,
      "grad_norm": 1.881635010547325,
      "language_loss": 0.73275822,
      "learning_rate": 5.315268420182437e-07,
      "loss": 0.7545355,
      "num_input_tokens_seen": 137744195,
      "step": 6400,
      "time_per_iteration": 2.72330641746521
    },
    {
      "auxiliary_loss_clip": 0.01180025,
      "auxiliary_loss_mlp": 0.01026584,
      "balance_loss_clip": 1.05465221,
      "balance_loss_mlp": 1.01814961,
      "epoch": 0.7696747429808213,
      "flos": 28001273708160.0,
      "grad_norm": 3.0807326000464896,
      "language_loss": 0.76537293,
      "learning_rate": 5.309981144765221e-07,
      "loss": 0.78743899,
      "num_input_tokens_seen": 137764340,
      "step": 6401,
      "time_per_iteration": 2.713146448135376
    },
    {
      "auxiliary_loss_clip": 0.01082111,
      "auxiliary_loss_mlp": 0.01025091,
      "balance_loss_clip": 1.04207027,
      "balance_loss_mlp": 1.01690137,
      "epoch": 0.7697949858714603,
      "flos": 11509550323200.0,
      "grad_norm": 2.996260076154353,
      "language_loss": 0.7596451,
      "learning_rate": 5.304696097812196e-07,
      "loss": 0.78071713,
      "num_input_tokens_seen": 137780940,
      "step": 6402,
      "time_per_iteration": 2.8461618423461914
    },
    {
      "auxiliary_loss_clip": 0.01153166,
      "auxiliary_loss_mlp": 0.01030729,
      "balance_loss_clip": 1.0502919,
      "balance_loss_mlp": 1.02265811,
      "epoch": 0.7699152287620994,
      "flos": 26688271956480.0,
      "grad_norm": 3.2701950173111527,
      "language_loss": 0.59853566,
      "learning_rate": 5.299413280125078e-07,
      "loss": 0.62037456,
      "num_input_tokens_seen": 137799250,
      "step": 6403,
      "time_per_iteration": 2.7684788703918457
    },
    {
      "auxiliary_loss_clip": 0.01140918,
      "auxiliary_loss_mlp": 0.01031498,
      "balance_loss_clip": 1.04565609,
      "balance_loss_mlp": 1.02357388,
      "epoch": 0.7700354716527386,
      "flos": 16544944362240.0,
      "grad_norm": 5.383506381043043,
      "language_loss": 0.72618455,
      "learning_rate": 5.294132692505284e-07,
      "loss": 0.74790871,
      "num_input_tokens_seen": 137817660,
      "step": 6404,
      "time_per_iteration": 2.7442123889923096
    },
    {
      "auxiliary_loss_clip": 0.01099327,
      "auxiliary_loss_mlp": 0.01028371,
      "balance_loss_clip": 1.04446268,
      "balance_loss_mlp": 1.02039599,
      "epoch": 0.7701557145433776,
      "flos": 19242733196160.0,
      "grad_norm": 2.7682822264947227,
      "language_loss": 0.79366827,
      "learning_rate": 5.288854335753861e-07,
      "loss": 0.81494522,
      "num_input_tokens_seen": 137835920,
      "step": 6405,
      "time_per_iteration": 2.7865607738494873
    },
    {
      "auxiliary_loss_clip": 0.01161043,
      "auxiliary_loss_mlp": 0.00762214,
      "balance_loss_clip": 1.04933941,
      "balance_loss_mlp": 1.00055742,
      "epoch": 0.7702759574340167,
      "flos": 31685744211840.0,
      "grad_norm": 9.28865259584993,
      "language_loss": 0.75322402,
      "learning_rate": 5.283578210671551e-07,
      "loss": 0.77245653,
      "num_input_tokens_seen": 137858160,
      "step": 6406,
      "time_per_iteration": 2.7489888668060303
    },
    {
      "auxiliary_loss_clip": 0.01154367,
      "auxiliary_loss_mlp": 0.00761871,
      "balance_loss_clip": 1.05057967,
      "balance_loss_mlp": 1.00058174,
      "epoch": 0.7703962003246558,
      "flos": 16800089644800.0,
      "grad_norm": 2.060523139063819,
      "language_loss": 0.77037013,
      "learning_rate": 5.278304318058719e-07,
      "loss": 0.78953248,
      "num_input_tokens_seen": 137876015,
      "step": 6407,
      "time_per_iteration": 2.6565492153167725
    },
    {
      "auxiliary_loss_clip": 0.01103975,
      "auxiliary_loss_mlp": 0.01033254,
      "balance_loss_clip": 1.04008186,
      "balance_loss_mlp": 1.02501035,
      "epoch": 0.7705164432152949,
      "flos": 35736072693120.0,
      "grad_norm": 2.3352859869484854,
      "language_loss": 0.79363459,
      "learning_rate": 5.273032658715411e-07,
      "loss": 0.81500685,
      "num_input_tokens_seen": 137898825,
      "step": 6408,
      "time_per_iteration": 2.9214820861816406
    },
    {
      "auxiliary_loss_clip": 0.01158507,
      "auxiliary_loss_mlp": 0.01024603,
      "balance_loss_clip": 1.04844522,
      "balance_loss_mlp": 1.01688993,
      "epoch": 0.7706366861059339,
      "flos": 23365960329600.0,
      "grad_norm": 2.1399221788734915,
      "language_loss": 0.76742828,
      "learning_rate": 5.267763233441347e-07,
      "loss": 0.78925943,
      "num_input_tokens_seen": 137919455,
      "step": 6409,
      "time_per_iteration": 2.6718294620513916
    },
    {
      "auxiliary_loss_clip": 0.01119413,
      "auxiliary_loss_mlp": 0.01028338,
      "balance_loss_clip": 1.0486412,
      "balance_loss_mlp": 1.02018356,
      "epoch": 0.7707569289965731,
      "flos": 22929897219840.0,
      "grad_norm": 4.189557938294882,
      "language_loss": 0.70339018,
      "learning_rate": 5.26249604303588e-07,
      "loss": 0.7248677,
      "num_input_tokens_seen": 137937960,
      "step": 6410,
      "time_per_iteration": 2.743694305419922
    },
    {
      "auxiliary_loss_clip": 0.01174928,
      "auxiliary_loss_mlp": 0.01024946,
      "balance_loss_clip": 1.05287528,
      "balance_loss_mlp": 1.01733458,
      "epoch": 0.7708771718872122,
      "flos": 17420661941760.0,
      "grad_norm": 2.172168406507517,
      "language_loss": 0.78332961,
      "learning_rate": 5.257231088298057e-07,
      "loss": 0.80532837,
      "num_input_tokens_seen": 137956370,
      "step": 6411,
      "time_per_iteration": 3.5127265453338623
    },
    {
      "auxiliary_loss_clip": 0.01054271,
      "auxiliary_loss_mlp": 0.01003016,
      "balance_loss_clip": 1.00858867,
      "balance_loss_mlp": 1.00207472,
      "epoch": 0.7709974147778512,
      "flos": 72241316248320.0,
      "grad_norm": 0.8153429504490397,
      "language_loss": 0.53787005,
      "learning_rate": 5.25196837002655e-07,
      "loss": 0.55844289,
      "num_input_tokens_seen": 138016080,
      "step": 6412,
      "time_per_iteration": 4.1039955615997314
    },
    {
      "auxiliary_loss_clip": 0.0112408,
      "auxiliary_loss_mlp": 0.01026648,
      "balance_loss_clip": 1.045349,
      "balance_loss_mlp": 1.01884031,
      "epoch": 0.7711176576684904,
      "flos": 39859694876160.0,
      "grad_norm": 2.2065205028842203,
      "language_loss": 0.68744981,
      "learning_rate": 5.24670788901971e-07,
      "loss": 0.70895708,
      "num_input_tokens_seen": 138039170,
      "step": 6413,
      "time_per_iteration": 2.870457410812378
    },
    {
      "auxiliary_loss_clip": 0.01125808,
      "auxiliary_loss_mlp": 0.01033868,
      "balance_loss_clip": 1.04921317,
      "balance_loss_mlp": 1.02526128,
      "epoch": 0.7712379005591294,
      "flos": 36976391274240.0,
      "grad_norm": 2.146255478699111,
      "language_loss": 0.68455565,
      "learning_rate": 5.241449646075557e-07,
      "loss": 0.70615244,
      "num_input_tokens_seen": 138062395,
      "step": 6414,
      "time_per_iteration": 2.9799368381500244
    },
    {
      "auxiliary_loss_clip": 0.01119761,
      "auxiliary_loss_mlp": 0.01028283,
      "balance_loss_clip": 1.04284036,
      "balance_loss_mlp": 1.01985526,
      "epoch": 0.7713581434497685,
      "flos": 22776773541120.0,
      "grad_norm": 2.149523670534378,
      "language_loss": 0.72476512,
      "learning_rate": 5.236193641991762e-07,
      "loss": 0.74624562,
      "num_input_tokens_seen": 138080325,
      "step": 6415,
      "time_per_iteration": 2.7539267539978027
    },
    {
      "auxiliary_loss_clip": 0.01147325,
      "auxiliary_loss_mlp": 0.01025131,
      "balance_loss_clip": 1.05031741,
      "balance_loss_mlp": 1.01714969,
      "epoch": 0.7714783863404077,
      "flos": 24097460803200.0,
      "grad_norm": 2.1999162949733275,
      "language_loss": 0.69777536,
      "learning_rate": 5.23093987756565e-07,
      "loss": 0.71949995,
      "num_input_tokens_seen": 138099020,
      "step": 6416,
      "time_per_iteration": 2.7776989936828613
    },
    {
      "auxiliary_loss_clip": 0.01125775,
      "auxiliary_loss_mlp": 0.01027692,
      "balance_loss_clip": 1.04494786,
      "balance_loss_mlp": 1.01878071,
      "epoch": 0.7715986292310467,
      "flos": 21063655215360.0,
      "grad_norm": 1.8442041379018155,
      "language_loss": 0.7557267,
      "learning_rate": 5.225688353594217e-07,
      "loss": 0.77726138,
      "num_input_tokens_seen": 138118650,
      "step": 6417,
      "time_per_iteration": 3.6786293983459473
    },
    {
      "auxiliary_loss_clip": 0.0116378,
      "auxiliary_loss_mlp": 0.01029875,
      "balance_loss_clip": 1.05386734,
      "balance_loss_mlp": 1.02217102,
      "epoch": 0.7717188721216858,
      "flos": 20594877793920.0,
      "grad_norm": 2.0550077023522006,
      "language_loss": 0.77560043,
      "learning_rate": 5.220439070874108e-07,
      "loss": 0.79753697,
      "num_input_tokens_seen": 138137890,
      "step": 6418,
      "time_per_iteration": 2.7148141860961914
    },
    {
      "auxiliary_loss_clip": 0.01168956,
      "auxiliary_loss_mlp": 0.01037807,
      "balance_loss_clip": 1.05481172,
      "balance_loss_mlp": 1.02954018,
      "epoch": 0.7718391150123249,
      "flos": 26250951870720.0,
      "grad_norm": 1.5157757558592908,
      "language_loss": 0.70933187,
      "learning_rate": 5.215192030201652e-07,
      "loss": 0.73139942,
      "num_input_tokens_seen": 138158880,
      "step": 6419,
      "time_per_iteration": 2.730255603790283
    },
    {
      "auxiliary_loss_clip": 0.01159053,
      "auxiliary_loss_mlp": 0.01028088,
      "balance_loss_clip": 1.05123007,
      "balance_loss_mlp": 1.02072072,
      "epoch": 0.771959357902964,
      "flos": 22049762267520.0,
      "grad_norm": 2.9902217109762304,
      "language_loss": 0.86370194,
      "learning_rate": 5.209947232372798e-07,
      "loss": 0.88557339,
      "num_input_tokens_seen": 138176370,
      "step": 6420,
      "time_per_iteration": 2.6555373668670654
    },
    {
      "auxiliary_loss_clip": 0.01178272,
      "auxiliary_loss_mlp": 0.01031466,
      "balance_loss_clip": 1.05340815,
      "balance_loss_mlp": 1.0236752,
      "epoch": 0.772079600793603,
      "flos": 30446000248320.0,
      "grad_norm": 1.769022959848317,
      "language_loss": 0.81578285,
      "learning_rate": 5.204704678183196e-07,
      "loss": 0.83788025,
      "num_input_tokens_seen": 138195105,
      "step": 6421,
      "time_per_iteration": 2.7138538360595703
    },
    {
      "auxiliary_loss_clip": 0.0116369,
      "auxiliary_loss_mlp": 0.01031001,
      "balance_loss_clip": 1.05213892,
      "balance_loss_mlp": 1.0227313,
      "epoch": 0.7721998436842422,
      "flos": 12969857750400.0,
      "grad_norm": 2.034272573149519,
      "language_loss": 0.85651857,
      "learning_rate": 5.19946436842813e-07,
      "loss": 0.87846541,
      "num_input_tokens_seen": 138212235,
      "step": 6422,
      "time_per_iteration": 2.5986554622650146
    },
    {
      "auxiliary_loss_clip": 0.0114518,
      "auxiliary_loss_mlp": 0.0076142,
      "balance_loss_clip": 1.04726958,
      "balance_loss_mlp": 1.00058734,
      "epoch": 0.7723200865748813,
      "flos": 32635509678720.0,
      "grad_norm": 2.126021414857862,
      "language_loss": 0.68569422,
      "learning_rate": 5.194226303902546e-07,
      "loss": 0.70476019,
      "num_input_tokens_seen": 138231970,
      "step": 6423,
      "time_per_iteration": 3.700244665145874
    },
    {
      "auxiliary_loss_clip": 0.01106938,
      "auxiliary_loss_mlp": 0.01030456,
      "balance_loss_clip": 1.0447793,
      "balance_loss_mlp": 1.02277923,
      "epoch": 0.7724403294655203,
      "flos": 21105707063040.0,
      "grad_norm": 3.0892159747869767,
      "language_loss": 0.70808554,
      "learning_rate": 5.188990485401072e-07,
      "loss": 0.72945952,
      "num_input_tokens_seen": 138251175,
      "step": 6424,
      "time_per_iteration": 2.7239034175872803
    },
    {
      "auxiliary_loss_clip": 0.01162433,
      "auxiliary_loss_mlp": 0.01027089,
      "balance_loss_clip": 1.05270326,
      "balance_loss_mlp": 1.01915526,
      "epoch": 0.7725605723561595,
      "flos": 22090736707200.0,
      "grad_norm": 2.0177761330430273,
      "language_loss": 0.86245656,
      "learning_rate": 5.183756913717954e-07,
      "loss": 0.88435185,
      "num_input_tokens_seen": 138270950,
      "step": 6425,
      "time_per_iteration": 2.7165780067443848
    },
    {
      "auxiliary_loss_clip": 0.01147908,
      "auxiliary_loss_mlp": 0.01025436,
      "balance_loss_clip": 1.05127692,
      "balance_loss_mlp": 1.01747274,
      "epoch": 0.7726808152467985,
      "flos": 34495610457600.0,
      "grad_norm": 2.277227945288217,
      "language_loss": 0.73177624,
      "learning_rate": 5.178525589647136e-07,
      "loss": 0.75350964,
      "num_input_tokens_seen": 138292590,
      "step": 6426,
      "time_per_iteration": 2.778651475906372
    },
    {
      "auxiliary_loss_clip": 0.01143369,
      "auxiliary_loss_mlp": 0.01028732,
      "balance_loss_clip": 1.04639792,
      "balance_loss_mlp": 1.02109098,
      "epoch": 0.7728010581374376,
      "flos": 22306344094080.0,
      "grad_norm": 2.172048792571827,
      "language_loss": 0.79027975,
      "learning_rate": 5.173296513982197e-07,
      "loss": 0.81200081,
      "num_input_tokens_seen": 138311115,
      "step": 6427,
      "time_per_iteration": 2.6694300174713135
    },
    {
      "auxiliary_loss_clip": 0.0116381,
      "auxiliary_loss_mlp": 0.00762315,
      "balance_loss_clip": 1.05412304,
      "balance_loss_mlp": 1.00049663,
      "epoch": 0.7729213010280768,
      "flos": 27126453968640.0,
      "grad_norm": 2.7857898550300817,
      "language_loss": 0.65258288,
      "learning_rate": 5.168069687516398e-07,
      "loss": 0.67184412,
      "num_input_tokens_seen": 138330885,
      "step": 6428,
      "time_per_iteration": 2.7555384635925293
    },
    {
      "auxiliary_loss_clip": 0.01163575,
      "auxiliary_loss_mlp": 0.0102916,
      "balance_loss_clip": 1.05265212,
      "balance_loss_mlp": 1.02096391,
      "epoch": 0.7730415439187158,
      "flos": 18150223080960.0,
      "grad_norm": 1.8549419817344752,
      "language_loss": 0.72087026,
      "learning_rate": 5.16284511104263e-07,
      "loss": 0.74279755,
      "num_input_tokens_seen": 138350020,
      "step": 6429,
      "time_per_iteration": 2.61348295211792
    },
    {
      "auxiliary_loss_clip": 0.01178445,
      "auxiliary_loss_mlp": 0.01033256,
      "balance_loss_clip": 1.05194533,
      "balance_loss_mlp": 1.02535796,
      "epoch": 0.7731617868093549,
      "flos": 11947480940160.0,
      "grad_norm": 31.12583178166849,
      "language_loss": 0.81418943,
      "learning_rate": 5.157622785353457e-07,
      "loss": 0.83630645,
      "num_input_tokens_seen": 138368135,
      "step": 6430,
      "time_per_iteration": 2.5973587036132812
    },
    {
      "auxiliary_loss_clip": 0.01053926,
      "auxiliary_loss_mlp": 0.01001816,
      "balance_loss_clip": 1.00813484,
      "balance_loss_mlp": 1.00089824,
      "epoch": 0.7732820296999939,
      "flos": 64201027069440.0,
      "grad_norm": 0.6446454985221144,
      "language_loss": 0.60290414,
      "learning_rate": 5.152402711241113e-07,
      "loss": 0.6234616,
      "num_input_tokens_seen": 138436040,
      "step": 6431,
      "time_per_iteration": 3.2758734226226807
    },
    {
      "auxiliary_loss_clip": 0.01146302,
      "auxiliary_loss_mlp": 0.01025818,
      "balance_loss_clip": 1.05128396,
      "balance_loss_mlp": 1.01800132,
      "epoch": 0.7734022725906331,
      "flos": 25302191984640.0,
      "grad_norm": 1.6810955624462853,
      "language_loss": 0.83332044,
      "learning_rate": 5.147184889497465e-07,
      "loss": 0.85504162,
      "num_input_tokens_seen": 138455510,
      "step": 6432,
      "time_per_iteration": 2.7446963787078857
    },
    {
      "auxiliary_loss_clip": 0.01175817,
      "auxiliary_loss_mlp": 0.01034963,
      "balance_loss_clip": 1.0510546,
      "balance_loss_mlp": 1.02646351,
      "epoch": 0.7735225154812722,
      "flos": 17347440067200.0,
      "grad_norm": 2.157479206782476,
      "language_loss": 0.80268395,
      "learning_rate": 5.141969320914072e-07,
      "loss": 0.82479179,
      "num_input_tokens_seen": 138473015,
      "step": 6433,
      "time_per_iteration": 2.5751795768737793
    },
    {
      "auxiliary_loss_clip": 0.01139045,
      "auxiliary_loss_mlp": 0.01028686,
      "balance_loss_clip": 1.04751849,
      "balance_loss_mlp": 1.02032399,
      "epoch": 0.7736427583719112,
      "flos": 32630086725120.0,
      "grad_norm": 11.363290982721102,
      "language_loss": 0.62978739,
      "learning_rate": 5.136756006282113e-07,
      "loss": 0.6514647,
      "num_input_tokens_seen": 138491680,
      "step": 6434,
      "time_per_iteration": 2.7785329818725586
    },
    {
      "auxiliary_loss_clip": 0.01136402,
      "auxiliary_loss_mlp": 0.01026691,
      "balance_loss_clip": 1.04650545,
      "balance_loss_mlp": 1.01857889,
      "epoch": 0.7737630012625504,
      "flos": 19860073269120.0,
      "grad_norm": 2.1419555795518983,
      "language_loss": 0.8500222,
      "learning_rate": 5.131544946392446e-07,
      "loss": 0.8716532,
      "num_input_tokens_seen": 138506960,
      "step": 6435,
      "time_per_iteration": 2.6667253971099854
    },
    {
      "auxiliary_loss_clip": 0.01125676,
      "auxiliary_loss_mlp": 0.0102986,
      "balance_loss_clip": 1.04781318,
      "balance_loss_mlp": 1.02159846,
      "epoch": 0.7738832441531894,
      "flos": 36022639397760.0,
      "grad_norm": 2.2325949995552627,
      "language_loss": 0.63981533,
      "learning_rate": 5.126336142035592e-07,
      "loss": 0.66137075,
      "num_input_tokens_seen": 138526995,
      "step": 6436,
      "time_per_iteration": 2.9217910766601562
    },
    {
      "auxiliary_loss_clip": 0.01118162,
      "auxiliary_loss_mlp": 0.01025789,
      "balance_loss_clip": 1.04554963,
      "balance_loss_mlp": 1.01846373,
      "epoch": 0.7740034870438285,
      "flos": 13405274415360.0,
      "grad_norm": 2.364940690476555,
      "language_loss": 0.72421694,
      "learning_rate": 5.121129594001721e-07,
      "loss": 0.74565649,
      "num_input_tokens_seen": 138541260,
      "step": 6437,
      "time_per_iteration": 3.6449170112609863
    },
    {
      "auxiliary_loss_clip": 0.01083866,
      "auxiliary_loss_mlp": 0.01026584,
      "balance_loss_clip": 1.04007792,
      "balance_loss_mlp": 1.0185796,
      "epoch": 0.7741237299344677,
      "flos": 22086714384000.0,
      "grad_norm": 1.770462186924809,
      "language_loss": 0.81493729,
      "learning_rate": 5.115925303080661e-07,
      "loss": 0.83604181,
      "num_input_tokens_seen": 138560970,
      "step": 6438,
      "time_per_iteration": 3.781648635864258
    },
    {
      "auxiliary_loss_clip": 0.01121573,
      "auxiliary_loss_mlp": 0.01025133,
      "balance_loss_clip": 1.04470801,
      "balance_loss_mlp": 1.01724482,
      "epoch": 0.7742439728251067,
      "flos": 19864777950720.0,
      "grad_norm": 2.4638428753983823,
      "language_loss": 0.7933647,
      "learning_rate": 5.110723270061899e-07,
      "loss": 0.81483173,
      "num_input_tokens_seen": 138577460,
      "step": 6439,
      "time_per_iteration": 2.73195219039917
    },
    {
      "auxiliary_loss_clip": 0.01173269,
      "auxiliary_loss_mlp": 0.01031437,
      "balance_loss_clip": 1.05168617,
      "balance_loss_mlp": 1.02390289,
      "epoch": 0.7743642157157458,
      "flos": 16690167048960.0,
      "grad_norm": 2.7279406869838985,
      "language_loss": 0.7970103,
      "learning_rate": 5.105523495734572e-07,
      "loss": 0.8190574,
      "num_input_tokens_seen": 138594860,
      "step": 6440,
      "time_per_iteration": 2.638916015625
    },
    {
      "auxiliary_loss_clip": 0.01137086,
      "auxiliary_loss_mlp": 0.01032874,
      "balance_loss_clip": 1.04904795,
      "balance_loss_mlp": 1.02542663,
      "epoch": 0.7744844586063849,
      "flos": 20304360593280.0,
      "grad_norm": 2.301213638939511,
      "language_loss": 0.75683266,
      "learning_rate": 5.100325980887499e-07,
      "loss": 0.77853227,
      "num_input_tokens_seen": 138614785,
      "step": 6441,
      "time_per_iteration": 2.7329819202423096
    },
    {
      "auxiliary_loss_clip": 0.0115111,
      "auxiliary_loss_mlp": 0.01024776,
      "balance_loss_clip": 1.05084062,
      "balance_loss_mlp": 1.01761174,
      "epoch": 0.774604701497024,
      "flos": 22966705681920.0,
      "grad_norm": 1.856223118400299,
      "language_loss": 0.83484131,
      "learning_rate": 5.095130726309116e-07,
      "loss": 0.85660017,
      "num_input_tokens_seen": 138634960,
      "step": 6442,
      "time_per_iteration": 2.7490057945251465
    },
    {
      "auxiliary_loss_clip": 0.01027801,
      "auxiliary_loss_mlp": 0.0100242,
      "balance_loss_clip": 1.00848138,
      "balance_loss_mlp": 1.00147843,
      "epoch": 0.774724944387663,
      "flos": 60288523073280.0,
      "grad_norm": 0.8001778574950944,
      "language_loss": 0.58921599,
      "learning_rate": 5.089937732787559e-07,
      "loss": 0.60951817,
      "num_input_tokens_seen": 138699520,
      "step": 6443,
      "time_per_iteration": 4.62028956413269
    },
    {
      "auxiliary_loss_clip": 0.01177862,
      "auxiliary_loss_mlp": 0.01031723,
      "balance_loss_clip": 1.05323553,
      "balance_loss_mlp": 1.02400994,
      "epoch": 0.7748451872783022,
      "flos": 26761026954240.0,
      "grad_norm": 2.3827959397583527,
      "language_loss": 0.6701116,
      "learning_rate": 5.084747001110592e-07,
      "loss": 0.6922074,
      "num_input_tokens_seen": 138719145,
      "step": 6444,
      "time_per_iteration": 2.637817144393921
    },
    {
      "auxiliary_loss_clip": 0.01164107,
      "auxiliary_loss_mlp": 0.00762111,
      "balance_loss_clip": 1.05352688,
      "balance_loss_mlp": 1.00058842,
      "epoch": 0.7749654301689413,
      "flos": 30338627518080.0,
      "grad_norm": 1.6920646788813094,
      "language_loss": 0.70346367,
      "learning_rate": 5.07955853206564e-07,
      "loss": 0.72272587,
      "num_input_tokens_seen": 138743850,
      "step": 6445,
      "time_per_iteration": 2.771981716156006
    },
    {
      "auxiliary_loss_clip": 0.01111525,
      "auxiliary_loss_mlp": 0.01022619,
      "balance_loss_clip": 1.0452683,
      "balance_loss_mlp": 1.01502526,
      "epoch": 0.7750856730595803,
      "flos": 43179851687040.0,
      "grad_norm": 2.0527837842985894,
      "language_loss": 0.71005398,
      "learning_rate": 5.074372326439807e-07,
      "loss": 0.73139536,
      "num_input_tokens_seen": 138766860,
      "step": 6446,
      "time_per_iteration": 2.913680076599121
    },
    {
      "auxiliary_loss_clip": 0.01140189,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.04725146,
      "balance_loss_mlp": 1.02212405,
      "epoch": 0.7752059159502195,
      "flos": 17640040256640.0,
      "grad_norm": 2.3725344093805876,
      "language_loss": 0.73623872,
      "learning_rate": 5.069188385019814e-07,
      "loss": 0.75794303,
      "num_input_tokens_seen": 138784560,
      "step": 6447,
      "time_per_iteration": 2.7308928966522217
    },
    {
      "auxiliary_loss_clip": 0.01165716,
      "auxiliary_loss_mlp": 0.01028418,
      "balance_loss_clip": 1.05057657,
      "balance_loss_mlp": 1.02004373,
      "epoch": 0.7753261588408585,
      "flos": 12677688524160.0,
      "grad_norm": 2.924588778478469,
      "language_loss": 0.61645472,
      "learning_rate": 5.064006708592077e-07,
      "loss": 0.63839602,
      "num_input_tokens_seen": 138800805,
      "step": 6448,
      "time_per_iteration": 2.564300298690796
    },
    {
      "auxiliary_loss_clip": 0.01099656,
      "auxiliary_loss_mlp": 0.0102649,
      "balance_loss_clip": 1.0448916,
      "balance_loss_mlp": 1.01891756,
      "epoch": 0.7754464017314976,
      "flos": 16690741666560.0,
      "grad_norm": 2.8456062149780643,
      "language_loss": 0.76024228,
      "learning_rate": 5.058827297942641e-07,
      "loss": 0.7815038,
      "num_input_tokens_seen": 138815910,
      "step": 6449,
      "time_per_iteration": 3.7232749462127686
    },
    {
      "auxiliary_loss_clip": 0.01155815,
      "auxiliary_loss_mlp": 0.01025954,
      "balance_loss_clip": 1.04904962,
      "balance_loss_mlp": 1.01800299,
      "epoch": 0.7755666446221368,
      "flos": 19718944732800.0,
      "grad_norm": 9.56913397445243,
      "language_loss": 0.75336552,
      "learning_rate": 5.053650153857237e-07,
      "loss": 0.7751832,
      "num_input_tokens_seen": 138834920,
      "step": 6450,
      "time_per_iteration": 2.7125484943389893
    },
    {
      "auxiliary_loss_clip": 0.01136301,
      "auxiliary_loss_mlp": 0.00761404,
      "balance_loss_clip": 1.04727483,
      "balance_loss_mlp": 1.00054145,
      "epoch": 0.7756868875127758,
      "flos": 18693623007360.0,
      "grad_norm": 2.2362531328759485,
      "language_loss": 0.70357955,
      "learning_rate": 5.048475277121214e-07,
      "loss": 0.72255659,
      "num_input_tokens_seen": 138852135,
      "step": 6451,
      "time_per_iteration": 2.6666595935821533
    },
    {
      "auxiliary_loss_clip": 0.0115763,
      "auxiliary_loss_mlp": 0.00761737,
      "balance_loss_clip": 1.0494262,
      "balance_loss_mlp": 1.00052488,
      "epoch": 0.7758071304034149,
      "flos": 28404191543040.0,
      "grad_norm": 1.8002526377874377,
      "language_loss": 0.76922524,
      "learning_rate": 5.043302668519598e-07,
      "loss": 0.78841895,
      "num_input_tokens_seen": 138871470,
      "step": 6452,
      "time_per_iteration": 2.771744728088379
    },
    {
      "auxiliary_loss_clip": 0.01175484,
      "auxiliary_loss_mlp": 0.01027529,
      "balance_loss_clip": 1.05351686,
      "balance_loss_mlp": 1.01919055,
      "epoch": 0.775927373294054,
      "flos": 20595344670720.0,
      "grad_norm": 2.76309143667217,
      "language_loss": 0.72509354,
      "learning_rate": 5.038132328837079e-07,
      "loss": 0.7471236,
      "num_input_tokens_seen": 138889860,
      "step": 6453,
      "time_per_iteration": 2.6062569618225098
    },
    {
      "auxiliary_loss_clip": 0.01088061,
      "auxiliary_loss_mlp": 0.0102668,
      "balance_loss_clip": 1.0395546,
      "balance_loss_mlp": 1.01882982,
      "epoch": 0.7760476161846931,
      "flos": 22526368853760.0,
      "grad_norm": 2.354912323139526,
      "language_loss": 0.74386632,
      "learning_rate": 5.032964258857993e-07,
      "loss": 0.76501375,
      "num_input_tokens_seen": 138909955,
      "step": 6454,
      "time_per_iteration": 2.9108448028564453
    },
    {
      "auxiliary_loss_clip": 0.0111206,
      "auxiliary_loss_mlp": 0.00762121,
      "balance_loss_clip": 1.04387283,
      "balance_loss_mlp": 1.00048041,
      "epoch": 0.7761678590753321,
      "flos": 48651488403840.0,
      "grad_norm": 2.0069910096418613,
      "language_loss": 0.68625081,
      "learning_rate": 5.027798459366329e-07,
      "loss": 0.70499265,
      "num_input_tokens_seen": 138935320,
      "step": 6455,
      "time_per_iteration": 2.9981119632720947
    },
    {
      "auxiliary_loss_clip": 0.01139367,
      "auxiliary_loss_mlp": 0.01034807,
      "balance_loss_clip": 1.04876304,
      "balance_loss_mlp": 1.02672434,
      "epoch": 0.7762881019659713,
      "flos": 26177047637760.0,
      "grad_norm": 1.5265898521438583,
      "language_loss": 0.6391108,
      "learning_rate": 5.02263493114573e-07,
      "loss": 0.66085255,
      "num_input_tokens_seen": 138957115,
      "step": 6456,
      "time_per_iteration": 2.8894007205963135
    },
    {
      "auxiliary_loss_clip": 0.01094873,
      "auxiliary_loss_mlp": 0.01022164,
      "balance_loss_clip": 1.04267025,
      "balance_loss_mlp": 1.01449299,
      "epoch": 0.7764083448566104,
      "flos": 20588341518720.0,
      "grad_norm": 9.190546415330926,
      "language_loss": 0.76640338,
      "learning_rate": 5.017473674979502e-07,
      "loss": 0.78757375,
      "num_input_tokens_seen": 138973140,
      "step": 6457,
      "time_per_iteration": 2.874450445175171
    },
    {
      "auxiliary_loss_clip": 0.01008331,
      "auxiliary_loss_mlp": 0.01003435,
      "balance_loss_clip": 1.00786066,
      "balance_loss_mlp": 1.0024513,
      "epoch": 0.7765285877472494,
      "flos": 67293078560640.0,
      "grad_norm": 0.7439882553994707,
      "language_loss": 0.58178097,
      "learning_rate": 5.01231469165061e-07,
      "loss": 0.60189861,
      "num_input_tokens_seen": 139028965,
      "step": 6458,
      "time_per_iteration": 3.210190773010254
    },
    {
      "auxiliary_loss_clip": 0.01051775,
      "auxiliary_loss_mlp": 0.01001519,
      "balance_loss_clip": 1.00920475,
      "balance_loss_mlp": 1.00058937,
      "epoch": 0.7766488306378886,
      "flos": 61344476121600.0,
      "grad_norm": 0.8273635435936552,
      "language_loss": 0.56753922,
      "learning_rate": 5.007157981941663e-07,
      "loss": 0.58807218,
      "num_input_tokens_seen": 139094325,
      "step": 6459,
      "time_per_iteration": 3.3557043075561523
    },
    {
      "auxiliary_loss_clip": 0.01020729,
      "auxiliary_loss_mlp": 0.01000932,
      "balance_loss_clip": 1.00900054,
      "balance_loss_mlp": 0.99998397,
      "epoch": 0.7767690735285276,
      "flos": 62946199393920.0,
      "grad_norm": 0.8771650795557407,
      "language_loss": 0.67317438,
      "learning_rate": 5.002003546634928e-07,
      "loss": 0.69339097,
      "num_input_tokens_seen": 139150425,
      "step": 6460,
      "time_per_iteration": 3.2287368774414062
    },
    {
      "auxiliary_loss_clip": 0.01112291,
      "auxiliary_loss_mlp": 0.01027263,
      "balance_loss_clip": 1.0439682,
      "balance_loss_mlp": 1.02013445,
      "epoch": 0.7768893164191667,
      "flos": 20886400575360.0,
      "grad_norm": 11.900544026634169,
      "language_loss": 0.76253879,
      "learning_rate": 4.996851386512331e-07,
      "loss": 0.78393424,
      "num_input_tokens_seen": 139169130,
      "step": 6461,
      "time_per_iteration": 2.737602472305298
    },
    {
      "auxiliary_loss_clip": 0.01112954,
      "auxiliary_loss_mlp": 0.00763231,
      "balance_loss_clip": 1.04098797,
      "balance_loss_mlp": 1.0005722,
      "epoch": 0.7770095593098058,
      "flos": 20704584908160.0,
      "grad_norm": 2.5993435402777627,
      "language_loss": 0.83239013,
      "learning_rate": 4.991701502355444e-07,
      "loss": 0.85115194,
      "num_input_tokens_seen": 139189595,
      "step": 6462,
      "time_per_iteration": 2.7813968658447266
    },
    {
      "auxiliary_loss_clip": 0.01128972,
      "auxiliary_loss_mlp": 0.01024977,
      "balance_loss_clip": 1.04573393,
      "balance_loss_mlp": 1.01772881,
      "epoch": 0.7771298022004449,
      "flos": 24717709877760.0,
      "grad_norm": 1.5442471582771888,
      "language_loss": 0.75764406,
      "learning_rate": 4.986553894945518e-07,
      "loss": 0.77918351,
      "num_input_tokens_seen": 139210805,
      "step": 6463,
      "time_per_iteration": 3.6172165870666504
    },
    {
      "auxiliary_loss_clip": 0.01147701,
      "auxiliary_loss_mlp": 0.00761264,
      "balance_loss_clip": 1.04575765,
      "balance_loss_mlp": 1.00053251,
      "epoch": 0.777250045091084,
      "flos": 25009232659200.0,
      "grad_norm": 2.7443225612583086,
      "language_loss": 0.86291981,
      "learning_rate": 4.981408565063416e-07,
      "loss": 0.88200951,
      "num_input_tokens_seen": 139230750,
      "step": 6464,
      "time_per_iteration": 3.649953603744507
    },
    {
      "auxiliary_loss_clip": 0.01147522,
      "auxiliary_loss_mlp": 0.01027621,
      "balance_loss_clip": 1.04946911,
      "balance_loss_mlp": 1.01958609,
      "epoch": 0.777370287981723,
      "flos": 20119887319680.0,
      "grad_norm": 1.9284826989495587,
      "language_loss": 0.76095879,
      "learning_rate": 4.976265513489701e-07,
      "loss": 0.78271019,
      "num_input_tokens_seen": 139250720,
      "step": 6465,
      "time_per_iteration": 2.6883842945098877
    },
    {
      "auxiliary_loss_clip": 0.01130142,
      "auxiliary_loss_mlp": 0.01031128,
      "balance_loss_clip": 1.04991639,
      "balance_loss_mlp": 1.02373731,
      "epoch": 0.7774905308723622,
      "flos": 21718809331200.0,
      "grad_norm": 1.9635716286704006,
      "language_loss": 0.80333358,
      "learning_rate": 4.971124741004562e-07,
      "loss": 0.82494628,
      "num_input_tokens_seen": 139269720,
      "step": 6466,
      "time_per_iteration": 2.6813104152679443
    },
    {
      "auxiliary_loss_clip": 0.01135562,
      "auxiliary_loss_mlp": 0.0103241,
      "balance_loss_clip": 1.04693985,
      "balance_loss_mlp": 1.02428293,
      "epoch": 0.7776107737630013,
      "flos": 16034115093120.0,
      "grad_norm": 1.849121714421261,
      "language_loss": 0.76275849,
      "learning_rate": 4.965986248387846e-07,
      "loss": 0.78443825,
      "num_input_tokens_seen": 139288035,
      "step": 6467,
      "time_per_iteration": 2.750213146209717
    },
    {
      "auxiliary_loss_clip": 0.01096219,
      "auxiliary_loss_mlp": 0.01026311,
      "balance_loss_clip": 1.04202795,
      "balance_loss_mlp": 1.01891422,
      "epoch": 0.7777310166536403,
      "flos": 24790895838720.0,
      "grad_norm": 1.6584320573352627,
      "language_loss": 0.77514899,
      "learning_rate": 4.960850036419073e-07,
      "loss": 0.79637432,
      "num_input_tokens_seen": 139307135,
      "step": 6468,
      "time_per_iteration": 3.8397164344787598
    },
    {
      "auxiliary_loss_clip": 0.01163216,
      "auxiliary_loss_mlp": 0.00762124,
      "balance_loss_clip": 1.04897022,
      "balance_loss_mlp": 1.00047457,
      "epoch": 0.7778512595442795,
      "flos": 17272530253440.0,
      "grad_norm": 2.1196596017962084,
      "language_loss": 0.78343499,
      "learning_rate": 4.955716105877378e-07,
      "loss": 0.80268836,
      "num_input_tokens_seen": 139325905,
      "step": 6469,
      "time_per_iteration": 2.621429920196533
    },
    {
      "auxiliary_loss_clip": 0.01076284,
      "auxiliary_loss_mlp": 0.01027498,
      "balance_loss_clip": 1.03912187,
      "balance_loss_mlp": 1.01940322,
      "epoch": 0.7779715024349185,
      "flos": 17748418567680.0,
      "grad_norm": 1.878695316068758,
      "language_loss": 0.83144343,
      "learning_rate": 4.950584457541598e-07,
      "loss": 0.85248125,
      "num_input_tokens_seen": 139344370,
      "step": 6470,
      "time_per_iteration": 2.8296329975128174
    },
    {
      "auxiliary_loss_clip": 0.01145816,
      "auxiliary_loss_mlp": 0.01029829,
      "balance_loss_clip": 1.04800081,
      "balance_loss_mlp": 1.0223074,
      "epoch": 0.7780917453255576,
      "flos": 24316875031680.0,
      "grad_norm": 1.4385790042318727,
      "language_loss": 0.82018316,
      "learning_rate": 4.945455092190183e-07,
      "loss": 0.84193957,
      "num_input_tokens_seen": 139365625,
      "step": 6471,
      "time_per_iteration": 2.738776922225952
    },
    {
      "auxiliary_loss_clip": 0.01046152,
      "auxiliary_loss_mlp": 0.01000147,
      "balance_loss_clip": 1.01748443,
      "balance_loss_mlp": 0.99915141,
      "epoch": 0.7782119882161967,
      "flos": 56364601530240.0,
      "grad_norm": 0.6847560886011522,
      "language_loss": 0.55897373,
      "learning_rate": 4.940328010601271e-07,
      "loss": 0.57943672,
      "num_input_tokens_seen": 139430540,
      "step": 6472,
      "time_per_iteration": 3.269343614578247
    },
    {
      "auxiliary_loss_clip": 0.01153743,
      "auxiliary_loss_mlp": 0.01036615,
      "balance_loss_clip": 1.05446386,
      "balance_loss_mlp": 1.02836537,
      "epoch": 0.7783322311068358,
      "flos": 46789986994560.0,
      "grad_norm": 2.933397651853436,
      "language_loss": 0.7695024,
      "learning_rate": 4.935203213552621e-07,
      "loss": 0.79140592,
      "num_input_tokens_seen": 139454280,
      "step": 6473,
      "time_per_iteration": 2.871560573577881
    },
    {
      "auxiliary_loss_clip": 0.01164991,
      "auxiliary_loss_mlp": 0.01026073,
      "balance_loss_clip": 1.05016553,
      "balance_loss_mlp": 1.01737654,
      "epoch": 0.7784524739974749,
      "flos": 19057864872960.0,
      "grad_norm": 2.402767611474073,
      "language_loss": 0.67407089,
      "learning_rate": 4.930080701821662e-07,
      "loss": 0.69598156,
      "num_input_tokens_seen": 139471745,
      "step": 6474,
      "time_per_iteration": 2.7054102420806885
    },
    {
      "auxiliary_loss_clip": 0.01162471,
      "auxiliary_loss_mlp": 0.0076175,
      "balance_loss_clip": 1.04938769,
      "balance_loss_mlp": 1.00052571,
      "epoch": 0.778572716888114,
      "flos": 24791111320320.0,
      "grad_norm": 2.0421369186918175,
      "language_loss": 0.76948017,
      "learning_rate": 4.92496047618548e-07,
      "loss": 0.7887224,
      "num_input_tokens_seen": 139491505,
      "step": 6475,
      "time_per_iteration": 3.661471128463745
    },
    {
      "auxiliary_loss_clip": 0.01137294,
      "auxiliary_loss_mlp": 0.01028461,
      "balance_loss_clip": 1.05074418,
      "balance_loss_mlp": 1.02091801,
      "epoch": 0.7786929597787531,
      "flos": 20078086867200.0,
      "grad_norm": 2.069219236219637,
      "language_loss": 0.77943528,
      "learning_rate": 4.919842537420811e-07,
      "loss": 0.80109286,
      "num_input_tokens_seen": 139508620,
      "step": 6476,
      "time_per_iteration": 2.682210922241211
    },
    {
      "auxiliary_loss_clip": 0.01115408,
      "auxiliary_loss_mlp": 0.01026082,
      "balance_loss_clip": 1.0468266,
      "balance_loss_mlp": 1.01780891,
      "epoch": 0.7788132026693921,
      "flos": 21872220318720.0,
      "grad_norm": 1.5687931184695942,
      "language_loss": 0.79369909,
      "learning_rate": 4.91472688630404e-07,
      "loss": 0.81511402,
      "num_input_tokens_seen": 139529360,
      "step": 6477,
      "time_per_iteration": 2.7557482719421387
    },
    {
      "auxiliary_loss_clip": 0.01119895,
      "auxiliary_loss_mlp": 0.01022785,
      "balance_loss_clip": 1.04180741,
      "balance_loss_mlp": 1.01505995,
      "epoch": 0.7789334455600313,
      "flos": 11181937351680.0,
      "grad_norm": 1.7902989178309419,
      "language_loss": 0.74428928,
      "learning_rate": 4.909613523611202e-07,
      "loss": 0.76571608,
      "num_input_tokens_seen": 139546240,
      "step": 6478,
      "time_per_iteration": 2.720857858657837
    },
    {
      "auxiliary_loss_clip": 0.01121492,
      "auxiliary_loss_mlp": 0.01032283,
      "balance_loss_clip": 1.04463279,
      "balance_loss_mlp": 1.02399206,
      "epoch": 0.7790536884506704,
      "flos": 28695427015680.0,
      "grad_norm": 2.322943088018663,
      "language_loss": 0.7478317,
      "learning_rate": 4.904502450117991e-07,
      "loss": 0.76936948,
      "num_input_tokens_seen": 139567200,
      "step": 6479,
      "time_per_iteration": 2.8305773735046387
    },
    {
      "auxiliary_loss_clip": 0.01162985,
      "auxiliary_loss_mlp": 0.0102879,
      "balance_loss_clip": 1.05256748,
      "balance_loss_mlp": 1.02101755,
      "epoch": 0.7791739313413094,
      "flos": 11072302064640.0,
      "grad_norm": 2.511678029434563,
      "language_loss": 0.72678387,
      "learning_rate": 4.899393666599762e-07,
      "loss": 0.74870169,
      "num_input_tokens_seen": 139583775,
      "step": 6480,
      "time_per_iteration": 2.5878286361694336
    },
    {
      "auxiliary_loss_clip": 0.01157061,
      "auxiliary_loss_mlp": 0.01025612,
      "balance_loss_clip": 1.04985976,
      "balance_loss_mlp": 1.01792336,
      "epoch": 0.7792941742319486,
      "flos": 14679276975360.0,
      "grad_norm": 3.837090622557391,
      "language_loss": 0.7302922,
      "learning_rate": 4.894287173831506e-07,
      "loss": 0.752119,
      "num_input_tokens_seen": 139599735,
      "step": 6481,
      "time_per_iteration": 2.625014543533325
    },
    {
      "auxiliary_loss_clip": 0.01176174,
      "auxiliary_loss_mlp": 0.0102826,
      "balance_loss_clip": 1.05163836,
      "balance_loss_mlp": 1.02051759,
      "epoch": 0.7794144171225876,
      "flos": 23258874908160.0,
      "grad_norm": 2.3261633857767903,
      "language_loss": 0.84593141,
      "learning_rate": 4.889182972587877e-07,
      "loss": 0.86797577,
      "num_input_tokens_seen": 139619030,
      "step": 6482,
      "time_per_iteration": 2.6726245880126953
    },
    {
      "auxiliary_loss_clip": 0.01091796,
      "auxiliary_loss_mlp": 0.01024643,
      "balance_loss_clip": 1.04994845,
      "balance_loss_mlp": 1.01764584,
      "epoch": 0.7795346600132267,
      "flos": 21507080613120.0,
      "grad_norm": 1.8189155113642286,
      "language_loss": 0.66345155,
      "learning_rate": 4.884081063643177e-07,
      "loss": 0.68461597,
      "num_input_tokens_seen": 139637690,
      "step": 6483,
      "time_per_iteration": 3.0186638832092285
    },
    {
      "auxiliary_loss_clip": 0.00996038,
      "auxiliary_loss_mlp": 0.01000226,
      "balance_loss_clip": 1.00833035,
      "balance_loss_mlp": 0.99920648,
      "epoch": 0.7796549029038659,
      "flos": 70052273694720.0,
      "grad_norm": 0.8454396306909576,
      "language_loss": 0.52473748,
      "learning_rate": 4.878981447771353e-07,
      "loss": 0.54470009,
      "num_input_tokens_seen": 139692070,
      "step": 6484,
      "time_per_iteration": 3.575129270553589
    },
    {
      "auxiliary_loss_clip": 0.01139072,
      "auxiliary_loss_mlp": 0.01028067,
      "balance_loss_clip": 1.04475021,
      "balance_loss_mlp": 1.02028275,
      "epoch": 0.7797751457945049,
      "flos": 23989405714560.0,
      "grad_norm": 1.5417040631582697,
      "language_loss": 0.73313057,
      "learning_rate": 4.873884125746035e-07,
      "loss": 0.75480199,
      "num_input_tokens_seen": 139713745,
      "step": 6485,
      "time_per_iteration": 3.0048582553863525
    },
    {
      "auxiliary_loss_clip": 0.01161156,
      "auxiliary_loss_mlp": 0.01028601,
      "balance_loss_clip": 1.05223572,
      "balance_loss_mlp": 1.0209446,
      "epoch": 0.779895388685144,
      "flos": 22674751937280.0,
      "grad_norm": 13.69630129949891,
      "language_loss": 0.71991438,
      "learning_rate": 4.868789098340456e-07,
      "loss": 0.74181193,
      "num_input_tokens_seen": 139731650,
      "step": 6486,
      "time_per_iteration": 2.648087501525879
    },
    {
      "auxiliary_loss_clip": 0.01148449,
      "auxiliary_loss_mlp": 0.01025263,
      "balance_loss_clip": 1.04686451,
      "balance_loss_mlp": 1.01756513,
      "epoch": 0.7800156315757831,
      "flos": 23768698596480.0,
      "grad_norm": 2.18144171456008,
      "language_loss": 0.73322392,
      "learning_rate": 4.863696366327543e-07,
      "loss": 0.75496101,
      "num_input_tokens_seen": 139750820,
      "step": 6487,
      "time_per_iteration": 2.692596197128296
    },
    {
      "auxiliary_loss_clip": 0.01160829,
      "auxiliary_loss_mlp": 0.01025211,
      "balance_loss_clip": 1.05047178,
      "balance_loss_mlp": 1.01682496,
      "epoch": 0.7801358744664222,
      "flos": 26429714881920.0,
      "grad_norm": 2.311971873108395,
      "language_loss": 0.77825207,
      "learning_rate": 4.85860593047986e-07,
      "loss": 0.80011249,
      "num_input_tokens_seen": 139770885,
      "step": 6488,
      "time_per_iteration": 2.7041430473327637
    },
    {
      "auxiliary_loss_clip": 0.01131433,
      "auxiliary_loss_mlp": 0.01026616,
      "balance_loss_clip": 1.04596233,
      "balance_loss_mlp": 1.01926661,
      "epoch": 0.7802561173570612,
      "flos": 26322162583680.0,
      "grad_norm": 1.6994472736566264,
      "language_loss": 0.75182676,
      "learning_rate": 4.853517791569613e-07,
      "loss": 0.77340728,
      "num_input_tokens_seen": 139793065,
      "step": 6489,
      "time_per_iteration": 3.8674402236938477
    },
    {
      "auxiliary_loss_clip": 0.01122774,
      "auxiliary_loss_mlp": 0.01029362,
      "balance_loss_clip": 1.04429638,
      "balance_loss_mlp": 1.02098799,
      "epoch": 0.7803763602477004,
      "flos": 40333751596800.0,
      "grad_norm": 2.339244478906302,
      "language_loss": 0.66363394,
      "learning_rate": 4.848431950368684e-07,
      "loss": 0.68515527,
      "num_input_tokens_seen": 139815625,
      "step": 6490,
      "time_per_iteration": 3.9051809310913086
    },
    {
      "auxiliary_loss_clip": 0.0106294,
      "auxiliary_loss_mlp": 0.01000656,
      "balance_loss_clip": 1.00854921,
      "balance_loss_mlp": 0.99974424,
      "epoch": 0.7804966031383395,
      "flos": 67001448038400.0,
      "grad_norm": 0.710351622062481,
      "language_loss": 0.5566107,
      "learning_rate": 4.843348407648569e-07,
      "loss": 0.57724667,
      "num_input_tokens_seen": 139876905,
      "step": 6491,
      "time_per_iteration": 3.104855537414551
    },
    {
      "auxiliary_loss_clip": 0.01126284,
      "auxiliary_loss_mlp": 0.00762121,
      "balance_loss_clip": 1.04155159,
      "balance_loss_mlp": 1.00057769,
      "epoch": 0.7806168460289785,
      "flos": 17740733057280.0,
      "grad_norm": 4.030365292749472,
      "language_loss": 0.83447474,
      "learning_rate": 4.838267164180457e-07,
      "loss": 0.85335881,
      "num_input_tokens_seen": 139892575,
      "step": 6492,
      "time_per_iteration": 2.5790975093841553
    },
    {
      "auxiliary_loss_clip": 0.01163725,
      "auxiliary_loss_mlp": 0.0103052,
      "balance_loss_clip": 1.05218399,
      "balance_loss_mlp": 1.02216351,
      "epoch": 0.7807370889196176,
      "flos": 23946240545280.0,
      "grad_norm": 2.06239560721522,
      "language_loss": 0.83452737,
      "learning_rate": 4.833188220735156e-07,
      "loss": 0.85646987,
      "num_input_tokens_seen": 139912245,
      "step": 6493,
      "time_per_iteration": 2.503499746322632
    },
    {
      "auxiliary_loss_clip": 0.01118092,
      "auxiliary_loss_mlp": 0.01023734,
      "balance_loss_clip": 1.0460161,
      "balance_loss_mlp": 1.01604199,
      "epoch": 0.7808573318102567,
      "flos": 18989024457600.0,
      "grad_norm": 2.0559285959718947,
      "language_loss": 0.74938226,
      "learning_rate": 4.828111578083152e-07,
      "loss": 0.77080059,
      "num_input_tokens_seen": 139929150,
      "step": 6494,
      "time_per_iteration": 2.621903419494629
    },
    {
      "auxiliary_loss_clip": 0.01147367,
      "auxiliary_loss_mlp": 0.01029556,
      "balance_loss_clip": 1.04810333,
      "balance_loss_mlp": 1.0215807,
      "epoch": 0.7809775747008958,
      "flos": 23980750536960.0,
      "grad_norm": 1.968485027516594,
      "language_loss": 0.81386554,
      "learning_rate": 4.823037236994556e-07,
      "loss": 0.83563471,
      "num_input_tokens_seen": 139947315,
      "step": 6495,
      "time_per_iteration": 3.6113460063934326
    },
    {
      "auxiliary_loss_clip": 0.01026063,
      "auxiliary_loss_mlp": 0.00998345,
      "balance_loss_clip": 1.01912022,
      "balance_loss_mlp": 0.99736136,
      "epoch": 0.7810978175915348,
      "flos": 68535875180160.0,
      "grad_norm": 0.7152912923986376,
      "language_loss": 0.56282562,
      "learning_rate": 4.817965198239136e-07,
      "loss": 0.58306968,
      "num_input_tokens_seen": 140013775,
      "step": 6496,
      "time_per_iteration": 3.303980588912964
    },
    {
      "auxiliary_loss_clip": 0.01130647,
      "auxiliary_loss_mlp": 0.01027292,
      "balance_loss_clip": 1.04358685,
      "balance_loss_mlp": 1.01866722,
      "epoch": 0.781218060482174,
      "flos": 19642131498240.0,
      "grad_norm": 2.0268026169446403,
      "language_loss": 0.74855691,
      "learning_rate": 4.812895462586331e-07,
      "loss": 0.77013624,
      "num_input_tokens_seen": 140031600,
      "step": 6497,
      "time_per_iteration": 2.7527194023132324
    },
    {
      "auxiliary_loss_clip": 0.01118761,
      "auxiliary_loss_mlp": 0.01024469,
      "balance_loss_clip": 1.04640102,
      "balance_loss_mlp": 1.0167979,
      "epoch": 0.7813383033728131,
      "flos": 25627865621760.0,
      "grad_norm": 1.8127806662960375,
      "language_loss": 0.81732774,
      "learning_rate": 4.807828030805207e-07,
      "loss": 0.83876002,
      "num_input_tokens_seen": 140050590,
      "step": 6498,
      "time_per_iteration": 2.7412476539611816
    },
    {
      "auxiliary_loss_clip": 0.01085099,
      "auxiliary_loss_mlp": 0.01025075,
      "balance_loss_clip": 1.03791785,
      "balance_loss_mlp": 1.0173924,
      "epoch": 0.7814585462634521,
      "flos": 20485924865280.0,
      "grad_norm": 2.2356015621357614,
      "language_loss": 0.67804027,
      "learning_rate": 4.802762903664495e-07,
      "loss": 0.69914192,
      "num_input_tokens_seen": 140069770,
      "step": 6499,
      "time_per_iteration": 2.8853518962860107
    },
    {
      "auxiliary_loss_clip": 0.01165443,
      "auxiliary_loss_mlp": 0.01026202,
      "balance_loss_clip": 1.05130816,
      "balance_loss_mlp": 1.01777411,
      "epoch": 0.7815787891540913,
      "flos": 22304297018880.0,
      "grad_norm": 2.421219103356947,
      "language_loss": 0.74481583,
      "learning_rate": 4.797700081932565e-07,
      "loss": 0.76673228,
      "num_input_tokens_seen": 140087635,
      "step": 6500,
      "time_per_iteration": 2.5643625259399414
    },
    {
      "auxiliary_loss_clip": 0.0112849,
      "auxiliary_loss_mlp": 0.01030393,
      "balance_loss_clip": 1.04737818,
      "balance_loss_mlp": 1.0226115,
      "epoch": 0.7816990320447303,
      "flos": 22600668136320.0,
      "grad_norm": 2.2151669019778404,
      "language_loss": 0.81939328,
      "learning_rate": 4.792639566377442e-07,
      "loss": 0.84098214,
      "num_input_tokens_seen": 140105045,
      "step": 6501,
      "time_per_iteration": 3.952535390853882
    },
    {
      "auxiliary_loss_clip": 0.0116024,
      "auxiliary_loss_mlp": 0.0102745,
      "balance_loss_clip": 1.04813337,
      "balance_loss_mlp": 1.01989818,
      "epoch": 0.7818192749353694,
      "flos": 24935974871040.0,
      "grad_norm": 3.0253553017194434,
      "language_loss": 0.77529323,
      "learning_rate": 4.78758135776681e-07,
      "loss": 0.79717016,
      "num_input_tokens_seen": 140124900,
      "step": 6502,
      "time_per_iteration": 2.720484972000122
    },
    {
      "auxiliary_loss_clip": 0.01164448,
      "auxiliary_loss_mlp": 0.01030997,
      "balance_loss_clip": 1.05309939,
      "balance_loss_mlp": 1.02254534,
      "epoch": 0.7819395178260086,
      "flos": 23733039369600.0,
      "grad_norm": 1.9848490133738261,
      "language_loss": 0.78964794,
      "learning_rate": 4.782525456867989e-07,
      "loss": 0.81160247,
      "num_input_tokens_seen": 140143755,
      "step": 6503,
      "time_per_iteration": 2.7089622020721436
    },
    {
      "auxiliary_loss_clip": 0.0117808,
      "auxiliary_loss_mlp": 0.01031738,
      "balance_loss_clip": 1.05128944,
      "balance_loss_mlp": 1.02311957,
      "epoch": 0.7820597607166476,
      "flos": 23221671396480.0,
      "grad_norm": 2.2541658925907795,
      "language_loss": 0.83572513,
      "learning_rate": 4.777471864447959e-07,
      "loss": 0.85782331,
      "num_input_tokens_seen": 140164495,
      "step": 6504,
      "time_per_iteration": 2.647115707397461
    },
    {
      "auxiliary_loss_clip": 0.01127275,
      "auxiliary_loss_mlp": 0.01029303,
      "balance_loss_clip": 1.04506314,
      "balance_loss_mlp": 1.02132177,
      "epoch": 0.7821800036072867,
      "flos": 22309540404480.0,
      "grad_norm": 2.3255282522263605,
      "language_loss": 0.80863917,
      "learning_rate": 4.772420581273344e-07,
      "loss": 0.83020496,
      "num_input_tokens_seen": 140181980,
      "step": 6505,
      "time_per_iteration": 2.6857898235321045
    },
    {
      "auxiliary_loss_clip": 0.01084546,
      "auxiliary_loss_mlp": 0.01025344,
      "balance_loss_clip": 1.03932571,
      "balance_loss_mlp": 1.01745784,
      "epoch": 0.7823002464979258,
      "flos": 21544176384000.0,
      "grad_norm": 2.2463191094921875,
      "language_loss": 0.76520836,
      "learning_rate": 4.7673716081104134e-07,
      "loss": 0.78630722,
      "num_input_tokens_seen": 140202155,
      "step": 6506,
      "time_per_iteration": 2.8513174057006836
    },
    {
      "auxiliary_loss_clip": 0.01127619,
      "auxiliary_loss_mlp": 0.01032149,
      "balance_loss_clip": 1.04409862,
      "balance_loss_mlp": 1.02421582,
      "epoch": 0.7824204893885649,
      "flos": 24535642815360.0,
      "grad_norm": 2.0692399516990903,
      "language_loss": 0.84656727,
      "learning_rate": 4.762324945725109e-07,
      "loss": 0.86816496,
      "num_input_tokens_seen": 140221600,
      "step": 6507,
      "time_per_iteration": 2.7347333431243896
    },
    {
      "auxiliary_loss_clip": 0.01120293,
      "auxiliary_loss_mlp": 0.01025122,
      "balance_loss_clip": 1.04718304,
      "balance_loss_mlp": 1.01783204,
      "epoch": 0.782540732279204,
      "flos": 27415211402880.0,
      "grad_norm": 1.648392371951953,
      "language_loss": 0.7593596,
      "learning_rate": 4.7572805948829844e-07,
      "loss": 0.78081375,
      "num_input_tokens_seen": 140241860,
      "step": 6508,
      "time_per_iteration": 2.803790330886841
    },
    {
      "auxiliary_loss_clip": 0.0108047,
      "auxiliary_loss_mlp": 0.01030237,
      "balance_loss_clip": 1.03898191,
      "balance_loss_mlp": 1.02251208,
      "epoch": 0.7826609751698431,
      "flos": 24353216616960.0,
      "grad_norm": 2.0381406355550338,
      "language_loss": 0.71182084,
      "learning_rate": 4.7522385563492795e-07,
      "loss": 0.73292792,
      "num_input_tokens_seen": 140262160,
      "step": 6509,
      "time_per_iteration": 2.8685429096221924
    },
    {
      "auxiliary_loss_clip": 0.01119429,
      "auxiliary_loss_mlp": 0.01030418,
      "balance_loss_clip": 1.04270244,
      "balance_loss_mlp": 1.02175725,
      "epoch": 0.7827812180604822,
      "flos": 23988543788160.0,
      "grad_norm": 2.02179445681682,
      "language_loss": 0.70620549,
      "learning_rate": 4.747198830888863e-07,
      "loss": 0.72770393,
      "num_input_tokens_seen": 140282030,
      "step": 6510,
      "time_per_iteration": 2.87970232963562
    },
    {
      "auxiliary_loss_clip": 0.01156016,
      "auxiliary_loss_mlp": 0.01024819,
      "balance_loss_clip": 1.0514878,
      "balance_loss_mlp": 1.01668274,
      "epoch": 0.7829014609511212,
      "flos": 27454318335360.0,
      "grad_norm": 3.057656699998539,
      "language_loss": 0.68623018,
      "learning_rate": 4.742161419266251e-07,
      "loss": 0.70803857,
      "num_input_tokens_seen": 140301190,
      "step": 6511,
      "time_per_iteration": 2.7717907428741455
    },
    {
      "auxiliary_loss_clip": 0.01156529,
      "auxiliary_loss_mlp": 0.01026635,
      "balance_loss_clip": 1.05136395,
      "balance_loss_mlp": 1.01868963,
      "epoch": 0.7830217038417604,
      "flos": 29204532432000.0,
      "grad_norm": 4.660640777574948,
      "language_loss": 0.64724243,
      "learning_rate": 4.7371263222456304e-07,
      "loss": 0.66907406,
      "num_input_tokens_seen": 140318510,
      "step": 6512,
      "time_per_iteration": 2.7205052375793457
    },
    {
      "auxiliary_loss_clip": 0.01032165,
      "auxiliary_loss_mlp": 0.0100515,
      "balance_loss_clip": 1.0094595,
      "balance_loss_mlp": 1.00411248,
      "epoch": 0.7831419467323995,
      "flos": 60950895822720.0,
      "grad_norm": 0.8020945940365155,
      "language_loss": 0.61201835,
      "learning_rate": 4.7320935405908004e-07,
      "loss": 0.63239145,
      "num_input_tokens_seen": 140379380,
      "step": 6513,
      "time_per_iteration": 3.2608823776245117
    },
    {
      "auxiliary_loss_clip": 0.01149304,
      "auxiliary_loss_mlp": 0.01029577,
      "balance_loss_clip": 1.0485239,
      "balance_loss_mlp": 1.02007043,
      "epoch": 0.7832621896230385,
      "flos": 19682531320320.0,
      "grad_norm": 2.5101812781894037,
      "language_loss": 0.84252858,
      "learning_rate": 4.7270630750652475e-07,
      "loss": 0.86431742,
      "num_input_tokens_seen": 140395335,
      "step": 6514,
      "time_per_iteration": 2.6230897903442383
    },
    {
      "auxiliary_loss_clip": 0.01143423,
      "auxiliary_loss_mlp": 0.01026802,
      "balance_loss_clip": 1.04404747,
      "balance_loss_mlp": 1.01905966,
      "epoch": 0.7833824325136777,
      "flos": 25009232659200.0,
      "grad_norm": 2.1395968418012954,
      "language_loss": 0.80420399,
      "learning_rate": 4.7220349264320746e-07,
      "loss": 0.82590622,
      "num_input_tokens_seen": 140414420,
      "step": 6515,
      "time_per_iteration": 3.685209035873413
    },
    {
      "auxiliary_loss_clip": 0.01041279,
      "auxiliary_loss_mlp": 0.01006317,
      "balance_loss_clip": 1.00866449,
      "balance_loss_mlp": 1.00536919,
      "epoch": 0.7835026754043167,
      "flos": 68800142517120.0,
      "grad_norm": 0.7380086377863192,
      "language_loss": 0.54831421,
      "learning_rate": 4.71700909545407e-07,
      "loss": 0.5687902,
      "num_input_tokens_seen": 140477365,
      "step": 6516,
      "time_per_iteration": 4.2208638191223145
    },
    {
      "auxiliary_loss_clip": 0.01105217,
      "auxiliary_loss_mlp": 0.01026955,
      "balance_loss_clip": 1.04724216,
      "balance_loss_mlp": 1.01886678,
      "epoch": 0.7836229182949558,
      "flos": 19864598382720.0,
      "grad_norm": 2.2340207676390964,
      "language_loss": 0.7685737,
      "learning_rate": 4.711985582893627e-07,
      "loss": 0.78989542,
      "num_input_tokens_seen": 140495885,
      "step": 6517,
      "time_per_iteration": 2.7888240814208984
    },
    {
      "auxiliary_loss_clip": 0.01148365,
      "auxiliary_loss_mlp": 0.01024893,
      "balance_loss_clip": 1.05057299,
      "balance_loss_mlp": 1.01743627,
      "epoch": 0.783743161185595,
      "flos": 22965843755520.0,
      "grad_norm": 1.7765440662051615,
      "language_loss": 0.71994334,
      "learning_rate": 4.706964389512811e-07,
      "loss": 0.74167585,
      "num_input_tokens_seen": 140515920,
      "step": 6518,
      "time_per_iteration": 2.7605602741241455
    },
    {
      "auxiliary_loss_clip": 0.01130613,
      "auxiliary_loss_mlp": 0.01029517,
      "balance_loss_clip": 1.0467329,
      "balance_loss_mlp": 1.02159548,
      "epoch": 0.783863404076234,
      "flos": 12458489777280.0,
      "grad_norm": 1.928564229418628,
      "language_loss": 0.87991035,
      "learning_rate": 4.701945516073345e-07,
      "loss": 0.90151167,
      "num_input_tokens_seen": 140533395,
      "step": 6519,
      "time_per_iteration": 2.728820562362671
    },
    {
      "auxiliary_loss_clip": 0.01145206,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 1.04526663,
      "balance_loss_mlp": 1.01911545,
      "epoch": 0.7839836469668731,
      "flos": 24243940465920.0,
      "grad_norm": 2.522688426339622,
      "language_loss": 0.75293744,
      "learning_rate": 4.696928963336577e-07,
      "loss": 0.77465588,
      "num_input_tokens_seen": 140552825,
      "step": 6520,
      "time_per_iteration": 3.872843027114868
    },
    {
      "auxiliary_loss_clip": 0.01036614,
      "auxiliary_loss_mlp": 0.0100388,
      "balance_loss_clip": 1.00921404,
      "balance_loss_mlp": 1.00292003,
      "epoch": 0.7841038898575122,
      "flos": 62121978938880.0,
      "grad_norm": 0.852716453030842,
      "language_loss": 0.60931385,
      "learning_rate": 4.6919147320635224e-07,
      "loss": 0.62971878,
      "num_input_tokens_seen": 140615535,
      "step": 6521,
      "time_per_iteration": 3.3470804691314697
    },
    {
      "auxiliary_loss_clip": 0.01097447,
      "auxiliary_loss_mlp": 0.01033443,
      "balance_loss_clip": 1.04269385,
      "balance_loss_mlp": 1.02559888,
      "epoch": 0.7842241327481513,
      "flos": 20193899293440.0,
      "grad_norm": 2.3576499537958555,
      "language_loss": 0.73454523,
      "learning_rate": 4.6869028230148286e-07,
      "loss": 0.75585413,
      "num_input_tokens_seen": 140633330,
      "step": 6522,
      "time_per_iteration": 2.8599464893341064
    },
    {
      "auxiliary_loss_clip": 0.01147958,
      "auxiliary_loss_mlp": 0.01028103,
      "balance_loss_clip": 1.04730499,
      "balance_loss_mlp": 1.02013659,
      "epoch": 0.7843443756387903,
      "flos": 28074531496320.0,
      "grad_norm": 3.0824755090933724,
      "language_loss": 0.60092479,
      "learning_rate": 4.6818932369507957e-07,
      "loss": 0.62268543,
      "num_input_tokens_seen": 140652830,
      "step": 6523,
      "time_per_iteration": 2.7814037799835205
    },
    {
      "auxiliary_loss_clip": 0.01118679,
      "auxiliary_loss_mlp": 0.01028449,
      "balance_loss_clip": 1.04729784,
      "balance_loss_mlp": 1.02099252,
      "epoch": 0.7844646185294295,
      "flos": 21323397438720.0,
      "grad_norm": 2.0335725461317375,
      "language_loss": 0.8920331,
      "learning_rate": 4.676885974631386e-07,
      "loss": 0.91350436,
      "num_input_tokens_seen": 140671190,
      "step": 6524,
      "time_per_iteration": 2.813126802444458
    },
    {
      "auxiliary_loss_clip": 0.01088425,
      "auxiliary_loss_mlp": 0.01030579,
      "balance_loss_clip": 1.04030776,
      "balance_loss_mlp": 1.02253866,
      "epoch": 0.7845848614200686,
      "flos": 23656585271040.0,
      "grad_norm": 1.9806742158950257,
      "language_loss": 0.81198752,
      "learning_rate": 4.67188103681619e-07,
      "loss": 0.83317751,
      "num_input_tokens_seen": 140690975,
      "step": 6525,
      "time_per_iteration": 2.8946430683135986
    },
    {
      "auxiliary_loss_clip": 0.01100392,
      "auxiliary_loss_mlp": 0.01028013,
      "balance_loss_clip": 1.04014266,
      "balance_loss_mlp": 1.02003813,
      "epoch": 0.7847051043107076,
      "flos": 23402194174080.0,
      "grad_norm": 2.4307158346664774,
      "language_loss": 0.69162309,
      "learning_rate": 4.666878424264453e-07,
      "loss": 0.71290714,
      "num_input_tokens_seen": 140710930,
      "step": 6526,
      "time_per_iteration": 3.8096983432769775
    },
    {
      "auxiliary_loss_clip": 0.01139926,
      "auxiliary_loss_mlp": 0.01025192,
      "balance_loss_clip": 1.04803944,
      "balance_loss_mlp": 1.01825953,
      "epoch": 0.7848253472013467,
      "flos": 19022277473280.0,
      "grad_norm": 1.6601467609574772,
      "language_loss": 0.74096388,
      "learning_rate": 4.661878137735069e-07,
      "loss": 0.76261508,
      "num_input_tokens_seen": 140729120,
      "step": 6527,
      "time_per_iteration": 2.7095086574554443
    },
    {
      "auxiliary_loss_clip": 0.01139051,
      "auxiliary_loss_mlp": 0.01029146,
      "balance_loss_clip": 1.04539871,
      "balance_loss_mlp": 1.02179098,
      "epoch": 0.7849455900919858,
      "flos": 21179180332800.0,
      "grad_norm": 2.552814115053183,
      "language_loss": 0.75340307,
      "learning_rate": 4.656880177986571e-07,
      "loss": 0.77508509,
      "num_input_tokens_seen": 140747665,
      "step": 6528,
      "time_per_iteration": 2.6937594413757324
    },
    {
      "auxiliary_loss_clip": 0.01131296,
      "auxiliary_loss_mlp": 0.01028506,
      "balance_loss_clip": 1.04600859,
      "balance_loss_mlp": 1.01927948,
      "epoch": 0.7850658329826249,
      "flos": 19536482620800.0,
      "grad_norm": 2.041512317682308,
      "language_loss": 0.81583023,
      "learning_rate": 4.6518845457771607e-07,
      "loss": 0.83742833,
      "num_input_tokens_seen": 140766525,
      "step": 6529,
      "time_per_iteration": 2.763212203979492
    },
    {
      "auxiliary_loss_clip": 0.01117362,
      "auxiliary_loss_mlp": 0.01031673,
      "balance_loss_clip": 1.04674339,
      "balance_loss_mlp": 1.02367997,
      "epoch": 0.7851860758732639,
      "flos": 12495334152960.0,
      "grad_norm": 4.726836820932066,
      "language_loss": 0.79551697,
      "learning_rate": 4.646891241864652e-07,
      "loss": 0.8170073,
      "num_input_tokens_seen": 140785090,
      "step": 6530,
      "time_per_iteration": 2.723371982574463
    },
    {
      "auxiliary_loss_clip": 0.01162323,
      "auxiliary_loss_mlp": 0.01025271,
      "balance_loss_clip": 1.05081463,
      "balance_loss_mlp": 1.01707578,
      "epoch": 0.7853063187639031,
      "flos": 22960959505920.0,
      "grad_norm": 1.9250902850915848,
      "language_loss": 0.73391962,
      "learning_rate": 4.6419002670065397e-07,
      "loss": 0.75579554,
      "num_input_tokens_seen": 140804670,
      "step": 6531,
      "time_per_iteration": 2.742928981781006
    },
    {
      "auxiliary_loss_clip": 0.01141686,
      "auxiliary_loss_mlp": 0.01033505,
      "balance_loss_clip": 1.04675734,
      "balance_loss_mlp": 1.02579844,
      "epoch": 0.7854265616545422,
      "flos": 17347260499200.0,
      "grad_norm": 2.503273229104705,
      "language_loss": 0.86183017,
      "learning_rate": 4.6369116219599445e-07,
      "loss": 0.88358212,
      "num_input_tokens_seen": 140820655,
      "step": 6532,
      "time_per_iteration": 2.629345178604126
    },
    {
      "auxiliary_loss_clip": 0.01172066,
      "auxiliary_loss_mlp": 0.00761579,
      "balance_loss_clip": 1.04947972,
      "balance_loss_mlp": 1.00054741,
      "epoch": 0.7855468045451812,
      "flos": 23838293197440.0,
      "grad_norm": 1.6141806463622723,
      "language_loss": 0.7931155,
      "learning_rate": 4.631925307481637e-07,
      "loss": 0.81245196,
      "num_input_tokens_seen": 140840470,
      "step": 6533,
      "time_per_iteration": 2.710815191268921
    },
    {
      "auxiliary_loss_clip": 0.01111146,
      "auxiliary_loss_mlp": 0.01024529,
      "balance_loss_clip": 1.04391408,
      "balance_loss_mlp": 1.0166676,
      "epoch": 0.7856670474358204,
      "flos": 25666792986240.0,
      "grad_norm": 2.0908297842051597,
      "language_loss": 0.75495625,
      "learning_rate": 4.6269413243280533e-07,
      "loss": 0.77631307,
      "num_input_tokens_seen": 140859890,
      "step": 6534,
      "time_per_iteration": 2.7278568744659424
    },
    {
      "auxiliary_loss_clip": 0.01131413,
      "auxiliary_loss_mlp": 0.01033731,
      "balance_loss_clip": 1.04670954,
      "balance_loss_mlp": 1.02462935,
      "epoch": 0.7857872903264594,
      "flos": 18144656472960.0,
      "grad_norm": 2.6135255518667524,
      "language_loss": 0.74292088,
      "learning_rate": 4.621959673255236e-07,
      "loss": 0.76457232,
      "num_input_tokens_seen": 140876190,
      "step": 6535,
      "time_per_iteration": 2.8155508041381836
    },
    {
      "auxiliary_loss_clip": 0.01147664,
      "auxiliary_loss_mlp": 0.01030527,
      "balance_loss_clip": 1.04834926,
      "balance_loss_mlp": 1.02308202,
      "epoch": 0.7859075332170985,
      "flos": 14386138081920.0,
      "grad_norm": 2.0654382884945552,
      "language_loss": 0.9025017,
      "learning_rate": 4.6169803550189135e-07,
      "loss": 0.92428356,
      "num_input_tokens_seen": 140891885,
      "step": 6536,
      "time_per_iteration": 2.61706805229187
    },
    {
      "auxiliary_loss_clip": 0.01130229,
      "auxiliary_loss_mlp": 0.01029405,
      "balance_loss_clip": 1.04688025,
      "balance_loss_mlp": 1.02107215,
      "epoch": 0.7860277761077377,
      "flos": 19864059678720.0,
      "grad_norm": 1.9763823196296075,
      "language_loss": 0.77659178,
      "learning_rate": 4.6120033703744355e-07,
      "loss": 0.79818809,
      "num_input_tokens_seen": 140910780,
      "step": 6537,
      "time_per_iteration": 2.76107120513916
    },
    {
      "auxiliary_loss_clip": 0.01121246,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.04423451,
      "balance_loss_mlp": 1.02194524,
      "epoch": 0.7861480189983767,
      "flos": 26396174557440.0,
      "grad_norm": 1.9177548226478676,
      "language_loss": 0.78234029,
      "learning_rate": 4.607028720076822e-07,
      "loss": 0.80384612,
      "num_input_tokens_seen": 140927460,
      "step": 6538,
      "time_per_iteration": 2.8056533336639404
    },
    {
      "auxiliary_loss_clip": 0.01153712,
      "auxiliary_loss_mlp": 0.01022189,
      "balance_loss_clip": 1.0483191,
      "balance_loss_mlp": 1.01403522,
      "epoch": 0.7862682618890158,
      "flos": 24236578177920.0,
      "grad_norm": 3.236616880799861,
      "language_loss": 0.73549318,
      "learning_rate": 4.6020564048807074e-07,
      "loss": 0.75725222,
      "num_input_tokens_seen": 140945135,
      "step": 6539,
      "time_per_iteration": 2.6848702430725098
    },
    {
      "auxiliary_loss_clip": 0.01145795,
      "auxiliary_loss_mlp": 0.01029736,
      "balance_loss_clip": 1.0514071,
      "balance_loss_mlp": 1.02158177,
      "epoch": 0.7863885047796549,
      "flos": 47551508259840.0,
      "grad_norm": 2.2034214581122167,
      "language_loss": 0.71986252,
      "learning_rate": 4.5970864255403883e-07,
      "loss": 0.74161786,
      "num_input_tokens_seen": 140966660,
      "step": 6540,
      "time_per_iteration": 2.9956986904144287
    },
    {
      "auxiliary_loss_clip": 0.0112385,
      "auxiliary_loss_mlp": 0.01024884,
      "balance_loss_clip": 1.04591131,
      "balance_loss_mlp": 1.0176686,
      "epoch": 0.786508747670294,
      "flos": 24389234979840.0,
      "grad_norm": 1.8175910901486494,
      "language_loss": 0.82435107,
      "learning_rate": 4.59211878280982e-07,
      "loss": 0.84583843,
      "num_input_tokens_seen": 140986175,
      "step": 6541,
      "time_per_iteration": 3.7052857875823975
    },
    {
      "auxiliary_loss_clip": 0.01115368,
      "auxiliary_loss_mlp": 0.01027485,
      "balance_loss_clip": 1.04564488,
      "balance_loss_mlp": 1.01889563,
      "epoch": 0.786628990560933,
      "flos": 18041234238720.0,
      "grad_norm": 2.3026303482335475,
      "language_loss": 0.70072538,
      "learning_rate": 4.587153477442578e-07,
      "loss": 0.7221539,
      "num_input_tokens_seen": 141002490,
      "step": 6542,
      "time_per_iteration": 3.8601627349853516
    },
    {
      "auxiliary_loss_clip": 0.01163998,
      "auxiliary_loss_mlp": 0.01032434,
      "balance_loss_clip": 1.05137408,
      "balance_loss_mlp": 1.02343345,
      "epoch": 0.7867492334515722,
      "flos": 25848860048640.0,
      "grad_norm": 3.839896163988517,
      "language_loss": 0.81545246,
      "learning_rate": 4.582190510191899e-07,
      "loss": 0.83741677,
      "num_input_tokens_seen": 141021150,
      "step": 6543,
      "time_per_iteration": 2.7224395275115967
    },
    {
      "auxiliary_loss_clip": 0.01145142,
      "auxiliary_loss_mlp": 0.01031783,
      "balance_loss_clip": 1.04707885,
      "balance_loss_mlp": 1.02471948,
      "epoch": 0.7868694763422113,
      "flos": 16580819070720.0,
      "grad_norm": 2.1724038661775515,
      "language_loss": 0.87252522,
      "learning_rate": 4.5772298818106625e-07,
      "loss": 0.89429444,
      "num_input_tokens_seen": 141036940,
      "step": 6544,
      "time_per_iteration": 2.6589772701263428
    },
    {
      "auxiliary_loss_clip": 0.01179296,
      "auxiliary_loss_mlp": 0.01041369,
      "balance_loss_clip": 1.05456078,
      "balance_loss_mlp": 1.03301859,
      "epoch": 0.7869897192328503,
      "flos": 29386276272000.0,
      "grad_norm": 2.401295999175684,
      "language_loss": 0.71791852,
      "learning_rate": 4.572271593051384e-07,
      "loss": 0.74012518,
      "num_input_tokens_seen": 141054295,
      "step": 6545,
      "time_per_iteration": 2.758341073989868
    },
    {
      "auxiliary_loss_clip": 0.01121499,
      "auxiliary_loss_mlp": 0.01030426,
      "balance_loss_clip": 1.04533517,
      "balance_loss_mlp": 1.02227855,
      "epoch": 0.7871099621234895,
      "flos": 17128923678720.0,
      "grad_norm": 1.7617888577396446,
      "language_loss": 0.78226197,
      "learning_rate": 4.567315644666245e-07,
      "loss": 0.80378115,
      "num_input_tokens_seen": 141073090,
      "step": 6546,
      "time_per_iteration": 3.794665575027466
    },
    {
      "auxiliary_loss_clip": 0.01144387,
      "auxiliary_loss_mlp": 0.01031402,
      "balance_loss_clip": 1.05143535,
      "balance_loss_mlp": 1.02444673,
      "epoch": 0.7872302050141285,
      "flos": 23440187784960.0,
      "grad_norm": 2.3128878706151252,
      "language_loss": 0.84762001,
      "learning_rate": 4.5623620374070507e-07,
      "loss": 0.86937791,
      "num_input_tokens_seen": 141092405,
      "step": 6547,
      "time_per_iteration": 2.698028326034546
    },
    {
      "auxiliary_loss_clip": 0.01054525,
      "auxiliary_loss_mlp": 0.01000931,
      "balance_loss_clip": 1.00861847,
      "balance_loss_mlp": 0.99997711,
      "epoch": 0.7873504479047676,
      "flos": 65959752689280.0,
      "grad_norm": 0.7648238347801257,
      "language_loss": 0.58296567,
      "learning_rate": 4.557410772025263e-07,
      "loss": 0.60352021,
      "num_input_tokens_seen": 141154355,
      "step": 6548,
      "time_per_iteration": 3.352733850479126
    },
    {
      "auxiliary_loss_clip": 0.01121849,
      "auxiliary_loss_mlp": 0.01032544,
      "balance_loss_clip": 1.04689097,
      "balance_loss_mlp": 1.02493203,
      "epoch": 0.7874706907954068,
      "flos": 23258336204160.0,
      "grad_norm": 4.905530601757386,
      "language_loss": 0.66455632,
      "learning_rate": 4.5524618492719803e-07,
      "loss": 0.68610024,
      "num_input_tokens_seen": 141173575,
      "step": 6549,
      "time_per_iteration": 2.7314810752868652
    },
    {
      "auxiliary_loss_clip": 0.01123093,
      "auxiliary_loss_mlp": 0.0076188,
      "balance_loss_clip": 1.04685616,
      "balance_loss_mlp": 1.00046945,
      "epoch": 0.7875909336860458,
      "flos": 28767786963840.0,
      "grad_norm": 1.6482921866802722,
      "language_loss": 0.79413009,
      "learning_rate": 4.54751526989795e-07,
      "loss": 0.81297982,
      "num_input_tokens_seen": 141195415,
      "step": 6550,
      "time_per_iteration": 2.8311917781829834
    },
    {
      "auxiliary_loss_clip": 0.01130003,
      "auxiliary_loss_mlp": 0.01026856,
      "balance_loss_clip": 1.04708338,
      "balance_loss_mlp": 1.01947427,
      "epoch": 0.7877111765766849,
      "flos": 18697286194560.0,
      "grad_norm": 2.1155232834105586,
      "language_loss": 0.7896558,
      "learning_rate": 4.5425710346535775e-07,
      "loss": 0.8112244,
      "num_input_tokens_seen": 141213360,
      "step": 6551,
      "time_per_iteration": 2.669660806655884
    },
    {
      "auxiliary_loss_clip": 0.01133369,
      "auxiliary_loss_mlp": 0.01027095,
      "balance_loss_clip": 1.04735708,
      "balance_loss_mlp": 1.01922727,
      "epoch": 0.787831419467324,
      "flos": 27592968833280.0,
      "grad_norm": 2.0971103035333107,
      "language_loss": 0.8178947,
      "learning_rate": 4.537629144288877e-07,
      "loss": 0.83949935,
      "num_input_tokens_seen": 141230815,
      "step": 6552,
      "time_per_iteration": 3.7787435054779053
    },
    {
      "auxiliary_loss_clip": 0.01141426,
      "auxiliary_loss_mlp": 0.01028705,
      "balance_loss_clip": 1.04637253,
      "balance_loss_mlp": 1.02092695,
      "epoch": 0.7879516623579631,
      "flos": 18150187167360.0,
      "grad_norm": 2.846517037707431,
      "language_loss": 0.7524941,
      "learning_rate": 4.5326895995535477e-07,
      "loss": 0.77419543,
      "num_input_tokens_seen": 141249715,
      "step": 6553,
      "time_per_iteration": 2.7351739406585693
    },
    {
      "auxiliary_loss_clip": 0.01102693,
      "auxiliary_loss_mlp": 0.01025626,
      "balance_loss_clip": 1.04412806,
      "balance_loss_mlp": 1.01736462,
      "epoch": 0.7880719052486022,
      "flos": 20339193807360.0,
      "grad_norm": 2.3990180427054377,
      "language_loss": 0.84879923,
      "learning_rate": 4.527752401196907e-07,
      "loss": 0.87008244,
      "num_input_tokens_seen": 141267730,
      "step": 6554,
      "time_per_iteration": 2.774066209793091
    },
    {
      "auxiliary_loss_clip": 0.01154338,
      "auxiliary_loss_mlp": 0.01031343,
      "balance_loss_clip": 1.04935861,
      "balance_loss_mlp": 1.02328444,
      "epoch": 0.7881921481392413,
      "flos": 21653237053440.0,
      "grad_norm": 1.7329584564681768,
      "language_loss": 0.66608357,
      "learning_rate": 4.5228175499679254e-07,
      "loss": 0.68794036,
      "num_input_tokens_seen": 141287315,
      "step": 6555,
      "time_per_iteration": 2.6973791122436523
    },
    {
      "auxiliary_loss_clip": 0.01054765,
      "auxiliary_loss_mlp": 0.00999698,
      "balance_loss_clip": 1.01771808,
      "balance_loss_mlp": 0.99877977,
      "epoch": 0.7883123910298804,
      "flos": 68565860058240.0,
      "grad_norm": 0.8339049365844966,
      "language_loss": 0.54502678,
      "learning_rate": 4.5178850466152174e-07,
      "loss": 0.56557131,
      "num_input_tokens_seen": 141346145,
      "step": 6556,
      "time_per_iteration": 3.247497320175171
    },
    {
      "auxiliary_loss_clip": 0.0109801,
      "auxiliary_loss_mlp": 0.01024223,
      "balance_loss_clip": 1.03709066,
      "balance_loss_mlp": 1.01646256,
      "epoch": 0.7884326339205194,
      "flos": 19318217627520.0,
      "grad_norm": 1.7769839389281057,
      "language_loss": 0.81754148,
      "learning_rate": 4.512954891887031e-07,
      "loss": 0.83876383,
      "num_input_tokens_seen": 141364445,
      "step": 6557,
      "time_per_iteration": 2.7556378841400146
    },
    {
      "auxiliary_loss_clip": 0.01162525,
      "auxiliary_loss_mlp": 0.01030376,
      "balance_loss_clip": 1.04885197,
      "balance_loss_mlp": 1.02231193,
      "epoch": 0.7885528768111585,
      "flos": 17784903807360.0,
      "grad_norm": 2.741031822169146,
      "language_loss": 0.83763909,
      "learning_rate": 4.5080270865312806e-07,
      "loss": 0.85956812,
      "num_input_tokens_seen": 141381640,
      "step": 6558,
      "time_per_iteration": 2.638136863708496
    },
    {
      "auxiliary_loss_clip": 0.01115811,
      "auxiliary_loss_mlp": 0.01023895,
      "balance_loss_clip": 1.04313743,
      "balance_loss_mlp": 1.01635492,
      "epoch": 0.7886731197017977,
      "flos": 18807639753600.0,
      "grad_norm": 2.07265449005437,
      "language_loss": 0.71025383,
      "learning_rate": 4.5031016312954985e-07,
      "loss": 0.73165083,
      "num_input_tokens_seen": 141399955,
      "step": 6559,
      "time_per_iteration": 2.7365334033966064
    },
    {
      "auxiliary_loss_clip": 0.01156096,
      "auxiliary_loss_mlp": 0.01030069,
      "balance_loss_clip": 1.05320239,
      "balance_loss_mlp": 1.02167094,
      "epoch": 0.7887933625924367,
      "flos": 33365358126720.0,
      "grad_norm": 2.218956290372721,
      "language_loss": 0.74416322,
      "learning_rate": 4.498178526926886e-07,
      "loss": 0.76602489,
      "num_input_tokens_seen": 141420820,
      "step": 6560,
      "time_per_iteration": 2.783021926879883
    },
    {
      "auxiliary_loss_clip": 0.01132954,
      "auxiliary_loss_mlp": 0.01036373,
      "balance_loss_clip": 1.04844403,
      "balance_loss_mlp": 1.02835,
      "epoch": 0.7889136054830758,
      "flos": 17019360218880.0,
      "grad_norm": 2.1101935100424347,
      "language_loss": 0.72410977,
      "learning_rate": 4.4932577741722635e-07,
      "loss": 0.745803,
      "num_input_tokens_seen": 141439350,
      "step": 6561,
      "time_per_iteration": 2.7382333278656006
    },
    {
      "auxiliary_loss_clip": 0.01176031,
      "auxiliary_loss_mlp": 0.01030675,
      "balance_loss_clip": 1.05233741,
      "balance_loss_mlp": 1.02295637,
      "epoch": 0.7890338483737149,
      "flos": 29424629018880.0,
      "grad_norm": 2.0402036234079417,
      "language_loss": 0.74186146,
      "learning_rate": 4.4883393737780985e-07,
      "loss": 0.76392853,
      "num_input_tokens_seen": 141460300,
      "step": 6562,
      "time_per_iteration": 2.790055274963379
    },
    {
      "auxiliary_loss_clip": 0.01133218,
      "auxiliary_loss_mlp": 0.01027517,
      "balance_loss_clip": 1.04289794,
      "balance_loss_mlp": 1.02010202,
      "epoch": 0.789154091264354,
      "flos": 19971576063360.0,
      "grad_norm": 1.8996336853026612,
      "language_loss": 0.78340566,
      "learning_rate": 4.4834233264905254e-07,
      "loss": 0.80501306,
      "num_input_tokens_seen": 141477315,
      "step": 6563,
      "time_per_iteration": 2.7355408668518066
    },
    {
      "auxiliary_loss_clip": 0.01165574,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.0555011,
      "balance_loss_mlp": 1.02042925,
      "epoch": 0.789274334154993,
      "flos": 14537825216640.0,
      "grad_norm": 3.5625970967192364,
      "language_loss": 0.72135246,
      "learning_rate": 4.478509633055294e-07,
      "loss": 0.7432996,
      "num_input_tokens_seen": 141495025,
      "step": 6564,
      "time_per_iteration": 2.738330125808716
    },
    {
      "auxiliary_loss_clip": 0.01154547,
      "auxiliary_loss_mlp": 0.01025679,
      "balance_loss_clip": 1.05218458,
      "balance_loss_mlp": 1.01699424,
      "epoch": 0.7893945770456322,
      "flos": 21827403123840.0,
      "grad_norm": 2.6154277816741027,
      "language_loss": 0.80680799,
      "learning_rate": 4.473598294217813e-07,
      "loss": 0.8286103,
      "num_input_tokens_seen": 141510450,
      "step": 6565,
      "time_per_iteration": 2.7345077991485596
    },
    {
      "auxiliary_loss_clip": 0.01120231,
      "auxiliary_loss_mlp": 0.01026709,
      "balance_loss_clip": 1.04608047,
      "balance_loss_mlp": 1.01950574,
      "epoch": 0.7895148199362713,
      "flos": 20740639184640.0,
      "grad_norm": 2.263564941127847,
      "language_loss": 0.71302044,
      "learning_rate": 4.468689310723124e-07,
      "loss": 0.7344898,
      "num_input_tokens_seen": 141528265,
      "step": 6566,
      "time_per_iteration": 3.7031922340393066
    },
    {
      "auxiliary_loss_clip": 0.01157267,
      "auxiliary_loss_mlp": 0.01028599,
      "balance_loss_clip": 1.05153227,
      "balance_loss_mlp": 1.02118385,
      "epoch": 0.7896350628269103,
      "flos": 16690669839360.0,
      "grad_norm": 1.6842306247856134,
      "language_loss": 0.78642148,
      "learning_rate": 4.463782683315913e-07,
      "loss": 0.80828017,
      "num_input_tokens_seen": 141547270,
      "step": 6567,
      "time_per_iteration": 2.7222518920898438
    },
    {
      "auxiliary_loss_clip": 0.01161348,
      "auxiliary_loss_mlp": 0.01028228,
      "balance_loss_clip": 1.04912567,
      "balance_loss_mlp": 1.02071214,
      "epoch": 0.7897553057175495,
      "flos": 22638374438400.0,
      "grad_norm": 1.8128560065228418,
      "language_loss": 0.73564875,
      "learning_rate": 4.458878412740523e-07,
      "loss": 0.75754452,
      "num_input_tokens_seen": 141566050,
      "step": 6568,
      "time_per_iteration": 3.6097116470336914
    },
    {
      "auxiliary_loss_clip": 0.01133647,
      "auxiliary_loss_mlp": 0.01026874,
      "balance_loss_clip": 1.04640937,
      "balance_loss_mlp": 1.01826763,
      "epoch": 0.7898755486081885,
      "flos": 14537573821440.0,
      "grad_norm": 2.102333169090262,
      "language_loss": 0.77802515,
      "learning_rate": 4.453976499740919e-07,
      "loss": 0.79963034,
      "num_input_tokens_seen": 141583695,
      "step": 6569,
      "time_per_iteration": 2.734466791152954
    },
    {
      "auxiliary_loss_clip": 0.01161427,
      "auxiliary_loss_mlp": 0.01030998,
      "balance_loss_clip": 1.05271029,
      "balance_loss_mlp": 1.02325535,
      "epoch": 0.7899957914988276,
      "flos": 17238487138560.0,
      "grad_norm": 2.1774719213436415,
      "language_loss": 0.78123736,
      "learning_rate": 4.4490769450607215e-07,
      "loss": 0.80316162,
      "num_input_tokens_seen": 141601320,
      "step": 6570,
      "time_per_iteration": 2.658299684524536
    },
    {
      "auxiliary_loss_clip": 0.01154579,
      "auxiliary_loss_mlp": 0.0102599,
      "balance_loss_clip": 1.05270016,
      "balance_loss_mlp": 1.01788092,
      "epoch": 0.7901160343894668,
      "flos": 41279351086080.0,
      "grad_norm": 2.0600984206978805,
      "language_loss": 0.72564435,
      "learning_rate": 4.4441797494431845e-07,
      "loss": 0.74745011,
      "num_input_tokens_seen": 141623125,
      "step": 6571,
      "time_per_iteration": 2.8589112758636475
    },
    {
      "auxiliary_loss_clip": 0.01129492,
      "auxiliary_loss_mlp": 0.01030177,
      "balance_loss_clip": 1.04303491,
      "balance_loss_mlp": 1.02238631,
      "epoch": 0.7902362772801058,
      "flos": 16837005847680.0,
      "grad_norm": 1.9561281696774528,
      "language_loss": 0.77763337,
      "learning_rate": 4.439284913631207e-07,
      "loss": 0.79923004,
      "num_input_tokens_seen": 141640335,
      "step": 6572,
      "time_per_iteration": 3.71374773979187
    },
    {
      "auxiliary_loss_clip": 0.01104386,
      "auxiliary_loss_mlp": 0.01028828,
      "balance_loss_clip": 1.04616439,
      "balance_loss_mlp": 1.01978588,
      "epoch": 0.7903565201707449,
      "flos": 27125987091840.0,
      "grad_norm": 6.450331075901278,
      "language_loss": 0.84299821,
      "learning_rate": 4.434392438367347e-07,
      "loss": 0.86433041,
      "num_input_tokens_seen": 141659760,
      "step": 6573,
      "time_per_iteration": 2.803213596343994
    },
    {
      "auxiliary_loss_clip": 0.01123734,
      "auxiliary_loss_mlp": 0.01031137,
      "balance_loss_clip": 1.04584444,
      "balance_loss_mlp": 1.02317977,
      "epoch": 0.790476763061384,
      "flos": 31025167142400.0,
      "grad_norm": 1.950518052922414,
      "language_loss": 0.74138725,
      "learning_rate": 4.4295023243937677e-07,
      "loss": 0.762936,
      "num_input_tokens_seen": 141679965,
      "step": 6574,
      "time_per_iteration": 2.900132656097412
    },
    {
      "auxiliary_loss_clip": 0.01134725,
      "auxiliary_loss_mlp": 0.01030224,
      "balance_loss_clip": 1.04692912,
      "balance_loss_mlp": 1.02112818,
      "epoch": 0.7905970059520231,
      "flos": 22089084681600.0,
      "grad_norm": 2.8527981238483027,
      "language_loss": 0.80408478,
      "learning_rate": 4.4246145724523123e-07,
      "loss": 0.82573438,
      "num_input_tokens_seen": 141697710,
      "step": 6575,
      "time_per_iteration": 2.772280693054199
    },
    {
      "auxiliary_loss_clip": 0.01147344,
      "auxiliary_loss_mlp": 0.01029677,
      "balance_loss_clip": 1.04978538,
      "balance_loss_mlp": 1.02212477,
      "epoch": 0.7907172488426621,
      "flos": 20558141159040.0,
      "grad_norm": 2.2172828801405484,
      "language_loss": 0.77583241,
      "learning_rate": 4.41972918328444e-07,
      "loss": 0.79760265,
      "num_input_tokens_seen": 141715145,
      "step": 6576,
      "time_per_iteration": 2.6497437953948975
    },
    {
      "auxiliary_loss_clip": 0.01097038,
      "auxiliary_loss_mlp": 0.01032669,
      "balance_loss_clip": 1.0438199,
      "balance_loss_mlp": 1.02475381,
      "epoch": 0.7908374917333013,
      "flos": 30081542901120.0,
      "grad_norm": 2.358256369254762,
      "language_loss": 0.77359957,
      "learning_rate": 4.4148461576312646e-07,
      "loss": 0.79489666,
      "num_input_tokens_seen": 141734810,
      "step": 6577,
      "time_per_iteration": 2.8509061336517334
    },
    {
      "auxiliary_loss_clip": 0.01144715,
      "auxiliary_loss_mlp": 0.01031208,
      "balance_loss_clip": 1.05464339,
      "balance_loss_mlp": 1.02375162,
      "epoch": 0.7909577346239404,
      "flos": 20996359084800.0,
      "grad_norm": 2.3586800300317985,
      "language_loss": 0.7490716,
      "learning_rate": 4.4099654962335343e-07,
      "loss": 0.77083081,
      "num_input_tokens_seen": 141755260,
      "step": 6578,
      "time_per_iteration": 3.652156352996826
    },
    {
      "auxiliary_loss_clip": 0.01148524,
      "auxiliary_loss_mlp": 0.01030507,
      "balance_loss_clip": 1.04856873,
      "balance_loss_mlp": 1.0220108,
      "epoch": 0.7910779775145794,
      "flos": 26247935128320.0,
      "grad_norm": 1.9894293196743946,
      "language_loss": 0.75040758,
      "learning_rate": 4.405087199831636e-07,
      "loss": 0.7721979,
      "num_input_tokens_seen": 141775500,
      "step": 6579,
      "time_per_iteration": 2.7340314388275146
    },
    {
      "auxiliary_loss_clip": 0.01116948,
      "auxiliary_loss_mlp": 0.01025953,
      "balance_loss_clip": 1.04604185,
      "balance_loss_mlp": 1.01852012,
      "epoch": 0.7911982204052186,
      "flos": 22564434291840.0,
      "grad_norm": 9.263562546374816,
      "language_loss": 0.67398393,
      "learning_rate": 4.400211269165619e-07,
      "loss": 0.69541299,
      "num_input_tokens_seen": 141791955,
      "step": 6580,
      "time_per_iteration": 2.767995834350586
    },
    {
      "auxiliary_loss_clip": 0.01144234,
      "auxiliary_loss_mlp": 0.0103091,
      "balance_loss_clip": 1.05078554,
      "balance_loss_mlp": 1.02385581,
      "epoch": 0.7913184632958576,
      "flos": 23112538899840.0,
      "grad_norm": 1.6546176275938362,
      "language_loss": 0.76924527,
      "learning_rate": 4.3953377049751416e-07,
      "loss": 0.79099667,
      "num_input_tokens_seen": 141812380,
      "step": 6581,
      "time_per_iteration": 2.73873233795166
    },
    {
      "auxiliary_loss_clip": 0.01179682,
      "auxiliary_loss_mlp": 0.01030653,
      "balance_loss_clip": 1.05531311,
      "balance_loss_mlp": 1.02286863,
      "epoch": 0.7914387061864967,
      "flos": 12311758719360.0,
      "grad_norm": 2.2643973656385064,
      "language_loss": 0.78313065,
      "learning_rate": 4.390466507999537e-07,
      "loss": 0.80523396,
      "num_input_tokens_seen": 141828130,
      "step": 6582,
      "time_per_iteration": 2.635554075241089
    },
    {
      "auxiliary_loss_clip": 0.01119342,
      "auxiliary_loss_mlp": 0.01031627,
      "balance_loss_clip": 1.04570436,
      "balance_loss_mlp": 1.0238421,
      "epoch": 0.7915589490771359,
      "flos": 17603267708160.0,
      "grad_norm": 2.2982201043849386,
      "language_loss": 0.75846022,
      "learning_rate": 4.385597678977748e-07,
      "loss": 0.77996993,
      "num_input_tokens_seen": 141846965,
      "step": 6583,
      "time_per_iteration": 2.7157740592956543
    },
    {
      "auxiliary_loss_clip": 0.01160818,
      "auxiliary_loss_mlp": 0.01026736,
      "balance_loss_clip": 1.04998279,
      "balance_loss_mlp": 1.01843357,
      "epoch": 0.7916791919677749,
      "flos": 25591272641280.0,
      "grad_norm": 1.601220449335496,
      "language_loss": 0.7567082,
      "learning_rate": 4.3807312186483726e-07,
      "loss": 0.77858377,
      "num_input_tokens_seen": 141867685,
      "step": 6584,
      "time_per_iteration": 2.6857552528381348
    },
    {
      "auxiliary_loss_clip": 0.01138576,
      "auxiliary_loss_mlp": 0.01032257,
      "balance_loss_clip": 1.0497036,
      "balance_loss_mlp": 1.02369213,
      "epoch": 0.791799434858414,
      "flos": 18844340474880.0,
      "grad_norm": 1.8077497467008805,
      "language_loss": 0.7839303,
      "learning_rate": 4.375867127749655e-07,
      "loss": 0.80563861,
      "num_input_tokens_seen": 141885960,
      "step": 6585,
      "time_per_iteration": 2.6997225284576416
    },
    {
      "auxiliary_loss_clip": 0.01159751,
      "auxiliary_loss_mlp": 0.01024347,
      "balance_loss_clip": 1.05122852,
      "balance_loss_mlp": 1.01701009,
      "epoch": 0.7919196777490531,
      "flos": 25812015672960.0,
      "grad_norm": 2.1798530037888066,
      "language_loss": 0.67191279,
      "learning_rate": 4.3710054070194744e-07,
      "loss": 0.69375372,
      "num_input_tokens_seen": 141905655,
      "step": 6586,
      "time_per_iteration": 2.7358930110931396
    },
    {
      "auxiliary_loss_clip": 0.01140577,
      "auxiliary_loss_mlp": 0.01025606,
      "balance_loss_clip": 1.04650044,
      "balance_loss_mlp": 1.01708245,
      "epoch": 0.7920399206396922,
      "flos": 11947624594560.0,
      "grad_norm": 3.0152834650770126,
      "language_loss": 0.66501331,
      "learning_rate": 4.3661460571953455e-07,
      "loss": 0.68667507,
      "num_input_tokens_seen": 141922390,
      "step": 6587,
      "time_per_iteration": 2.6502344608306885
    },
    {
      "auxiliary_loss_clip": 0.01150462,
      "auxiliary_loss_mlp": 0.01025969,
      "balance_loss_clip": 1.04669118,
      "balance_loss_mlp": 1.01836967,
      "epoch": 0.7921601635303313,
      "flos": 21579907438080.0,
      "grad_norm": 1.5763546865112996,
      "language_loss": 0.68684053,
      "learning_rate": 4.36128907901443e-07,
      "loss": 0.70860487,
      "num_input_tokens_seen": 141941985,
      "step": 6588,
      "time_per_iteration": 2.7049267292022705
    },
    {
      "auxiliary_loss_clip": 0.01143564,
      "auxiliary_loss_mlp": 0.01030975,
      "balance_loss_clip": 1.04799783,
      "balance_loss_mlp": 1.02322602,
      "epoch": 0.7922804064209703,
      "flos": 18113989236480.0,
      "grad_norm": 3.4115074972292554,
      "language_loss": 0.72961044,
      "learning_rate": 4.356434473213519e-07,
      "loss": 0.75135577,
      "num_input_tokens_seen": 141959435,
      "step": 6589,
      "time_per_iteration": 2.6720001697540283
    },
    {
      "auxiliary_loss_clip": 0.01123448,
      "auxiliary_loss_mlp": 0.0102401,
      "balance_loss_clip": 1.04927266,
      "balance_loss_mlp": 1.01665521,
      "epoch": 0.7924006493116095,
      "flos": 21652806090240.0,
      "grad_norm": 1.898653484909071,
      "language_loss": 0.79767811,
      "learning_rate": 4.351582240529068e-07,
      "loss": 0.81915271,
      "num_input_tokens_seen": 141980265,
      "step": 6590,
      "time_per_iteration": 2.754121780395508
    },
    {
      "auxiliary_loss_clip": 0.01053677,
      "auxiliary_loss_mlp": 0.01002687,
      "balance_loss_clip": 1.00944924,
      "balance_loss_mlp": 1.00174546,
      "epoch": 0.7925208922022485,
      "flos": 64242755694720.0,
      "grad_norm": 0.7305052555111803,
      "language_loss": 0.58182919,
      "learning_rate": 4.346732381697149e-07,
      "loss": 0.60239285,
      "num_input_tokens_seen": 142044395,
      "step": 6591,
      "time_per_iteration": 3.2661056518554688
    },
    {
      "auxiliary_loss_clip": 0.01159839,
      "auxiliary_loss_mlp": 0.01027016,
      "balance_loss_clip": 1.05120409,
      "balance_loss_mlp": 1.01910067,
      "epoch": 0.7926411350928876,
      "flos": 16941541403520.0,
      "grad_norm": 2.229386477176805,
      "language_loss": 0.80852884,
      "learning_rate": 4.3418848974534825e-07,
      "loss": 0.83039737,
      "num_input_tokens_seen": 142061335,
      "step": 6592,
      "time_per_iteration": 3.512274980545044
    },
    {
      "auxiliary_loss_clip": 0.0110553,
      "auxiliary_loss_mlp": 0.01031648,
      "balance_loss_clip": 1.04980755,
      "balance_loss_mlp": 1.02411437,
      "epoch": 0.7927613779835267,
      "flos": 34459987144320.0,
      "grad_norm": 1.7862590017696196,
      "language_loss": 0.69225198,
      "learning_rate": 4.3370397885334276e-07,
      "loss": 0.71362376,
      "num_input_tokens_seen": 142081965,
      "step": 6593,
      "time_per_iteration": 2.9616682529449463
    },
    {
      "auxiliary_loss_clip": 0.01134553,
      "auxiliary_loss_mlp": 0.01028377,
      "balance_loss_clip": 1.04977763,
      "balance_loss_mlp": 1.02097964,
      "epoch": 0.7928816208741658,
      "flos": 18951174501120.0,
      "grad_norm": 1.8684972401607838,
      "language_loss": 0.7583288,
      "learning_rate": 4.3321970556719777e-07,
      "loss": 0.77995813,
      "num_input_tokens_seen": 142100260,
      "step": 6594,
      "time_per_iteration": 3.7312018871307373
    },
    {
      "auxiliary_loss_clip": 0.01136302,
      "auxiliary_loss_mlp": 0.00762885,
      "balance_loss_clip": 1.05040407,
      "balance_loss_mlp": 1.00042009,
      "epoch": 0.7930018637648049,
      "flos": 18623022825600.0,
      "grad_norm": 2.5768417640665877,
      "language_loss": 0.72163308,
      "learning_rate": 4.3273566996037856e-07,
      "loss": 0.7406249,
      "num_input_tokens_seen": 142116955,
      "step": 6595,
      "time_per_iteration": 2.752380847930908
    },
    {
      "auxiliary_loss_clip": 0.01141437,
      "auxiliary_loss_mlp": 0.00762153,
      "balance_loss_clip": 1.04958546,
      "balance_loss_mlp": 1.00049043,
      "epoch": 0.793122106655444,
      "flos": 24530650824960.0,
      "grad_norm": 2.3111694465119994,
      "language_loss": 0.80493033,
      "learning_rate": 4.322518721063113e-07,
      "loss": 0.82396626,
      "num_input_tokens_seen": 142135505,
      "step": 6596,
      "time_per_iteration": 2.759385585784912
    },
    {
      "auxiliary_loss_clip": 0.01133947,
      "auxiliary_loss_mlp": 0.01031136,
      "balance_loss_clip": 1.05154312,
      "balance_loss_mlp": 1.023507,
      "epoch": 0.7932423495460831,
      "flos": 34421203434240.0,
      "grad_norm": 1.926222151574546,
      "language_loss": 0.70666265,
      "learning_rate": 4.3176831207838906e-07,
      "loss": 0.72831351,
      "num_input_tokens_seen": 142158915,
      "step": 6597,
      "time_per_iteration": 3.552210807800293
    },
    {
      "auxiliary_loss_clip": 0.01122754,
      "auxiliary_loss_mlp": 0.01030605,
      "balance_loss_clip": 1.04580665,
      "balance_loss_mlp": 1.02282095,
      "epoch": 0.7933625924367221,
      "flos": 26980333441920.0,
      "grad_norm": 1.9579907601846962,
      "language_loss": 0.74862808,
      "learning_rate": 4.3128498994996685e-07,
      "loss": 0.77016169,
      "num_input_tokens_seen": 142178390,
      "step": 6598,
      "time_per_iteration": 2.810161590576172
    },
    {
      "auxiliary_loss_clip": 0.01122469,
      "auxiliary_loss_mlp": 0.01028717,
      "balance_loss_clip": 1.04363871,
      "balance_loss_mlp": 1.02045012,
      "epoch": 0.7934828353273613,
      "flos": 29568630643200.0,
      "grad_norm": 2.4970823324882145,
      "language_loss": 0.71296227,
      "learning_rate": 4.308019057943646e-07,
      "loss": 0.73447412,
      "num_input_tokens_seen": 142200115,
      "step": 6599,
      "time_per_iteration": 2.748793601989746
    },
    {
      "auxiliary_loss_clip": 0.01161695,
      "auxiliary_loss_mlp": 0.01022687,
      "balance_loss_clip": 1.05041623,
      "balance_loss_mlp": 1.01488435,
      "epoch": 0.7936030782180004,
      "flos": 28615381557120.0,
      "grad_norm": 1.690989175367658,
      "language_loss": 0.74713069,
      "learning_rate": 4.3031905968486535e-07,
      "loss": 0.76897448,
      "num_input_tokens_seen": 142220945,
      "step": 6600,
      "time_per_iteration": 2.785210371017456
    },
    {
      "auxiliary_loss_clip": 0.0116038,
      "auxiliary_loss_mlp": 0.01032215,
      "balance_loss_clip": 1.04912877,
      "balance_loss_mlp": 1.02436793,
      "epoch": 0.7937233211086394,
      "flos": 16392574869120.0,
      "grad_norm": 2.1974918910271524,
      "language_loss": 0.6852392,
      "learning_rate": 4.298364516947162e-07,
      "loss": 0.70716518,
      "num_input_tokens_seen": 142238175,
      "step": 6601,
      "time_per_iteration": 2.711463689804077
    },
    {
      "auxiliary_loss_clip": 0.01137204,
      "auxiliary_loss_mlp": 0.01020779,
      "balance_loss_clip": 1.04714108,
      "balance_loss_mlp": 1.01279151,
      "epoch": 0.7938435639992786,
      "flos": 22013420682240.0,
      "grad_norm": 2.04892974909972,
      "language_loss": 0.65549099,
      "learning_rate": 4.293540818971295e-07,
      "loss": 0.67707086,
      "num_input_tokens_seen": 142255980,
      "step": 6602,
      "time_per_iteration": 2.6832005977630615
    },
    {
      "auxiliary_loss_clip": 0.01176741,
      "auxiliary_loss_mlp": 0.01026852,
      "balance_loss_clip": 1.05233133,
      "balance_loss_mlp": 1.0182091,
      "epoch": 0.7939638068899176,
      "flos": 22197032029440.0,
      "grad_norm": 2.284860685976398,
      "language_loss": 0.76626873,
      "learning_rate": 4.2887195036527934e-07,
      "loss": 0.78830475,
      "num_input_tokens_seen": 142274785,
      "step": 6603,
      "time_per_iteration": 2.7134459018707275
    },
    {
      "auxiliary_loss_clip": 0.01157434,
      "auxiliary_loss_mlp": 0.00762096,
      "balance_loss_clip": 1.04860425,
      "balance_loss_mlp": 1.00047779,
      "epoch": 0.7940840497805567,
      "flos": 17745186343680.0,
      "grad_norm": 2.4851104337127734,
      "language_loss": 0.73838294,
      "learning_rate": 4.28390057172306e-07,
      "loss": 0.75757825,
      "num_input_tokens_seen": 142291290,
      "step": 6604,
      "time_per_iteration": 3.533923387527466
    },
    {
      "auxiliary_loss_clip": 0.01106939,
      "auxiliary_loss_mlp": 0.01026616,
      "balance_loss_clip": 1.04670429,
      "balance_loss_mlp": 1.01815844,
      "epoch": 0.7942042926711959,
      "flos": 23805435231360.0,
      "grad_norm": 2.0409678877129855,
      "language_loss": 0.72061706,
      "learning_rate": 4.279084023913111e-07,
      "loss": 0.74195266,
      "num_input_tokens_seen": 142309165,
      "step": 6605,
      "time_per_iteration": 2.8125126361846924
    },
    {
      "auxiliary_loss_clip": 0.01109104,
      "auxiliary_loss_mlp": 0.01031063,
      "balance_loss_clip": 1.04861093,
      "balance_loss_mlp": 1.02326715,
      "epoch": 0.7943245355618349,
      "flos": 19244959839360.0,
      "grad_norm": 2.6287922133460793,
      "language_loss": 0.70013857,
      "learning_rate": 4.2742698609536096e-07,
      "loss": 0.72154033,
      "num_input_tokens_seen": 142327475,
      "step": 6606,
      "time_per_iteration": 2.8250679969787598
    },
    {
      "auxiliary_loss_clip": 0.01166118,
      "auxiliary_loss_mlp": 0.010325,
      "balance_loss_clip": 1.05219126,
      "balance_loss_mlp": 1.0242095,
      "epoch": 0.794444778452474,
      "flos": 25007616547200.0,
      "grad_norm": 2.829918459903907,
      "language_loss": 0.78832537,
      "learning_rate": 4.2694580835748706e-07,
      "loss": 0.81031156,
      "num_input_tokens_seen": 142347335,
      "step": 6607,
      "time_per_iteration": 2.700805425643921
    },
    {
      "auxiliary_loss_clip": 0.01120011,
      "auxiliary_loss_mlp": 0.01028266,
      "balance_loss_clip": 1.04614925,
      "balance_loss_mlp": 1.02046323,
      "epoch": 0.7945650213431131,
      "flos": 23221491828480.0,
      "grad_norm": 2.1853677458691707,
      "language_loss": 0.74203587,
      "learning_rate": 4.264648692506836e-07,
      "loss": 0.76351857,
      "num_input_tokens_seen": 142366125,
      "step": 6608,
      "time_per_iteration": 2.7831926345825195
    },
    {
      "auxiliary_loss_clip": 0.01163189,
      "auxiliary_loss_mlp": 0.01027591,
      "balance_loss_clip": 1.05227196,
      "balance_loss_mlp": 1.01932383,
      "epoch": 0.7946852642337522,
      "flos": 26062887237120.0,
      "grad_norm": 5.368577758653654,
      "language_loss": 0.72123909,
      "learning_rate": 4.2598416884790824e-07,
      "loss": 0.7431469,
      "num_input_tokens_seen": 142385175,
      "step": 6609,
      "time_per_iteration": 2.67820143699646
    },
    {
      "auxiliary_loss_clip": 0.01124713,
      "auxiliary_loss_mlp": 0.01029517,
      "balance_loss_clip": 1.04814947,
      "balance_loss_mlp": 1.0210588,
      "epoch": 0.7948055071243912,
      "flos": 23769704177280.0,
      "grad_norm": 2.799855949585238,
      "language_loss": 0.81081808,
      "learning_rate": 4.255037072220828e-07,
      "loss": 0.83236045,
      "num_input_tokens_seen": 142406545,
      "step": 6610,
      "time_per_iteration": 2.737438201904297
    },
    {
      "auxiliary_loss_clip": 0.01109412,
      "auxiliary_loss_mlp": 0.01025932,
      "balance_loss_clip": 1.04106283,
      "balance_loss_mlp": 1.01815319,
      "epoch": 0.7949257500150304,
      "flos": 21980814111360.0,
      "grad_norm": 6.41942333460549,
      "language_loss": 0.718651,
      "learning_rate": 4.2502348444609293e-07,
      "loss": 0.74000442,
      "num_input_tokens_seen": 142426165,
      "step": 6611,
      "time_per_iteration": 2.710751533508301
    },
    {
      "auxiliary_loss_clip": 0.01177345,
      "auxiliary_loss_mlp": 0.0102918,
      "balance_loss_clip": 1.05384505,
      "balance_loss_mlp": 1.02144909,
      "epoch": 0.7950459929056695,
      "flos": 25774129802880.0,
      "grad_norm": 1.978315695349598,
      "language_loss": 0.694947,
      "learning_rate": 4.2454350059278844e-07,
      "loss": 0.71701229,
      "num_input_tokens_seen": 142447225,
      "step": 6612,
      "time_per_iteration": 2.826298713684082
    },
    {
      "auxiliary_loss_clip": 0.01106314,
      "auxiliary_loss_mlp": 0.01031862,
      "balance_loss_clip": 1.0429666,
      "balance_loss_mlp": 1.02396441,
      "epoch": 0.7951662357963085,
      "flos": 22158068751360.0,
      "grad_norm": 1.911148881152028,
      "language_loss": 0.84491336,
      "learning_rate": 4.240637557349824e-07,
      "loss": 0.8662951,
      "num_input_tokens_seen": 142464440,
      "step": 6613,
      "time_per_iteration": 2.752376079559326
    },
    {
      "auxiliary_loss_clip": 0.01174297,
      "auxiliary_loss_mlp": 0.01028701,
      "balance_loss_clip": 1.05310082,
      "balance_loss_mlp": 1.0209198,
      "epoch": 0.7952864786869477,
      "flos": 24641938137600.0,
      "grad_norm": 2.1412847650192672,
      "language_loss": 0.67045861,
      "learning_rate": 4.235842499454516e-07,
      "loss": 0.69248861,
      "num_input_tokens_seen": 142484355,
      "step": 6614,
      "time_per_iteration": 2.7244162559509277
    },
    {
      "auxiliary_loss_clip": 0.01157917,
      "auxiliary_loss_mlp": 0.01025198,
      "balance_loss_clip": 1.05084515,
      "balance_loss_mlp": 1.01703203,
      "epoch": 0.7954067215775867,
      "flos": 21830922656640.0,
      "grad_norm": 1.6945534289952324,
      "language_loss": 0.82903916,
      "learning_rate": 4.2310498329693687e-07,
      "loss": 0.85087037,
      "num_input_tokens_seen": 142505255,
      "step": 6615,
      "time_per_iteration": 2.6123523712158203
    },
    {
      "auxiliary_loss_clip": 0.01124005,
      "auxiliary_loss_mlp": 0.01029015,
      "balance_loss_clip": 1.04971516,
      "balance_loss_mlp": 1.0202353,
      "epoch": 0.7955269644682258,
      "flos": 24060652341120.0,
      "grad_norm": 1.8686461724841763,
      "language_loss": 0.81116331,
      "learning_rate": 4.2262595586214164e-07,
      "loss": 0.83269352,
      "num_input_tokens_seen": 142526350,
      "step": 6616,
      "time_per_iteration": 2.7980778217315674
    },
    {
      "auxiliary_loss_clip": 0.01138316,
      "auxiliary_loss_mlp": 0.0103097,
      "balance_loss_clip": 1.04882383,
      "balance_loss_mlp": 1.02261913,
      "epoch": 0.795647207358865,
      "flos": 25010741030400.0,
      "grad_norm": 1.728419404177329,
      "language_loss": 0.76812488,
      "learning_rate": 4.221471677137358e-07,
      "loss": 0.78981775,
      "num_input_tokens_seen": 142547165,
      "step": 6617,
      "time_per_iteration": 3.6867973804473877
    },
    {
      "auxiliary_loss_clip": 0.01144661,
      "auxiliary_loss_mlp": 0.01026831,
      "balance_loss_clip": 1.05037117,
      "balance_loss_mlp": 1.01928174,
      "epoch": 0.795767450249504,
      "flos": 14648358343680.0,
      "grad_norm": 2.0831235606330294,
      "language_loss": 0.7052114,
      "learning_rate": 4.216686189243492e-07,
      "loss": 0.72692633,
      "num_input_tokens_seen": 142565955,
      "step": 6618,
      "time_per_iteration": 2.7008748054504395
    },
    {
      "auxiliary_loss_clip": 0.01146612,
      "auxiliary_loss_mlp": 0.01024227,
      "balance_loss_clip": 1.04745936,
      "balance_loss_mlp": 1.01556027,
      "epoch": 0.7958876931401431,
      "flos": 18547897530240.0,
      "grad_norm": 3.4557425122469883,
      "language_loss": 0.72807229,
      "learning_rate": 4.211903095665785e-07,
      "loss": 0.74978071,
      "num_input_tokens_seen": 142585340,
      "step": 6619,
      "time_per_iteration": 2.675614833831787
    },
    {
      "auxiliary_loss_clip": 0.01131126,
      "auxiliary_loss_mlp": 0.01031427,
      "balance_loss_clip": 1.04541814,
      "balance_loss_mlp": 1.02324939,
      "epoch": 0.7960079360307821,
      "flos": 21543960902400.0,
      "grad_norm": 5.407004307941558,
      "language_loss": 0.75262451,
      "learning_rate": 4.2071223971298277e-07,
      "loss": 0.77425009,
      "num_input_tokens_seen": 142602525,
      "step": 6620,
      "time_per_iteration": 3.558868885040283
    },
    {
      "auxiliary_loss_clip": 0.01164048,
      "auxiliary_loss_mlp": 0.01030677,
      "balance_loss_clip": 1.05260634,
      "balance_loss_mlp": 1.02218938,
      "epoch": 0.7961281789214213,
      "flos": 25481745095040.0,
      "grad_norm": 2.5069324971764764,
      "language_loss": 0.61318302,
      "learning_rate": 4.2023440943608433e-07,
      "loss": 0.63513029,
      "num_input_tokens_seen": 142622490,
      "step": 6621,
      "time_per_iteration": 2.679164409637451
    },
    {
      "auxiliary_loss_clip": 0.01154226,
      "auxiliary_loss_mlp": 0.01024416,
      "balance_loss_clip": 1.0488801,
      "balance_loss_mlp": 1.01718044,
      "epoch": 0.7962484218120603,
      "flos": 21944436612480.0,
      "grad_norm": 1.828222200910977,
      "language_loss": 0.78036773,
      "learning_rate": 4.1975681880837023e-07,
      "loss": 0.80215418,
      "num_input_tokens_seen": 142642495,
      "step": 6622,
      "time_per_iteration": 2.8504717350006104
    },
    {
      "auxiliary_loss_clip": 0.01145157,
      "auxiliary_loss_mlp": 0.00762105,
      "balance_loss_clip": 1.04657102,
      "balance_loss_mlp": 1.00049114,
      "epoch": 0.7963686647026994,
      "flos": 18876264687360.0,
      "grad_norm": 1.6606905097186588,
      "language_loss": 0.83137327,
      "learning_rate": 4.192794679022895e-07,
      "loss": 0.85044593,
      "num_input_tokens_seen": 142660820,
      "step": 6623,
      "time_per_iteration": 3.6483516693115234
    },
    {
      "auxiliary_loss_clip": 0.01127801,
      "auxiliary_loss_mlp": 0.01028044,
      "balance_loss_clip": 1.04661274,
      "balance_loss_mlp": 1.02026522,
      "epoch": 0.7964889075933386,
      "flos": 29716582763520.0,
      "grad_norm": 3.243447327303591,
      "language_loss": 0.72277522,
      "learning_rate": 4.1880235679025743e-07,
      "loss": 0.74433368,
      "num_input_tokens_seen": 142680915,
      "step": 6624,
      "time_per_iteration": 2.784419059753418
    },
    {
      "auxiliary_loss_clip": 0.01165916,
      "auxiliary_loss_mlp": 0.01025892,
      "balance_loss_clip": 1.05048108,
      "balance_loss_mlp": 1.01774383,
      "epoch": 0.7966091504839776,
      "flos": 29491458272640.0,
      "grad_norm": 1.838869716904733,
      "language_loss": 0.64007449,
      "learning_rate": 4.1832548554464986e-07,
      "loss": 0.66199261,
      "num_input_tokens_seen": 142699210,
      "step": 6625,
      "time_per_iteration": 2.711611270904541
    },
    {
      "auxiliary_loss_clip": 0.01021734,
      "auxiliary_loss_mlp": 0.01003728,
      "balance_loss_clip": 1.00897467,
      "balance_loss_mlp": 1.00277388,
      "epoch": 0.7967293933746167,
      "flos": 67288697101440.0,
      "grad_norm": 0.7411063138506084,
      "language_loss": 0.58667612,
      "learning_rate": 4.178488542378098e-07,
      "loss": 0.60693085,
      "num_input_tokens_seen": 142756790,
      "step": 6626,
      "time_per_iteration": 3.2449169158935547
    },
    {
      "auxiliary_loss_clip": 0.01179262,
      "auxiliary_loss_mlp": 0.01029536,
      "balance_loss_clip": 1.05274832,
      "balance_loss_mlp": 1.02089965,
      "epoch": 0.7968496362652558,
      "flos": 25554679660800.0,
      "grad_norm": 2.1975062869791038,
      "language_loss": 0.89115232,
      "learning_rate": 4.173724629420401e-07,
      "loss": 0.91324031,
      "num_input_tokens_seen": 142778150,
      "step": 6627,
      "time_per_iteration": 2.6545827388763428
    },
    {
      "auxiliary_loss_clip": 0.01163522,
      "auxiliary_loss_mlp": 0.0103013,
      "balance_loss_clip": 1.05018377,
      "balance_loss_mlp": 1.02191663,
      "epoch": 0.7969698791558949,
      "flos": 14501088581760.0,
      "grad_norm": 2.61536500595949,
      "language_loss": 0.68512279,
      "learning_rate": 4.168963117296087e-07,
      "loss": 0.70705932,
      "num_input_tokens_seen": 142795485,
      "step": 6628,
      "time_per_iteration": 2.620272397994995
    },
    {
      "auxiliary_loss_clip": 0.01145089,
      "auxiliary_loss_mlp": 0.01032111,
      "balance_loss_clip": 1.04720747,
      "balance_loss_mlp": 1.024333,
      "epoch": 0.797090122046534,
      "flos": 22127545169280.0,
      "grad_norm": 2.1973474153027315,
      "language_loss": 0.75875348,
      "learning_rate": 4.1642040067274876e-07,
      "loss": 0.78052545,
      "num_input_tokens_seen": 142815155,
      "step": 6629,
      "time_per_iteration": 2.7441518306732178
    },
    {
      "auxiliary_loss_clip": 0.0116043,
      "auxiliary_loss_mlp": 0.01023492,
      "balance_loss_clip": 1.05082655,
      "balance_loss_mlp": 1.01617837,
      "epoch": 0.7972103649371731,
      "flos": 19897671830400.0,
      "grad_norm": 1.6188746144566661,
      "language_loss": 0.7288174,
      "learning_rate": 4.1594472984365493e-07,
      "loss": 0.7506566,
      "num_input_tokens_seen": 142833840,
      "step": 6630,
      "time_per_iteration": 3.6105880737304688
    },
    {
      "auxiliary_loss_clip": 0.01129629,
      "auxiliary_loss_mlp": 0.01031275,
      "balance_loss_clip": 1.04653895,
      "balance_loss_mlp": 1.02371407,
      "epoch": 0.7973306078278122,
      "flos": 36058621847040.0,
      "grad_norm": 1.8402040348067505,
      "language_loss": 0.778018,
      "learning_rate": 4.154692993144862e-07,
      "loss": 0.79962707,
      "num_input_tokens_seen": 142853610,
      "step": 6631,
      "time_per_iteration": 2.8935699462890625
    },
    {
      "auxiliary_loss_clip": 0.01092132,
      "auxiliary_loss_mlp": 0.01028473,
      "balance_loss_clip": 1.04533291,
      "balance_loss_mlp": 1.02059281,
      "epoch": 0.7974508507184512,
      "flos": 21360600950400.0,
      "grad_norm": 2.0240081225080413,
      "language_loss": 0.71564102,
      "learning_rate": 4.1499410915736476e-07,
      "loss": 0.73684704,
      "num_input_tokens_seen": 142872540,
      "step": 6632,
      "time_per_iteration": 2.813936710357666
    },
    {
      "auxiliary_loss_clip": 0.01034511,
      "auxiliary_loss_mlp": 0.00998684,
      "balance_loss_clip": 1.01005876,
      "balance_loss_mlp": 0.9978258,
      "epoch": 0.7975710936090904,
      "flos": 68253115317120.0,
      "grad_norm": 0.7729208496879311,
      "language_loss": 0.64184511,
      "learning_rate": 4.145191594443762e-07,
      "loss": 0.66217703,
      "num_input_tokens_seen": 142936895,
      "step": 6633,
      "time_per_iteration": 3.411149263381958
    },
    {
      "auxiliary_loss_clip": 0.01124638,
      "auxiliary_loss_mlp": 0.0103283,
      "balance_loss_clip": 1.04634345,
      "balance_loss_mlp": 1.02389562,
      "epoch": 0.7976913364997295,
      "flos": 22492433479680.0,
      "grad_norm": 2.3041080409157493,
      "language_loss": 0.70569599,
      "learning_rate": 4.140444502475713e-07,
      "loss": 0.72727072,
      "num_input_tokens_seen": 142956445,
      "step": 6634,
      "time_per_iteration": 2.7810139656066895
    },
    {
      "auxiliary_loss_clip": 0.01102257,
      "auxiliary_loss_mlp": 0.01031664,
      "balance_loss_clip": 1.04078555,
      "balance_loss_mlp": 1.02431464,
      "epoch": 0.7978115793903685,
      "flos": 15263220378240.0,
      "grad_norm": 2.307976776578536,
      "language_loss": 0.70159233,
      "learning_rate": 4.1356998163896216e-07,
      "loss": 0.7229315,
      "num_input_tokens_seen": 142973495,
      "step": 6635,
      "time_per_iteration": 2.7393131256103516
    },
    {
      "auxiliary_loss_clip": 0.01146258,
      "auxiliary_loss_mlp": 0.01024783,
      "balance_loss_clip": 1.05160594,
      "balance_loss_mlp": 1.01712942,
      "epoch": 0.7979318222810077,
      "flos": 19719232041600.0,
      "grad_norm": 5.607175546145064,
      "language_loss": 0.74851185,
      "learning_rate": 4.130957536905255e-07,
      "loss": 0.77022231,
      "num_input_tokens_seen": 142991510,
      "step": 6636,
      "time_per_iteration": 2.8401427268981934
    },
    {
      "auxiliary_loss_clip": 0.01154861,
      "auxiliary_loss_mlp": 0.0102476,
      "balance_loss_clip": 1.05169797,
      "balance_loss_mlp": 1.01667166,
      "epoch": 0.7980520651716467,
      "flos": 15560273854080.0,
      "grad_norm": 2.9345526912637996,
      "language_loss": 0.71841776,
      "learning_rate": 4.1262176647420134e-07,
      "loss": 0.74021393,
      "num_input_tokens_seen": 143009675,
      "step": 6637,
      "time_per_iteration": 2.6888022422790527
    },
    {
      "auxiliary_loss_clip": 0.01144323,
      "auxiliary_loss_mlp": 0.01027584,
      "balance_loss_clip": 1.04937458,
      "balance_loss_mlp": 1.01956725,
      "epoch": 0.7981723080622858,
      "flos": 22309432663680.0,
      "grad_norm": 1.774579453687472,
      "language_loss": 0.80298233,
      "learning_rate": 4.121480200618923e-07,
      "loss": 0.82470137,
      "num_input_tokens_seen": 143029330,
      "step": 6638,
      "time_per_iteration": 2.7074337005615234
    },
    {
      "auxiliary_loss_clip": 0.0114124,
      "auxiliary_loss_mlp": 0.0103283,
      "balance_loss_clip": 1.0481317,
      "balance_loss_mlp": 1.02537334,
      "epoch": 0.798292550952925,
      "flos": 22929573997440.0,
      "grad_norm": 4.581648818769683,
      "language_loss": 0.80125535,
      "learning_rate": 4.116745145254674e-07,
      "loss": 0.82299608,
      "num_input_tokens_seen": 143048865,
      "step": 6639,
      "time_per_iteration": 2.7172436714172363
    },
    {
      "auxiliary_loss_clip": 0.0106263,
      "auxiliary_loss_mlp": 0.01003518,
      "balance_loss_clip": 1.008564,
      "balance_loss_mlp": 1.00264788,
      "epoch": 0.798412793843564,
      "flos": 64497936890880.0,
      "grad_norm": 0.764095325970675,
      "language_loss": 0.57818681,
      "learning_rate": 4.1120124993675476e-07,
      "loss": 0.59884828,
      "num_input_tokens_seen": 143113295,
      "step": 6640,
      "time_per_iteration": 3.2340164184570312
    },
    {
      "auxiliary_loss_clip": 0.01179327,
      "auxiliary_loss_mlp": 0.01029912,
      "balance_loss_clip": 1.05299711,
      "balance_loss_mlp": 1.02193081,
      "epoch": 0.7985330367342031,
      "flos": 13586910514560.0,
      "grad_norm": 3.8394055730136096,
      "language_loss": 0.62053019,
      "learning_rate": 4.107282263675498e-07,
      "loss": 0.64262259,
      "num_input_tokens_seen": 143130965,
      "step": 6641,
      "time_per_iteration": 2.664626359939575
    },
    {
      "auxiliary_loss_clip": 0.01030888,
      "auxiliary_loss_mlp": 0.0100364,
      "balance_loss_clip": 1.00978088,
      "balance_loss_mlp": 1.00268006,
      "epoch": 0.7986532796248422,
      "flos": 67698797656320.0,
      "grad_norm": 0.7590572870069934,
      "language_loss": 0.52395993,
      "learning_rate": 4.1025544388960907e-07,
      "loss": 0.54430526,
      "num_input_tokens_seen": 143192005,
      "step": 6642,
      "time_per_iteration": 3.2288930416107178
    },
    {
      "auxiliary_loss_clip": 0.01150663,
      "auxiliary_loss_mlp": 0.01024987,
      "balance_loss_clip": 1.04997325,
      "balance_loss_mlp": 1.01766133,
      "epoch": 0.7987735225154813,
      "flos": 22455373622400.0,
      "grad_norm": 1.799350817057805,
      "language_loss": 0.72279173,
      "learning_rate": 4.097829025746538e-07,
      "loss": 0.7445482,
      "num_input_tokens_seen": 143213550,
      "step": 6643,
      "time_per_iteration": 3.6464555263519287
    },
    {
      "auxiliary_loss_clip": 0.01039995,
      "auxiliary_loss_mlp": 0.0100096,
      "balance_loss_clip": 1.00893259,
      "balance_loss_mlp": 1.0000422,
      "epoch": 0.7988937654061203,
      "flos": 68864098682880.0,
      "grad_norm": 0.6716132803550738,
      "language_loss": 0.60928613,
      "learning_rate": 4.0931060249436757e-07,
      "loss": 0.62969565,
      "num_input_tokens_seen": 143277390,
      "step": 6644,
      "time_per_iteration": 3.280885696411133
    },
    {
      "auxiliary_loss_clip": 0.01142675,
      "auxiliary_loss_mlp": 0.01025459,
      "balance_loss_clip": 1.04976749,
      "balance_loss_mlp": 1.01811576,
      "epoch": 0.7990140082967595,
      "flos": 20806893820800.0,
      "grad_norm": 2.8620196417337778,
      "language_loss": 0.6957466,
      "learning_rate": 4.088385437203978e-07,
      "loss": 0.71742797,
      "num_input_tokens_seen": 143294400,
      "step": 6645,
      "time_per_iteration": 2.64083194732666
    },
    {
      "auxiliary_loss_clip": 0.01123,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.04321647,
      "balance_loss_mlp": 1.01939225,
      "epoch": 0.7991342511873986,
      "flos": 18985289443200.0,
      "grad_norm": 2.2277832527522134,
      "language_loss": 0.77495718,
      "learning_rate": 4.083667263243564e-07,
      "loss": 0.79646719,
      "num_input_tokens_seen": 143312745,
      "step": 6646,
      "time_per_iteration": 3.6455729007720947
    },
    {
      "auxiliary_loss_clip": 0.01126198,
      "auxiliary_loss_mlp": 0.01029888,
      "balance_loss_clip": 1.0437423,
      "balance_loss_mlp": 1.02183509,
      "epoch": 0.7992544940780376,
      "flos": 20816805974400.0,
      "grad_norm": 1.7159035321606306,
      "language_loss": 0.71815151,
      "learning_rate": 4.0789515037781653e-07,
      "loss": 0.73971236,
      "num_input_tokens_seen": 143333470,
      "step": 6647,
      "time_per_iteration": 2.7900118827819824
    },
    {
      "auxiliary_loss_clip": 0.01100596,
      "auxiliary_loss_mlp": 0.01027071,
      "balance_loss_clip": 1.04563105,
      "balance_loss_mlp": 1.01927495,
      "epoch": 0.7993747369686768,
      "flos": 12640772321280.0,
      "grad_norm": 1.725988625093956,
      "language_loss": 0.82838142,
      "learning_rate": 4.0742381595231755e-07,
      "loss": 0.84965801,
      "num_input_tokens_seen": 143350195,
      "step": 6648,
      "time_per_iteration": 2.7369577884674072
    },
    {
      "auxiliary_loss_clip": 0.01159431,
      "auxiliary_loss_mlp": 0.01027601,
      "balance_loss_clip": 1.05018055,
      "balance_loss_mlp": 1.01995993,
      "epoch": 0.7994949798593158,
      "flos": 20078769225600.0,
      "grad_norm": 1.8277313582844166,
      "language_loss": 0.78514177,
      "learning_rate": 4.06952723119359e-07,
      "loss": 0.80701208,
      "num_input_tokens_seen": 143370070,
      "step": 6649,
      "time_per_iteration": 3.6091573238372803
    },
    {
      "auxiliary_loss_clip": 0.01110711,
      "auxiliary_loss_mlp": 0.01027604,
      "balance_loss_clip": 1.04445088,
      "balance_loss_mlp": 1.01952446,
      "epoch": 0.7996152227499549,
      "flos": 38654209509120.0,
      "grad_norm": 2.254046818006814,
      "language_loss": 0.67529184,
      "learning_rate": 4.0648187195040504e-07,
      "loss": 0.69667494,
      "num_input_tokens_seen": 143392275,
      "step": 6650,
      "time_per_iteration": 2.9046828746795654
    },
    {
      "auxiliary_loss_clip": 0.01016125,
      "auxiliary_loss_mlp": 0.01004386,
      "balance_loss_clip": 1.00797093,
      "balance_loss_mlp": 1.00345027,
      "epoch": 0.799735465640594,
      "flos": 70243821947520.0,
      "grad_norm": 0.8152009386253118,
      "language_loss": 0.6747548,
      "learning_rate": 4.060112625168848e-07,
      "loss": 0.69495994,
      "num_input_tokens_seen": 143457385,
      "step": 6651,
      "time_per_iteration": 3.4105679988861084
    },
    {
      "auxiliary_loss_clip": 0.01166158,
      "auxiliary_loss_mlp": 0.01033712,
      "balance_loss_clip": 1.05340981,
      "balance_loss_mlp": 1.02535534,
      "epoch": 0.7998557085312331,
      "flos": 24240995550720.0,
      "grad_norm": 1.7550620890732518,
      "language_loss": 0.74105793,
      "learning_rate": 4.055408948901886e-07,
      "loss": 0.76305664,
      "num_input_tokens_seen": 143478785,
      "step": 6652,
      "time_per_iteration": 2.7131717205047607
    },
    {
      "auxiliary_loss_clip": 0.0108351,
      "auxiliary_loss_mlp": 0.00762857,
      "balance_loss_clip": 1.04323494,
      "balance_loss_mlp": 1.00047231,
      "epoch": 0.7999759514218722,
      "flos": 27564025449600.0,
      "grad_norm": 2.627335804912432,
      "language_loss": 0.71349943,
      "learning_rate": 4.050707691416708e-07,
      "loss": 0.73196316,
      "num_input_tokens_seen": 143500095,
      "step": 6653,
      "time_per_iteration": 2.830726146697998
    },
    {
      "auxiliary_loss_clip": 0.01031916,
      "auxiliary_loss_mlp": 0.00999363,
      "balance_loss_clip": 1.01987755,
      "balance_loss_mlp": 0.99841559,
      "epoch": 0.8000961943125112,
      "flos": 67337428878720.0,
      "grad_norm": 0.6801636129310252,
      "language_loss": 0.59751576,
      "learning_rate": 4.046008853426495e-07,
      "loss": 0.61782855,
      "num_input_tokens_seen": 143563410,
      "step": 6654,
      "time_per_iteration": 3.350985288619995
    },
    {
      "auxiliary_loss_clip": 0.01164464,
      "auxiliary_loss_mlp": 0.01028588,
      "balance_loss_clip": 1.04941511,
      "balance_loss_mlp": 1.02054691,
      "epoch": 0.8002164372031504,
      "flos": 28733815676160.0,
      "grad_norm": 1.7121269320102586,
      "language_loss": 0.6304965,
      "learning_rate": 4.0413124356440464e-07,
      "loss": 0.65242702,
      "num_input_tokens_seen": 143587455,
      "step": 6655,
      "time_per_iteration": 2.7166550159454346
    },
    {
      "auxiliary_loss_clip": 0.01089003,
      "auxiliary_loss_mlp": 0.01026359,
      "balance_loss_clip": 1.04052043,
      "balance_loss_mlp": 1.01843476,
      "epoch": 0.8003366800937894,
      "flos": 17639429725440.0,
      "grad_norm": 1.7944019739265336,
      "language_loss": 0.82322919,
      "learning_rate": 4.0366184387818223e-07,
      "loss": 0.84438276,
      "num_input_tokens_seen": 143605915,
      "step": 6656,
      "time_per_iteration": 3.718862533569336
    },
    {
      "auxiliary_loss_clip": 0.01151662,
      "auxiliary_loss_mlp": 0.010279,
      "balance_loss_clip": 1.04939651,
      "balance_loss_mlp": 1.01932847,
      "epoch": 0.8004569229844285,
      "flos": 25995303797760.0,
      "grad_norm": 1.7353953126544517,
      "language_loss": 0.85317534,
      "learning_rate": 4.0319268635518797e-07,
      "loss": 0.87497091,
      "num_input_tokens_seen": 143626490,
      "step": 6657,
      "time_per_iteration": 2.922426462173462
    },
    {
      "auxiliary_loss_clip": 0.01099784,
      "auxiliary_loss_mlp": 0.01026714,
      "balance_loss_clip": 1.04400444,
      "balance_loss_mlp": 1.01902533,
      "epoch": 0.8005771658750677,
      "flos": 20812352688000.0,
      "grad_norm": 1.5449647249657426,
      "language_loss": 0.75239813,
      "learning_rate": 4.027237710665943e-07,
      "loss": 0.7736631,
      "num_input_tokens_seen": 143644955,
      "step": 6658,
      "time_per_iteration": 2.78908109664917
    },
    {
      "auxiliary_loss_clip": 0.01141658,
      "auxiliary_loss_mlp": 0.01030605,
      "balance_loss_clip": 1.04752088,
      "balance_loss_mlp": 1.02236736,
      "epoch": 0.8006974087657067,
      "flos": 25812626204160.0,
      "grad_norm": 2.069341295963505,
      "language_loss": 0.69610643,
      "learning_rate": 4.022550980835344e-07,
      "loss": 0.71782905,
      "num_input_tokens_seen": 143667200,
      "step": 6659,
      "time_per_iteration": 2.792482376098633
    },
    {
      "auxiliary_loss_clip": 0.01155032,
      "auxiliary_loss_mlp": 0.010267,
      "balance_loss_clip": 1.0478934,
      "balance_loss_mlp": 1.01837921,
      "epoch": 0.8008176516563458,
      "flos": 17164690646400.0,
      "grad_norm": 1.937656404886313,
      "language_loss": 0.79665273,
      "learning_rate": 4.017866674771051e-07,
      "loss": 0.81847012,
      "num_input_tokens_seen": 143684685,
      "step": 6660,
      "time_per_iteration": 2.692063331604004
    },
    {
      "auxiliary_loss_clip": 0.01150033,
      "auxiliary_loss_mlp": 0.01031557,
      "balance_loss_clip": 1.04738426,
      "balance_loss_mlp": 1.02368069,
      "epoch": 0.8009378945469849,
      "flos": 24207311571840.0,
      "grad_norm": 3.1510992753133173,
      "language_loss": 0.74717611,
      "learning_rate": 4.013184793183688e-07,
      "loss": 0.76899201,
      "num_input_tokens_seen": 143706780,
      "step": 6661,
      "time_per_iteration": 2.77470064163208
    },
    {
      "auxiliary_loss_clip": 0.0104712,
      "auxiliary_loss_mlp": 0.00762536,
      "balance_loss_clip": 1.03352988,
      "balance_loss_mlp": 1.00056958,
      "epoch": 0.801058137437624,
      "flos": 19787318271360.0,
      "grad_norm": 4.116979138122877,
      "language_loss": 0.72805119,
      "learning_rate": 4.008505336783472e-07,
      "loss": 0.74614775,
      "num_input_tokens_seen": 143724505,
      "step": 6662,
      "time_per_iteration": 3.1127448081970215
    },
    {
      "auxiliary_loss_clip": 0.01157358,
      "auxiliary_loss_mlp": 0.01027704,
      "balance_loss_clip": 1.04915261,
      "balance_loss_mlp": 1.02018166,
      "epoch": 0.801178380328263,
      "flos": 18659400324480.0,
      "grad_norm": 1.9609440998514673,
      "language_loss": 0.80952793,
      "learning_rate": 4.003828306280284e-07,
      "loss": 0.83137852,
      "num_input_tokens_seen": 143742180,
      "step": 6663,
      "time_per_iteration": 2.8510801792144775
    },
    {
      "auxiliary_loss_clip": 0.01144424,
      "auxiliary_loss_mlp": 0.01024838,
      "balance_loss_clip": 1.04921889,
      "balance_loss_mlp": 1.01766205,
      "epoch": 0.8012986232189022,
      "flos": 15706573948800.0,
      "grad_norm": 2.0289238266308702,
      "language_loss": 0.78200293,
      "learning_rate": 3.999153702383626e-07,
      "loss": 0.8036955,
      "num_input_tokens_seen": 143760070,
      "step": 6664,
      "time_per_iteration": 2.6672568321228027
    },
    {
      "auxiliary_loss_clip": 0.01132116,
      "auxiliary_loss_mlp": 0.01030581,
      "balance_loss_clip": 1.04880381,
      "balance_loss_mlp": 1.02212298,
      "epoch": 0.8014188661095413,
      "flos": 28584139703040.0,
      "grad_norm": 2.018394669530696,
      "language_loss": 0.73772722,
      "learning_rate": 3.9944815258026263e-07,
      "loss": 0.75935417,
      "num_input_tokens_seen": 143781890,
      "step": 6665,
      "time_per_iteration": 2.74101185798645
    },
    {
      "auxiliary_loss_clip": 0.011689,
      "auxiliary_loss_mlp": 0.01031591,
      "balance_loss_clip": 1.05271518,
      "balance_loss_mlp": 1.02331161,
      "epoch": 0.8015391090001803,
      "flos": 29310360877440.0,
      "grad_norm": 1.7300685315962332,
      "language_loss": 0.83152878,
      "learning_rate": 3.989811777246057e-07,
      "loss": 0.85353369,
      "num_input_tokens_seen": 143802060,
      "step": 6666,
      "time_per_iteration": 2.7373104095458984
    },
    {
      "auxiliary_loss_clip": 0.01062569,
      "auxiliary_loss_mlp": 0.01002367,
      "balance_loss_clip": 1.00832379,
      "balance_loss_mlp": 1.00148439,
      "epoch": 0.8016593518908195,
      "flos": 70397340675840.0,
      "grad_norm": 0.850585099975037,
      "language_loss": 0.66143775,
      "learning_rate": 3.985144457422305e-07,
      "loss": 0.68208718,
      "num_input_tokens_seen": 143856345,
      "step": 6667,
      "time_per_iteration": 3.1700451374053955
    },
    {
      "auxiliary_loss_clip": 0.01124743,
      "auxiliary_loss_mlp": 0.00761346,
      "balance_loss_clip": 1.04781651,
      "balance_loss_mlp": 1.00051856,
      "epoch": 0.8017795947814585,
      "flos": 26026114688640.0,
      "grad_norm": 2.7691515920662217,
      "language_loss": 0.76943004,
      "learning_rate": 3.9804795670394096e-07,
      "loss": 0.78829092,
      "num_input_tokens_seen": 143876470,
      "step": 6668,
      "time_per_iteration": 3.9868950843811035
    },
    {
      "auxiliary_loss_clip": 0.01090334,
      "auxiliary_loss_mlp": 0.01023093,
      "balance_loss_clip": 1.04207754,
      "balance_loss_mlp": 1.01546955,
      "epoch": 0.8018998376720976,
      "flos": 22087181260800.0,
      "grad_norm": 1.810008238137935,
      "language_loss": 0.70298803,
      "learning_rate": 3.975817106805022e-07,
      "loss": 0.72412229,
      "num_input_tokens_seen": 143895170,
      "step": 6669,
      "time_per_iteration": 2.9369544982910156
    },
    {
      "auxiliary_loss_clip": 0.01145302,
      "auxiliary_loss_mlp": 0.01030764,
      "balance_loss_clip": 1.04730642,
      "balance_loss_mlp": 1.02173352,
      "epoch": 0.8020200805627368,
      "flos": 34568545023360.0,
      "grad_norm": 1.9794449778766092,
      "language_loss": 0.65348685,
      "learning_rate": 3.97115707742645e-07,
      "loss": 0.67524755,
      "num_input_tokens_seen": 143915845,
      "step": 6670,
      "time_per_iteration": 2.850377321243286
    },
    {
      "auxiliary_loss_clip": 0.01132707,
      "auxiliary_loss_mlp": 0.01029033,
      "balance_loss_clip": 1.04914856,
      "balance_loss_mlp": 1.02076614,
      "epoch": 0.8021403234533758,
      "flos": 20120354196480.0,
      "grad_norm": 2.5156560436198108,
      "language_loss": 0.6590488,
      "learning_rate": 3.966499479610599e-07,
      "loss": 0.68066621,
      "num_input_tokens_seen": 143933940,
      "step": 6671,
      "time_per_iteration": 3.899031639099121
    },
    {
      "auxiliary_loss_clip": 0.01132287,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.04937363,
      "balance_loss_mlp": 1.02104473,
      "epoch": 0.8022605663440149,
      "flos": 27746200252800.0,
      "grad_norm": 2.1016858327922288,
      "language_loss": 0.65068507,
      "learning_rate": 3.9618443140640225e-07,
      "loss": 0.67229527,
      "num_input_tokens_seen": 143952850,
      "step": 6672,
      "time_per_iteration": 2.854870557785034
    },
    {
      "auxiliary_loss_clip": 0.01002112,
      "auxiliary_loss_mlp": 0.00752001,
      "balance_loss_clip": 1.00788331,
      "balance_loss_mlp": 1.00030732,
      "epoch": 0.802380809234654,
      "flos": 60244998768000.0,
      "grad_norm": 0.6888279583609253,
      "language_loss": 0.51336777,
      "learning_rate": 3.957191581492918e-07,
      "loss": 0.53090894,
      "num_input_tokens_seen": 144013610,
      "step": 6673,
      "time_per_iteration": 3.533205986022949
    },
    {
      "auxiliary_loss_clip": 0.01141048,
      "auxiliary_loss_mlp": 0.01024915,
      "balance_loss_clip": 1.05029237,
      "balance_loss_mlp": 1.01615334,
      "epoch": 0.8025010521252931,
      "flos": 15080722352640.0,
      "grad_norm": 2.747884830142856,
      "language_loss": 0.70981961,
      "learning_rate": 3.952541282603097e-07,
      "loss": 0.73147929,
      "num_input_tokens_seen": 144028715,
      "step": 6674,
      "time_per_iteration": 3.287813901901245
    },
    {
      "auxiliary_loss_clip": 0.01140884,
      "auxiliary_loss_mlp": 0.01027723,
      "balance_loss_clip": 1.04664898,
      "balance_loss_mlp": 1.01943183,
      "epoch": 0.8026212950159322,
      "flos": 22163527618560.0,
      "grad_norm": 1.9360699907912304,
      "language_loss": 0.8374604,
      "learning_rate": 3.9478934181000013e-07,
      "loss": 0.85914648,
      "num_input_tokens_seen": 144048740,
      "step": 6675,
      "time_per_iteration": 3.6705873012542725
    },
    {
      "auxiliary_loss_clip": 0.01117555,
      "auxiliary_loss_mlp": 0.01029535,
      "balance_loss_clip": 1.04466534,
      "balance_loss_mlp": 1.02124381,
      "epoch": 0.8027415379065713,
      "flos": 17675986792320.0,
      "grad_norm": 2.351595558109996,
      "language_loss": 0.84703028,
      "learning_rate": 3.943247988688714e-07,
      "loss": 0.86850113,
      "num_input_tokens_seen": 144067435,
      "step": 6676,
      "time_per_iteration": 2.7795777320861816
    },
    {
      "auxiliary_loss_clip": 0.01111217,
      "auxiliary_loss_mlp": 0.01031402,
      "balance_loss_clip": 1.04575884,
      "balance_loss_mlp": 1.02386522,
      "epoch": 0.8028617807972104,
      "flos": 21979593048960.0,
      "grad_norm": 2.2001199884188454,
      "language_loss": 0.72230655,
      "learning_rate": 3.938604995073933e-07,
      "loss": 0.74373275,
      "num_input_tokens_seen": 144085905,
      "step": 6677,
      "time_per_iteration": 2.773204803466797
    },
    {
      "auxiliary_loss_clip": 0.01148568,
      "auxiliary_loss_mlp": 0.01028003,
      "balance_loss_clip": 1.05026007,
      "balance_loss_mlp": 1.02027249,
      "epoch": 0.8029820236878494,
      "flos": 26428457905920.0,
      "grad_norm": 4.415506755873739,
      "language_loss": 0.65096879,
      "learning_rate": 3.9339644379600157e-07,
      "loss": 0.6727345,
      "num_input_tokens_seen": 144105735,
      "step": 6678,
      "time_per_iteration": 2.809924840927124
    },
    {
      "auxiliary_loss_clip": 0.01175004,
      "auxiliary_loss_mlp": 0.0076214,
      "balance_loss_clip": 1.05424917,
      "balance_loss_mlp": 1.00049591,
      "epoch": 0.8031022665784886,
      "flos": 17676489582720.0,
      "grad_norm": 1.9665883510043056,
      "language_loss": 0.71404636,
      "learning_rate": 3.929326318050907e-07,
      "loss": 0.73341775,
      "num_input_tokens_seen": 144123405,
      "step": 6679,
      "time_per_iteration": 2.5971198081970215
    },
    {
      "auxiliary_loss_clip": 0.01158755,
      "auxiliary_loss_mlp": 0.01027421,
      "balance_loss_clip": 1.04803526,
      "balance_loss_mlp": 1.02014017,
      "epoch": 0.8032225094691277,
      "flos": 15450279431040.0,
      "grad_norm": 1.891147991663071,
      "language_loss": 0.7916429,
      "learning_rate": 3.924690636050225e-07,
      "loss": 0.81350464,
      "num_input_tokens_seen": 144140815,
      "step": 6680,
      "time_per_iteration": 2.698885917663574
    },
    {
      "auxiliary_loss_clip": 0.01175072,
      "auxiliary_loss_mlp": 0.01030147,
      "balance_loss_clip": 1.0505302,
      "balance_loss_mlp": 1.02158237,
      "epoch": 0.8033427523597667,
      "flos": 26179202453760.0,
      "grad_norm": 1.825436859276818,
      "language_loss": 0.7263087,
      "learning_rate": 3.9200573926611915e-07,
      "loss": 0.74836093,
      "num_input_tokens_seen": 144162230,
      "step": 6681,
      "time_per_iteration": 3.694399356842041
    },
    {
      "auxiliary_loss_clip": 0.01162088,
      "auxiliary_loss_mlp": 0.01030927,
      "balance_loss_clip": 1.05264819,
      "balance_loss_mlp": 1.0232321,
      "epoch": 0.8034629952504058,
      "flos": 21324905809920.0,
      "grad_norm": 2.109763413488445,
      "language_loss": 0.73071814,
      "learning_rate": 3.9154265885866613e-07,
      "loss": 0.75264829,
      "num_input_tokens_seen": 144181540,
      "step": 6682,
      "time_per_iteration": 2.8287291526794434
    },
    {
      "auxiliary_loss_clip": 0.0115382,
      "auxiliary_loss_mlp": 0.01028433,
      "balance_loss_clip": 1.05054951,
      "balance_loss_mlp": 1.02028513,
      "epoch": 0.8035832381410449,
      "flos": 21651585027840.0,
      "grad_norm": 2.3557833296385016,
      "language_loss": 0.74847811,
      "learning_rate": 3.9107982245291394e-07,
      "loss": 0.77030063,
      "num_input_tokens_seen": 144199665,
      "step": 6683,
      "time_per_iteration": 2.769474506378174
    },
    {
      "auxiliary_loss_clip": 0.01123354,
      "auxiliary_loss_mlp": 0.01029179,
      "balance_loss_clip": 1.04635382,
      "balance_loss_mlp": 1.02084041,
      "epoch": 0.803703481031684,
      "flos": 20518818744960.0,
      "grad_norm": 5.288358042304856,
      "language_loss": 0.77432418,
      "learning_rate": 3.9061723011907245e-07,
      "loss": 0.79584956,
      "num_input_tokens_seen": 144219020,
      "step": 6684,
      "time_per_iteration": 2.7863950729370117
    },
    {
      "auxiliary_loss_clip": 0.01134585,
      "auxiliary_loss_mlp": 0.01030039,
      "balance_loss_clip": 1.04813397,
      "balance_loss_mlp": 1.02159286,
      "epoch": 0.803823723922323,
      "flos": 22854807838080.0,
      "grad_norm": 1.6676002316177962,
      "language_loss": 0.79484057,
      "learning_rate": 3.901548819273179e-07,
      "loss": 0.81648672,
      "num_input_tokens_seen": 144239035,
      "step": 6685,
      "time_per_iteration": 2.770622491836548
    },
    {
      "auxiliary_loss_clip": 0.01162246,
      "auxiliary_loss_mlp": 0.01025829,
      "balance_loss_clip": 1.05379105,
      "balance_loss_mlp": 1.01775217,
      "epoch": 0.8039439668129622,
      "flos": 21362145235200.0,
      "grad_norm": 2.5274058556090817,
      "language_loss": 0.69433832,
      "learning_rate": 3.896927779477881e-07,
      "loss": 0.71621907,
      "num_input_tokens_seen": 144258295,
      "step": 6686,
      "time_per_iteration": 2.6338255405426025
    },
    {
      "auxiliary_loss_clip": 0.01137267,
      "auxiliary_loss_mlp": 0.01025078,
      "balance_loss_clip": 1.04915667,
      "balance_loss_mlp": 1.01750827,
      "epoch": 0.8040642097036013,
      "flos": 23802382575360.0,
      "grad_norm": 2.191833666457615,
      "language_loss": 0.67069352,
      "learning_rate": 3.892309182505833e-07,
      "loss": 0.69231701,
      "num_input_tokens_seen": 144276110,
      "step": 6687,
      "time_per_iteration": 2.7422938346862793
    },
    {
      "auxiliary_loss_clip": 0.01149043,
      "auxiliary_loss_mlp": 0.01027353,
      "balance_loss_clip": 1.04845762,
      "balance_loss_mlp": 1.01983356,
      "epoch": 0.8041844525942403,
      "flos": 25922046009600.0,
      "grad_norm": 3.260518224848855,
      "language_loss": 0.85954612,
      "learning_rate": 3.887693029057675e-07,
      "loss": 0.88130999,
      "num_input_tokens_seen": 144295620,
      "step": 6688,
      "time_per_iteration": 2.6801536083221436
    },
    {
      "auxiliary_loss_clip": 0.01158247,
      "auxiliary_loss_mlp": 0.01022456,
      "balance_loss_clip": 1.04771447,
      "balance_loss_mlp": 1.01528013,
      "epoch": 0.8043046954848795,
      "flos": 25191120153600.0,
      "grad_norm": 1.70766063096676,
      "language_loss": 0.8145498,
      "learning_rate": 3.8830793198336684e-07,
      "loss": 0.83635688,
      "num_input_tokens_seen": 144315210,
      "step": 6689,
      "time_per_iteration": 2.7755162715911865
    },
    {
      "auxiliary_loss_clip": 0.01147276,
      "auxiliary_loss_mlp": 0.01030385,
      "balance_loss_clip": 1.04790175,
      "balance_loss_mlp": 1.02208269,
      "epoch": 0.8044249383755185,
      "flos": 41719185123840.0,
      "grad_norm": 1.703059428738575,
      "language_loss": 0.70524341,
      "learning_rate": 3.878468055533721e-07,
      "loss": 0.72702003,
      "num_input_tokens_seen": 144337750,
      "step": 6690,
      "time_per_iteration": 2.8544504642486572
    },
    {
      "auxiliary_loss_clip": 0.01148671,
      "auxiliary_loss_mlp": 0.01031532,
      "balance_loss_clip": 1.04975557,
      "balance_loss_mlp": 1.023175,
      "epoch": 0.8045451812661576,
      "flos": 20631434860800.0,
      "grad_norm": 3.827014624168808,
      "language_loss": 0.85533112,
      "learning_rate": 3.8738592368573464e-07,
      "loss": 0.87713313,
      "num_input_tokens_seen": 144355305,
      "step": 6691,
      "time_per_iteration": 2.7044897079467773
    },
    {
      "auxiliary_loss_clip": 0.0115873,
      "auxiliary_loss_mlp": 0.01031372,
      "balance_loss_clip": 1.05148721,
      "balance_loss_mlp": 1.0227294,
      "epoch": 0.8046654241567968,
      "flos": 29711806254720.0,
      "grad_norm": 1.9665531052314138,
      "language_loss": 0.88055205,
      "learning_rate": 3.8692528645037137e-07,
      "loss": 0.90245306,
      "num_input_tokens_seen": 144374485,
      "step": 6692,
      "time_per_iteration": 2.714665651321411
    },
    {
      "auxiliary_loss_clip": 0.01150697,
      "auxiliary_loss_mlp": 0.01023971,
      "balance_loss_clip": 1.04912698,
      "balance_loss_mlp": 1.01635909,
      "epoch": 0.8047856670474358,
      "flos": 17671389851520.0,
      "grad_norm": 2.632763139522781,
      "language_loss": 0.77709401,
      "learning_rate": 3.8646489391715907e-07,
      "loss": 0.7988407,
      "num_input_tokens_seen": 144388780,
      "step": 6693,
      "time_per_iteration": 2.666910409927368
    },
    {
      "auxiliary_loss_clip": 0.01152414,
      "auxiliary_loss_mlp": 0.01025585,
      "balance_loss_clip": 1.04868543,
      "balance_loss_mlp": 1.01769018,
      "epoch": 0.8049059099380749,
      "flos": 17120699464320.0,
      "grad_norm": 9.528098290350533,
      "language_loss": 0.87991089,
      "learning_rate": 3.8600474615593903e-07,
      "loss": 0.9016909,
      "num_input_tokens_seen": 144403395,
      "step": 6694,
      "time_per_iteration": 3.529177188873291
    },
    {
      "auxiliary_loss_clip": 0.01041105,
      "auxiliary_loss_mlp": 0.01002199,
      "balance_loss_clip": 1.00880527,
      "balance_loss_mlp": 1.00123334,
      "epoch": 0.805026152828714,
      "flos": 62212903240320.0,
      "grad_norm": 0.7862762777795221,
      "language_loss": 0.59611833,
      "learning_rate": 3.8554484323651605e-07,
      "loss": 0.61655134,
      "num_input_tokens_seen": 144465265,
      "step": 6695,
      "time_per_iteration": 3.2759618759155273
    },
    {
      "auxiliary_loss_clip": 0.01125566,
      "auxiliary_loss_mlp": 0.01029598,
      "balance_loss_clip": 1.04411316,
      "balance_loss_mlp": 1.02215981,
      "epoch": 0.8051463957193531,
      "flos": 21688608971520.0,
      "grad_norm": 1.6745560917119466,
      "language_loss": 0.79292095,
      "learning_rate": 3.85085185228657e-07,
      "loss": 0.81447256,
      "num_input_tokens_seen": 144484235,
      "step": 6696,
      "time_per_iteration": 2.7642152309417725
    },
    {
      "auxiliary_loss_clip": 0.01135841,
      "auxiliary_loss_mlp": 0.01025885,
      "balance_loss_clip": 1.04885566,
      "balance_loss_mlp": 1.01782608,
      "epoch": 0.8052666386099921,
      "flos": 32051458535040.0,
      "grad_norm": 1.9965596703480115,
      "language_loss": 0.73443758,
      "learning_rate": 3.8462577220209114e-07,
      "loss": 0.75605482,
      "num_input_tokens_seen": 144504610,
      "step": 6697,
      "time_per_iteration": 3.587118625640869
    },
    {
      "auxiliary_loss_clip": 0.01025547,
      "auxiliary_loss_mlp": 0.01004977,
      "balance_loss_clip": 1.00964212,
      "balance_loss_mlp": 1.00391567,
      "epoch": 0.8053868815006313,
      "flos": 67157875768320.0,
      "grad_norm": 0.7046902216569718,
      "language_loss": 0.58868122,
      "learning_rate": 3.8416660422651127e-07,
      "loss": 0.60898644,
      "num_input_tokens_seen": 144574260,
      "step": 6698,
      "time_per_iteration": 3.4533913135528564
    },
    {
      "auxiliary_loss_clip": 0.01097866,
      "auxiliary_loss_mlp": 0.01028533,
      "balance_loss_clip": 1.04228294,
      "balance_loss_mlp": 1.02053428,
      "epoch": 0.8055071243912704,
      "flos": 23837000307840.0,
      "grad_norm": 3.995773779582425,
      "language_loss": 0.6805166,
      "learning_rate": 3.837076813715723e-07,
      "loss": 0.70178056,
      "num_input_tokens_seen": 144594145,
      "step": 6699,
      "time_per_iteration": 3.146897554397583
    },
    {
      "auxiliary_loss_clip": 0.01159189,
      "auxiliary_loss_mlp": 0.01025723,
      "balance_loss_clip": 1.05359006,
      "balance_loss_mlp": 1.01740229,
      "epoch": 0.8056273672819094,
      "flos": 21324510760320.0,
      "grad_norm": 2.6295377544374974,
      "language_loss": 0.7528075,
      "learning_rate": 3.832490037068941e-07,
      "loss": 0.77465665,
      "num_input_tokens_seen": 144612935,
      "step": 6700,
      "time_per_iteration": 2.8042945861816406
    },
    {
      "auxiliary_loss_clip": 0.0114692,
      "auxiliary_loss_mlp": 0.01026011,
      "balance_loss_clip": 1.04677498,
      "balance_loss_mlp": 1.01822639,
      "epoch": 0.8057476101725486,
      "flos": 25768383626880.0,
      "grad_norm": 1.9403397558668853,
      "language_loss": 0.76195109,
      "learning_rate": 3.827905713020554e-07,
      "loss": 0.78368044,
      "num_input_tokens_seen": 144630580,
      "step": 6701,
      "time_per_iteration": 3.620363235473633
    },
    {
      "auxiliary_loss_clip": 0.01116719,
      "auxiliary_loss_mlp": 0.01039531,
      "balance_loss_clip": 1.04834843,
      "balance_loss_mlp": 1.03036368,
      "epoch": 0.8058678530631876,
      "flos": 24535283679360.0,
      "grad_norm": 3.1260234241760987,
      "language_loss": 0.69040406,
      "learning_rate": 3.823323842266017e-07,
      "loss": 0.71196663,
      "num_input_tokens_seen": 144649975,
      "step": 6702,
      "time_per_iteration": 2.809138059616089
    },
    {
      "auxiliary_loss_clip": 0.01136598,
      "auxiliary_loss_mlp": 0.01025043,
      "balance_loss_clip": 1.04721928,
      "balance_loss_mlp": 1.0171932,
      "epoch": 0.8059880959538267,
      "flos": 24753728240640.0,
      "grad_norm": 4.243869794537535,
      "language_loss": 0.7347815,
      "learning_rate": 3.818744425500393e-07,
      "loss": 0.75639796,
      "num_input_tokens_seen": 144667990,
      "step": 6703,
      "time_per_iteration": 2.724634885787964
    },
    {
      "auxiliary_loss_clip": 0.01146423,
      "auxiliary_loss_mlp": 0.01030011,
      "balance_loss_clip": 1.04845548,
      "balance_loss_mlp": 1.02267373,
      "epoch": 0.8061083388444659,
      "flos": 22196349671040.0,
      "grad_norm": 2.400643462866497,
      "language_loss": 0.80964541,
      "learning_rate": 3.8141674634183675e-07,
      "loss": 0.83140981,
      "num_input_tokens_seen": 144687020,
      "step": 6704,
      "time_per_iteration": 2.687086582183838
    },
    {
      "auxiliary_loss_clip": 0.01170798,
      "auxiliary_loss_mlp": 0.01029884,
      "balance_loss_clip": 1.05137801,
      "balance_loss_mlp": 1.02268374,
      "epoch": 0.8062285817351049,
      "flos": 30044195735040.0,
      "grad_norm": 3.00364053625656,
      "language_loss": 0.66584992,
      "learning_rate": 3.809592956714278e-07,
      "loss": 0.68785679,
      "num_input_tokens_seen": 144710255,
      "step": 6705,
      "time_per_iteration": 2.6923928260803223
    },
    {
      "auxiliary_loss_clip": 0.01114411,
      "auxiliary_loss_mlp": 0.00762565,
      "balance_loss_clip": 1.04792714,
      "balance_loss_mlp": 1.00046945,
      "epoch": 0.806348824625744,
      "flos": 22782591544320.0,
      "grad_norm": 1.878336419957491,
      "language_loss": 0.75370312,
      "learning_rate": 3.805020906082057e-07,
      "loss": 0.77247286,
      "num_input_tokens_seen": 144728830,
      "step": 6706,
      "time_per_iteration": 2.7500288486480713
    },
    {
      "auxiliary_loss_clip": 0.01176392,
      "auxiliary_loss_mlp": 0.01036074,
      "balance_loss_clip": 1.05091715,
      "balance_loss_mlp": 1.02752042,
      "epoch": 0.8064690675163831,
      "flos": 23404600385280.0,
      "grad_norm": 2.5095839863586913,
      "language_loss": 0.81385267,
      "learning_rate": 3.8004513122152917e-07,
      "loss": 0.83597732,
      "num_input_tokens_seen": 144747140,
      "step": 6707,
      "time_per_iteration": 3.7006638050079346
    },
    {
      "auxiliary_loss_clip": 0.01143175,
      "auxiliary_loss_mlp": 0.01025853,
      "balance_loss_clip": 1.04979408,
      "balance_loss_mlp": 1.018641,
      "epoch": 0.8065893104070222,
      "flos": 24060903736320.0,
      "grad_norm": 1.7331608051092466,
      "language_loss": 0.6736353,
      "learning_rate": 3.79588417580718e-07,
      "loss": 0.69532561,
      "num_input_tokens_seen": 144765250,
      "step": 6708,
      "time_per_iteration": 2.6622025966644287
    },
    {
      "auxiliary_loss_clip": 0.01161672,
      "auxiliary_loss_mlp": 0.01029154,
      "balance_loss_clip": 1.04971313,
      "balance_loss_mlp": 1.02092314,
      "epoch": 0.8067095532976613,
      "flos": 22305410340480.0,
      "grad_norm": 2.046074056564922,
      "language_loss": 0.76526654,
      "learning_rate": 3.791319497550558e-07,
      "loss": 0.78717476,
      "num_input_tokens_seen": 144783080,
      "step": 6709,
      "time_per_iteration": 2.705259084701538
    },
    {
      "auxiliary_loss_clip": 0.01146877,
      "auxiliary_loss_mlp": 0.01026253,
      "balance_loss_clip": 1.05294871,
      "balance_loss_mlp": 1.01888001,
      "epoch": 0.8068297961883004,
      "flos": 17129498296320.0,
      "grad_norm": 2.4414048392167342,
      "language_loss": 0.71287453,
      "learning_rate": 3.78675727813788e-07,
      "loss": 0.73460579,
      "num_input_tokens_seen": 144800645,
      "step": 6710,
      "time_per_iteration": 2.670252561569214
    },
    {
      "auxiliary_loss_clip": 0.01165286,
      "auxiliary_loss_mlp": 0.01033151,
      "balance_loss_clip": 1.04929507,
      "balance_loss_mlp": 1.02476156,
      "epoch": 0.8069500390789395,
      "flos": 22018843635840.0,
      "grad_norm": 2.9551321594363635,
      "language_loss": 0.73539102,
      "learning_rate": 3.782197518261225e-07,
      "loss": 0.75737536,
      "num_input_tokens_seen": 144820085,
      "step": 6711,
      "time_per_iteration": 2.7123610973358154
    },
    {
      "auxiliary_loss_clip": 0.01143059,
      "auxiliary_loss_mlp": 0.01032965,
      "balance_loss_clip": 1.04768455,
      "balance_loss_mlp": 1.02513885,
      "epoch": 0.8070702819695785,
      "flos": 19244241567360.0,
      "grad_norm": 2.195695456387141,
      "language_loss": 0.95711625,
      "learning_rate": 3.777640218612319e-07,
      "loss": 0.97887653,
      "num_input_tokens_seen": 144838070,
      "step": 6712,
      "time_per_iteration": 2.7360572814941406
    },
    {
      "auxiliary_loss_clip": 0.01163599,
      "auxiliary_loss_mlp": 0.01017928,
      "balance_loss_clip": 1.05104125,
      "balance_loss_mlp": 1.01063228,
      "epoch": 0.8071905248602176,
      "flos": 21544320038400.0,
      "grad_norm": 2.228008258273178,
      "language_loss": 0.72154284,
      "learning_rate": 3.773085379882488e-07,
      "loss": 0.74335808,
      "num_input_tokens_seen": 144857125,
      "step": 6713,
      "time_per_iteration": 2.6762173175811768
    },
    {
      "auxiliary_loss_clip": 0.01123169,
      "auxiliary_loss_mlp": 0.01029318,
      "balance_loss_clip": 1.04469681,
      "balance_loss_mlp": 1.02125943,
      "epoch": 0.8073107677508568,
      "flos": 37268309105280.0,
      "grad_norm": 1.812022185576088,
      "language_loss": 0.75919956,
      "learning_rate": 3.768533002762715e-07,
      "loss": 0.78072441,
      "num_input_tokens_seen": 144880660,
      "step": 6714,
      "time_per_iteration": 2.8825652599334717
    },
    {
      "auxiliary_loss_clip": 0.01159958,
      "auxiliary_loss_mlp": 0.01027689,
      "balance_loss_clip": 1.04838765,
      "balance_loss_mlp": 1.02013075,
      "epoch": 0.8074310106414958,
      "flos": 28366269759360.0,
      "grad_norm": 2.0110886125950924,
      "language_loss": 0.77034819,
      "learning_rate": 3.763983087943572e-07,
      "loss": 0.79222465,
      "num_input_tokens_seen": 144900050,
      "step": 6715,
      "time_per_iteration": 2.668264150619507
    },
    {
      "auxiliary_loss_clip": 0.01112617,
      "auxiliary_loss_mlp": 0.00762267,
      "balance_loss_clip": 1.04275131,
      "balance_loss_mlp": 1.0005312,
      "epoch": 0.8075512535321349,
      "flos": 24281646768000.0,
      "grad_norm": 1.6631861717959473,
      "language_loss": 0.81408256,
      "learning_rate": 3.759435636115282e-07,
      "loss": 0.83283138,
      "num_input_tokens_seen": 144920835,
      "step": 6716,
      "time_per_iteration": 2.8070664405822754
    },
    {
      "auxiliary_loss_clip": 0.01163262,
      "auxiliary_loss_mlp": 0.01028966,
      "balance_loss_clip": 1.05302572,
      "balance_loss_mlp": 1.02112794,
      "epoch": 0.807671496422774,
      "flos": 26030855283840.0,
      "grad_norm": 1.9206948165185533,
      "language_loss": 0.73246908,
      "learning_rate": 3.7548906479676967e-07,
      "loss": 0.75439137,
      "num_input_tokens_seen": 144940430,
      "step": 6717,
      "time_per_iteration": 2.739016532897949
    },
    {
      "auxiliary_loss_clip": 0.01130189,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 1.04911458,
      "balance_loss_mlp": 1.02073956,
      "epoch": 0.8077917393134131,
      "flos": 23730740899200.0,
      "grad_norm": 1.730408130883691,
      "language_loss": 0.718692,
      "learning_rate": 3.7503481241902855e-07,
      "loss": 0.74028099,
      "num_input_tokens_seen": 144960405,
      "step": 6718,
      "time_per_iteration": 2.7752814292907715
    },
    {
      "auxiliary_loss_clip": 0.0115727,
      "auxiliary_loss_mlp": 0.01031941,
      "balance_loss_clip": 1.05119753,
      "balance_loss_mlp": 1.02449083,
      "epoch": 0.8079119822040521,
      "flos": 18402028398720.0,
      "grad_norm": 2.231667194787982,
      "language_loss": 0.80353749,
      "learning_rate": 3.745808065472145e-07,
      "loss": 0.82542968,
      "num_input_tokens_seen": 144977700,
      "step": 6719,
      "time_per_iteration": 2.5769529342651367
    },
    {
      "auxiliary_loss_clip": 0.01094426,
      "auxiliary_loss_mlp": 0.01028157,
      "balance_loss_clip": 1.04121172,
      "balance_loss_mlp": 1.02037895,
      "epoch": 0.8080322250946913,
      "flos": 23621787970560.0,
      "grad_norm": 1.5925608380800156,
      "language_loss": 0.76643407,
      "learning_rate": 3.741270472501994e-07,
      "loss": 0.78765988,
      "num_input_tokens_seen": 144998340,
      "step": 6720,
      "time_per_iteration": 3.7050259113311768
    },
    {
      "auxiliary_loss_clip": 0.0112816,
      "auxiliary_loss_mlp": 0.01033184,
      "balance_loss_clip": 1.04707551,
      "balance_loss_mlp": 1.02565861,
      "epoch": 0.8081524679853304,
      "flos": 22820692896000.0,
      "grad_norm": 1.9551024678249143,
      "language_loss": 0.73251224,
      "learning_rate": 3.736735345968183e-07,
      "loss": 0.75412565,
      "num_input_tokens_seen": 145017950,
      "step": 6721,
      "time_per_iteration": 2.7365498542785645
    },
    {
      "auxiliary_loss_clip": 0.01159833,
      "auxiliary_loss_mlp": 0.01028103,
      "balance_loss_clip": 1.05303979,
      "balance_loss_mlp": 1.0205605,
      "epoch": 0.8082727108759694,
      "flos": 17640004343040.0,
      "grad_norm": 1.8321050942255774,
      "language_loss": 0.79250014,
      "learning_rate": 3.7322026865586986e-07,
      "loss": 0.81437945,
      "num_input_tokens_seen": 145036985,
      "step": 6722,
      "time_per_iteration": 2.614715576171875
    },
    {
      "auxiliary_loss_clip": 0.01152205,
      "auxiliary_loss_mlp": 0.01026993,
      "balance_loss_clip": 1.05326092,
      "balance_loss_mlp": 1.01912546,
      "epoch": 0.8083929537666086,
      "flos": 25958172113280.0,
      "grad_norm": 1.914014238578193,
      "language_loss": 0.73489749,
      "learning_rate": 3.7276724949611206e-07,
      "loss": 0.75668949,
      "num_input_tokens_seen": 145057095,
      "step": 6723,
      "time_per_iteration": 3.6694912910461426
    },
    {
      "auxiliary_loss_clip": 0.01169066,
      "auxiliary_loss_mlp": 0.00762609,
      "balance_loss_clip": 1.05216146,
      "balance_loss_mlp": 1.00038755,
      "epoch": 0.8085131966572476,
      "flos": 27089178629760.0,
      "grad_norm": 4.160353547120726,
      "language_loss": 0.75198328,
      "learning_rate": 3.723144771862694e-07,
      "loss": 0.77130002,
      "num_input_tokens_seen": 145077735,
      "step": 6724,
      "time_per_iteration": 2.660292387008667
    },
    {
      "auxiliary_loss_clip": 0.01163643,
      "auxiliary_loss_mlp": 0.01028152,
      "balance_loss_clip": 1.05091643,
      "balance_loss_mlp": 1.02016544,
      "epoch": 0.8086334395478867,
      "flos": 23988543788160.0,
      "grad_norm": 1.7612662499941965,
      "language_loss": 0.77205813,
      "learning_rate": 3.718619517950263e-07,
      "loss": 0.79397607,
      "num_input_tokens_seen": 145098330,
      "step": 6725,
      "time_per_iteration": 3.040070056915283
    },
    {
      "auxiliary_loss_clip": 0.01145879,
      "auxiliary_loss_mlp": 0.01031251,
      "balance_loss_clip": 1.04816163,
      "balance_loss_mlp": 1.02399743,
      "epoch": 0.8087536824385259,
      "flos": 20405879406720.0,
      "grad_norm": 1.9451475301308379,
      "language_loss": 0.7682609,
      "learning_rate": 3.714096733910301e-07,
      "loss": 0.79003221,
      "num_input_tokens_seen": 145115855,
      "step": 6726,
      "time_per_iteration": 3.715348720550537
    },
    {
      "auxiliary_loss_clip": 0.01124115,
      "auxiliary_loss_mlp": 0.01030198,
      "balance_loss_clip": 1.04535758,
      "balance_loss_mlp": 1.02169299,
      "epoch": 0.8088739253291649,
      "flos": 25919639798400.0,
      "grad_norm": 7.745461333803276,
      "language_loss": 0.70318067,
      "learning_rate": 3.709576420428926e-07,
      "loss": 0.72472382,
      "num_input_tokens_seen": 145136655,
      "step": 6727,
      "time_per_iteration": 2.879307746887207
    },
    {
      "auxiliary_loss_clip": 0.01162127,
      "auxiliary_loss_mlp": 0.0102648,
      "balance_loss_clip": 1.04889059,
      "balance_loss_mlp": 1.01871896,
      "epoch": 0.808994168219804,
      "flos": 28402072640640.0,
      "grad_norm": 2.415963828945825,
      "language_loss": 0.73424089,
      "learning_rate": 3.7050585781918463e-07,
      "loss": 0.756127,
      "num_input_tokens_seen": 145156955,
      "step": 6728,
      "time_per_iteration": 2.741364002227783
    },
    {
      "auxiliary_loss_clip": 0.01148167,
      "auxiliary_loss_mlp": 0.01030396,
      "balance_loss_clip": 1.05329156,
      "balance_loss_mlp": 1.02162802,
      "epoch": 0.8091144111104431,
      "flos": 17421056991360.0,
      "grad_norm": 2.3313499132311244,
      "language_loss": 0.68516439,
      "learning_rate": 3.700543207884428e-07,
      "loss": 0.70695007,
      "num_input_tokens_seen": 145173865,
      "step": 6729,
      "time_per_iteration": 2.677032232284546
    },
    {
      "auxiliary_loss_clip": 0.01137874,
      "auxiliary_loss_mlp": 0.01024723,
      "balance_loss_clip": 1.04247367,
      "balance_loss_mlp": 1.01672375,
      "epoch": 0.8092346540010822,
      "flos": 32153803361280.0,
      "grad_norm": 2.104148996472874,
      "language_loss": 0.70975894,
      "learning_rate": 3.6960303101916466e-07,
      "loss": 0.73138493,
      "num_input_tokens_seen": 145193780,
      "step": 6730,
      "time_per_iteration": 2.7477011680603027
    },
    {
      "auxiliary_loss_clip": 0.01032162,
      "auxiliary_loss_mlp": 0.00998952,
      "balance_loss_clip": 1.0162555,
      "balance_loss_mlp": 0.99787277,
      "epoch": 0.8093548968917212,
      "flos": 58035093390720.0,
      "grad_norm": 0.7377722848061805,
      "language_loss": 0.55427068,
      "learning_rate": 3.6915198857981047e-07,
      "loss": 0.5745818,
      "num_input_tokens_seen": 145258980,
      "step": 6731,
      "time_per_iteration": 3.3009395599365234
    },
    {
      "auxiliary_loss_clip": 0.01147303,
      "auxiliary_loss_mlp": 0.00762022,
      "balance_loss_clip": 1.04872298,
      "balance_loss_mlp": 1.00039601,
      "epoch": 0.8094751397823604,
      "flos": 27381599251200.0,
      "grad_norm": 1.7080696380871614,
      "language_loss": 0.6797381,
      "learning_rate": 3.687011935388027e-07,
      "loss": 0.69883132,
      "num_input_tokens_seen": 145281875,
      "step": 6732,
      "time_per_iteration": 2.737877368927002
    },
    {
      "auxiliary_loss_clip": 0.0114799,
      "auxiliary_loss_mlp": 0.0102362,
      "balance_loss_clip": 1.04860973,
      "balance_loss_mlp": 1.0160923,
      "epoch": 0.8095953826729995,
      "flos": 24061083304320.0,
      "grad_norm": 2.3064404250210995,
      "language_loss": 0.7286855,
      "learning_rate": 3.6825064596452646e-07,
      "loss": 0.75040162,
      "num_input_tokens_seen": 145302220,
      "step": 6733,
      "time_per_iteration": 3.623119354248047
    },
    {
      "auxiliary_loss_clip": 0.01146424,
      "auxiliary_loss_mlp": 0.01031385,
      "balance_loss_clip": 1.04529655,
      "balance_loss_mlp": 1.02416086,
      "epoch": 0.8097156255636385,
      "flos": 23951412103680.0,
      "grad_norm": 1.7559993474243885,
      "language_loss": 0.7064299,
      "learning_rate": 3.678003459253305e-07,
      "loss": 0.72820801,
      "num_input_tokens_seen": 145323070,
      "step": 6734,
      "time_per_iteration": 2.687450647354126
    },
    {
      "auxiliary_loss_clip": 0.01159037,
      "auxiliary_loss_mlp": 0.01028818,
      "balance_loss_clip": 1.04993427,
      "balance_loss_mlp": 1.02052045,
      "epoch": 0.8098358684542777,
      "flos": 21799142098560.0,
      "grad_norm": 2.4621767516739146,
      "language_loss": 0.7416234,
      "learning_rate": 3.673502934895236e-07,
      "loss": 0.76350194,
      "num_input_tokens_seen": 145342575,
      "step": 6735,
      "time_per_iteration": 2.7182204723358154
    },
    {
      "auxiliary_loss_clip": 0.01046253,
      "auxiliary_loss_mlp": 0.01003939,
      "balance_loss_clip": 1.00919187,
      "balance_loss_mlp": 1.0029496,
      "epoch": 0.8099561113449167,
      "flos": 68809515966720.0,
      "grad_norm": 0.7160778193517595,
      "language_loss": 0.57850987,
      "learning_rate": 3.669004887253802e-07,
      "loss": 0.59901178,
      "num_input_tokens_seen": 145408865,
      "step": 6736,
      "time_per_iteration": 3.368457317352295
    },
    {
      "auxiliary_loss_clip": 0.01127829,
      "auxiliary_loss_mlp": 0.0103049,
      "balance_loss_clip": 1.04837441,
      "balance_loss_mlp": 1.023072,
      "epoch": 0.8100763542355558,
      "flos": 23586056916480.0,
      "grad_norm": 1.6027833459911347,
      "language_loss": 0.78980076,
      "learning_rate": 3.664509317011335e-07,
      "loss": 0.81138396,
      "num_input_tokens_seen": 145429200,
      "step": 6737,
      "time_per_iteration": 2.737008571624756
    },
    {
      "auxiliary_loss_clip": 0.01163631,
      "auxiliary_loss_mlp": 0.01028901,
      "balance_loss_clip": 1.05262923,
      "balance_loss_mlp": 1.02079487,
      "epoch": 0.810196597126195,
      "flos": 31650408207360.0,
      "grad_norm": 2.1536602436973205,
      "language_loss": 0.74046278,
      "learning_rate": 3.6600162248498134e-07,
      "loss": 0.76238805,
      "num_input_tokens_seen": 145452830,
      "step": 6738,
      "time_per_iteration": 2.7785227298736572
    },
    {
      "auxiliary_loss_clip": 0.01140456,
      "auxiliary_loss_mlp": 0.01029029,
      "balance_loss_clip": 1.0464412,
      "balance_loss_mlp": 1.02220142,
      "epoch": 0.810316840016834,
      "flos": 24900459298560.0,
      "grad_norm": 1.993610814035135,
      "language_loss": 0.7672019,
      "learning_rate": 3.6555256114508426e-07,
      "loss": 0.7888968,
      "num_input_tokens_seen": 145472625,
      "step": 6739,
      "time_per_iteration": 2.7148778438568115
    },
    {
      "auxiliary_loss_clip": 0.01156973,
      "auxiliary_loss_mlp": 0.01033592,
      "balance_loss_clip": 1.04973602,
      "balance_loss_mlp": 1.02566493,
      "epoch": 0.8104370829074731,
      "flos": 27965003950080.0,
      "grad_norm": 1.9071838278932391,
      "language_loss": 0.7365694,
      "learning_rate": 3.651037477495642e-07,
      "loss": 0.75847507,
      "num_input_tokens_seen": 145494075,
      "step": 6740,
      "time_per_iteration": 2.7748327255249023
    },
    {
      "auxiliary_loss_clip": 0.01129071,
      "auxiliary_loss_mlp": 0.01032504,
      "balance_loss_clip": 1.04669881,
      "balance_loss_mlp": 1.02469003,
      "epoch": 0.8105573257981122,
      "flos": 24640752988800.0,
      "grad_norm": 1.8562124220628018,
      "language_loss": 0.68330741,
      "learning_rate": 3.6465518236650584e-07,
      "loss": 0.70492315,
      "num_input_tokens_seen": 145514220,
      "step": 6741,
      "time_per_iteration": 2.7524383068084717
    },
    {
      "auxiliary_loss_clip": 0.01139544,
      "auxiliary_loss_mlp": 0.01026887,
      "balance_loss_clip": 1.04944026,
      "balance_loss_mlp": 1.01982391,
      "epoch": 0.8106775686887513,
      "flos": 26358935132160.0,
      "grad_norm": 3.14187509021194,
      "language_loss": 0.78639227,
      "learning_rate": 3.642068650639558e-07,
      "loss": 0.80805659,
      "num_input_tokens_seen": 145533965,
      "step": 6742,
      "time_per_iteration": 2.783442735671997
    },
    {
      "auxiliary_loss_clip": 0.01119451,
      "auxiliary_loss_mlp": 0.00762074,
      "balance_loss_clip": 1.04075575,
      "balance_loss_mlp": 1.00041962,
      "epoch": 0.8107978115793903,
      "flos": 27271892136960.0,
      "grad_norm": 1.7714859434642911,
      "language_loss": 0.64669204,
      "learning_rate": 3.6375879590992334e-07,
      "loss": 0.66550726,
      "num_input_tokens_seen": 145554310,
      "step": 6743,
      "time_per_iteration": 2.8059234619140625
    },
    {
      "auxiliary_loss_clip": 0.01116822,
      "auxiliary_loss_mlp": 0.01030512,
      "balance_loss_clip": 1.04833555,
      "balance_loss_mlp": 1.02267396,
      "epoch": 0.8109180544700295,
      "flos": 24934322845440.0,
      "grad_norm": 1.8481969716263542,
      "language_loss": 0.81125331,
      "learning_rate": 3.6331097497238173e-07,
      "loss": 0.8327266,
      "num_input_tokens_seen": 145573755,
      "step": 6744,
      "time_per_iteration": 2.774693012237549
    },
    {
      "auxiliary_loss_clip": 0.01143479,
      "auxiliary_loss_mlp": 0.01030488,
      "balance_loss_clip": 1.04410005,
      "balance_loss_mlp": 1.02268291,
      "epoch": 0.8110382973606686,
      "flos": 21105383840640.0,
      "grad_norm": 1.9808038003069828,
      "language_loss": 0.79998791,
      "learning_rate": 3.628634023192627e-07,
      "loss": 0.82172763,
      "num_input_tokens_seen": 145594000,
      "step": 6745,
      "time_per_iteration": 2.7552359104156494
    },
    {
      "auxiliary_loss_clip": 0.01141731,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.04614747,
      "balance_loss_mlp": 1.01832426,
      "epoch": 0.8111585402513076,
      "flos": 15414081500160.0,
      "grad_norm": 2.5918112451375244,
      "language_loss": 0.7543503,
      "learning_rate": 3.624160780184644e-07,
      "loss": 0.77603126,
      "num_input_tokens_seen": 145611215,
      "step": 6746,
      "time_per_iteration": 3.596928834915161
    },
    {
      "auxiliary_loss_clip": 0.01131192,
      "auxiliary_loss_mlp": 0.01029926,
      "balance_loss_clip": 1.04978776,
      "balance_loss_mlp": 1.02153921,
      "epoch": 0.8112787831419467,
      "flos": 24095736950400.0,
      "grad_norm": 2.8132205366724548,
      "language_loss": 0.74731195,
      "learning_rate": 3.6196900213784496e-07,
      "loss": 0.76892304,
      "num_input_tokens_seen": 145630530,
      "step": 6747,
      "time_per_iteration": 2.72702956199646
    },
    {
      "auxiliary_loss_clip": 0.01146115,
      "auxiliary_loss_mlp": 0.01028386,
      "balance_loss_clip": 1.0473212,
      "balance_loss_mlp": 1.02119493,
      "epoch": 0.8113990260325858,
      "flos": 20483374999680.0,
      "grad_norm": 2.3451084312275534,
      "language_loss": 0.8683672,
      "learning_rate": 3.6152217474522527e-07,
      "loss": 0.89011222,
      "num_input_tokens_seen": 145647345,
      "step": 6748,
      "time_per_iteration": 3.62182354927063
    },
    {
      "auxiliary_loss_clip": 0.01144464,
      "auxiliary_loss_mlp": 0.00762037,
      "balance_loss_clip": 1.051682,
      "balance_loss_mlp": 1.00059366,
      "epoch": 0.8115192689232249,
      "flos": 24901141656960.0,
      "grad_norm": 4.006665458974298,
      "language_loss": 0.73144352,
      "learning_rate": 3.6107559590838975e-07,
      "loss": 0.75050861,
      "num_input_tokens_seen": 145666330,
      "step": 6749,
      "time_per_iteration": 2.7100648880004883
    },
    {
      "auxiliary_loss_clip": 0.01131558,
      "auxiliary_loss_mlp": 0.0103421,
      "balance_loss_clip": 1.05176735,
      "balance_loss_mlp": 1.02587152,
      "epoch": 0.811639511813864,
      "flos": 24057204635520.0,
      "grad_norm": 2.442777271565622,
      "language_loss": 0.66458404,
      "learning_rate": 3.606292656950822e-07,
      "loss": 0.68624175,
      "num_input_tokens_seen": 145684740,
      "step": 6750,
      "time_per_iteration": 2.751707077026367
    },
    {
      "auxiliary_loss_clip": 0.01174605,
      "auxiliary_loss_mlp": 0.01030551,
      "balance_loss_clip": 1.05063653,
      "balance_loss_mlp": 1.02237272,
      "epoch": 0.8117597547045031,
      "flos": 23185150243200.0,
      "grad_norm": 2.6333398030071007,
      "language_loss": 0.86814427,
      "learning_rate": 3.601831841730121e-07,
      "loss": 0.89019585,
      "num_input_tokens_seen": 145702660,
      "step": 6751,
      "time_per_iteration": 2.618496894836426
    },
    {
      "auxiliary_loss_clip": 0.0112255,
      "auxiliary_loss_mlp": 0.01026726,
      "balance_loss_clip": 1.04855347,
      "balance_loss_mlp": 1.01959753,
      "epoch": 0.8118799975951422,
      "flos": 23040250778880.0,
      "grad_norm": 1.6357644012024026,
      "language_loss": 0.72829998,
      "learning_rate": 3.5973735140984916e-07,
      "loss": 0.74979281,
      "num_input_tokens_seen": 145722830,
      "step": 6752,
      "time_per_iteration": 3.7077713012695312
    },
    {
      "auxiliary_loss_clip": 0.01097906,
      "auxiliary_loss_mlp": 0.01021029,
      "balance_loss_clip": 1.04126143,
      "balance_loss_mlp": 1.01401067,
      "epoch": 0.8120002404857812,
      "flos": 24639962889600.0,
      "grad_norm": 2.442139012435254,
      "language_loss": 0.79846358,
      "learning_rate": 3.5929176747322607e-07,
      "loss": 0.81965292,
      "num_input_tokens_seen": 145741935,
      "step": 6753,
      "time_per_iteration": 2.894561290740967
    },
    {
      "auxiliary_loss_clip": 0.01029235,
      "auxiliary_loss_mlp": 0.01003267,
      "balance_loss_clip": 1.01200891,
      "balance_loss_mlp": 1.0024147,
      "epoch": 0.8121204833764204,
      "flos": 57415742156160.0,
      "grad_norm": 0.8098162129842164,
      "language_loss": 0.5612672,
      "learning_rate": 3.588464324307372e-07,
      "loss": 0.58159214,
      "num_input_tokens_seen": 145805560,
      "step": 6754,
      "time_per_iteration": 3.320638656616211
    },
    {
      "auxiliary_loss_clip": 0.01114336,
      "auxiliary_loss_mlp": 0.01027897,
      "balance_loss_clip": 1.04585385,
      "balance_loss_mlp": 1.02001691,
      "epoch": 0.8122407262670595,
      "flos": 19464589549440.0,
      "grad_norm": 1.747775203904938,
      "language_loss": 0.75595808,
      "learning_rate": 3.584013463499391e-07,
      "loss": 0.77738047,
      "num_input_tokens_seen": 145824180,
      "step": 6755,
      "time_per_iteration": 2.7768282890319824
    },
    {
      "auxiliary_loss_clip": 0.01054659,
      "auxiliary_loss_mlp": 0.01001686,
      "balance_loss_clip": 1.00927329,
      "balance_loss_mlp": 1.00078619,
      "epoch": 0.8123609691576985,
      "flos": 56425325472000.0,
      "grad_norm": 0.7320271273077165,
      "language_loss": 0.64404583,
      "learning_rate": 3.579565092983521e-07,
      "loss": 0.66460931,
      "num_input_tokens_seen": 145885300,
      "step": 6756,
      "time_per_iteration": 3.1018919944763184
    },
    {
      "auxiliary_loss_clip": 0.01164387,
      "auxiliary_loss_mlp": 0.01028348,
      "balance_loss_clip": 1.0512625,
      "balance_loss_mlp": 1.0208497,
      "epoch": 0.8124812120483377,
      "flos": 20631973564800.0,
      "grad_norm": 2.0442863345606392,
      "language_loss": 0.84082186,
      "learning_rate": 3.575119213434565e-07,
      "loss": 0.86274922,
      "num_input_tokens_seen": 145903815,
      "step": 6757,
      "time_per_iteration": 2.644428014755249
    },
    {
      "auxiliary_loss_clip": 0.01091101,
      "auxiliary_loss_mlp": 0.01028495,
      "balance_loss_clip": 1.04429781,
      "balance_loss_mlp": 1.02068663,
      "epoch": 0.8126014549389767,
      "flos": 22492397566080.0,
      "grad_norm": 1.8625825935739482,
      "language_loss": 0.8175962,
      "learning_rate": 3.5706758255269765e-07,
      "loss": 0.83879209,
      "num_input_tokens_seen": 145922270,
      "step": 6758,
      "time_per_iteration": 2.9014151096343994
    },
    {
      "auxiliary_loss_clip": 0.01134605,
      "auxiliary_loss_mlp": 0.00762158,
      "balance_loss_clip": 1.04660869,
      "balance_loss_mlp": 1.00045681,
      "epoch": 0.8127216978296158,
      "flos": 23287961946240.0,
      "grad_norm": 2.2662149816347683,
      "language_loss": 0.69982338,
      "learning_rate": 3.566234929934795e-07,
      "loss": 0.71879101,
      "num_input_tokens_seen": 145941470,
      "step": 6759,
      "time_per_iteration": 3.9398512840270996
    },
    {
      "auxiliary_loss_clip": 0.01113665,
      "auxiliary_loss_mlp": 0.01027832,
      "balance_loss_clip": 1.04456127,
      "balance_loss_mlp": 1.020298,
      "epoch": 0.812841940720255,
      "flos": 25154994049920.0,
      "grad_norm": 1.4846304377055577,
      "language_loss": 0.7194823,
      "learning_rate": 3.561796527331706e-07,
      "loss": 0.7408973,
      "num_input_tokens_seen": 145963145,
      "step": 6760,
      "time_per_iteration": 2.8026974201202393
    },
    {
      "auxiliary_loss_clip": 0.01160173,
      "auxiliary_loss_mlp": 0.01025704,
      "balance_loss_clip": 1.05099964,
      "balance_loss_mlp": 1.01741946,
      "epoch": 0.812962183610894,
      "flos": 26648446752000.0,
      "grad_norm": 1.9721950759498819,
      "language_loss": 0.77951485,
      "learning_rate": 3.5573606183910163e-07,
      "loss": 0.8013736,
      "num_input_tokens_seen": 145983150,
      "step": 6761,
      "time_per_iteration": 2.665048122406006
    },
    {
      "auxiliary_loss_clip": 0.01119037,
      "auxiliary_loss_mlp": 0.0103001,
      "balance_loss_clip": 1.0438956,
      "balance_loss_mlp": 1.02196336,
      "epoch": 0.8130824265015331,
      "flos": 24966965329920.0,
      "grad_norm": 2.1858450654446346,
      "language_loss": 0.78906995,
      "learning_rate": 3.5529272037856493e-07,
      "loss": 0.81056035,
      "num_input_tokens_seen": 146001365,
      "step": 6762,
      "time_per_iteration": 2.7619540691375732
    },
    {
      "auxiliary_loss_clip": 0.0105319,
      "auxiliary_loss_mlp": 0.01001782,
      "balance_loss_clip": 1.009637,
      "balance_loss_mlp": 1.0009234,
      "epoch": 0.8132026693921722,
      "flos": 67622918175360.0,
      "grad_norm": 0.7062062569866518,
      "language_loss": 0.5376358,
      "learning_rate": 3.548496284188149e-07,
      "loss": 0.55818558,
      "num_input_tokens_seen": 146061570,
      "step": 6763,
      "time_per_iteration": 3.2831504344940186
    },
    {
      "auxiliary_loss_clip": 0.01149431,
      "auxiliary_loss_mlp": 0.01024253,
      "balance_loss_clip": 1.04914713,
      "balance_loss_mlp": 1.01658213,
      "epoch": 0.8133229122828113,
      "flos": 19495149045120.0,
      "grad_norm": 2.9680328518747494,
      "language_loss": 0.79027009,
      "learning_rate": 3.544067860270681e-07,
      "loss": 0.81200689,
      "num_input_tokens_seen": 146079145,
      "step": 6764,
      "time_per_iteration": 2.7082366943359375
    },
    {
      "auxiliary_loss_clip": 0.01163384,
      "auxiliary_loss_mlp": 0.01030322,
      "balance_loss_clip": 1.04823923,
      "balance_loss_mlp": 1.02176881,
      "epoch": 0.8134431551734503,
      "flos": 20668135582080.0,
      "grad_norm": 1.6417571450531514,
      "language_loss": 0.71427631,
      "learning_rate": 3.539641932705029e-07,
      "loss": 0.73621333,
      "num_input_tokens_seen": 146097625,
      "step": 6765,
      "time_per_iteration": 2.6941726207733154
    },
    {
      "auxiliary_loss_clip": 0.0116596,
      "auxiliary_loss_mlp": 0.01026949,
      "balance_loss_clip": 1.04947746,
      "balance_loss_mlp": 1.01827109,
      "epoch": 0.8135633980640895,
      "flos": 21507332008320.0,
      "grad_norm": 2.947618553098232,
      "language_loss": 0.77364212,
      "learning_rate": 3.53521850216262e-07,
      "loss": 0.79557121,
      "num_input_tokens_seen": 146117195,
      "step": 6766,
      "time_per_iteration": 2.6423492431640625
    },
    {
      "auxiliary_loss_clip": 0.01119311,
      "auxiliary_loss_mlp": 0.01023536,
      "balance_loss_clip": 1.04397893,
      "balance_loss_mlp": 1.01529908,
      "epoch": 0.8136836409547286,
      "flos": 20554442058240.0,
      "grad_norm": 1.8589439640496748,
      "language_loss": 0.7687999,
      "learning_rate": 3.530797569314461e-07,
      "loss": 0.79022837,
      "num_input_tokens_seen": 146136220,
      "step": 6767,
      "time_per_iteration": 2.795497417449951
    },
    {
      "auxiliary_loss_clip": 0.01163019,
      "auxiliary_loss_mlp": 0.01028402,
      "balance_loss_clip": 1.05269325,
      "balance_loss_mlp": 1.01982534,
      "epoch": 0.8138038838453676,
      "flos": 20299045380480.0,
      "grad_norm": 2.1509479038842,
      "language_loss": 0.77863377,
      "learning_rate": 3.5263791348312235e-07,
      "loss": 0.80054796,
      "num_input_tokens_seen": 146155415,
      "step": 6768,
      "time_per_iteration": 2.6513121128082275
    },
    {
      "auxiliary_loss_clip": 0.01127146,
      "auxiliary_loss_mlp": 0.01023492,
      "balance_loss_clip": 1.04241371,
      "balance_loss_mlp": 1.01563048,
      "epoch": 0.8139241267360068,
      "flos": 29789840551680.0,
      "grad_norm": 1.9353054052586367,
      "language_loss": 0.70654857,
      "learning_rate": 3.521963199383171e-07,
      "loss": 0.728055,
      "num_input_tokens_seen": 146178370,
      "step": 6769,
      "time_per_iteration": 2.7977027893066406
    },
    {
      "auxiliary_loss_clip": 0.01066861,
      "auxiliary_loss_mlp": 0.0102468,
      "balance_loss_clip": 1.03605545,
      "balance_loss_mlp": 1.01566839,
      "epoch": 0.8140443696266458,
      "flos": 19713270384000.0,
      "grad_norm": 2.5578987555416646,
      "language_loss": 0.76754117,
      "learning_rate": 3.517549763640197e-07,
      "loss": 0.78845668,
      "num_input_tokens_seen": 146196010,
      "step": 6770,
      "time_per_iteration": 2.8506057262420654
    },
    {
      "auxiliary_loss_clip": 0.01148367,
      "auxiliary_loss_mlp": 0.01029122,
      "balance_loss_clip": 1.051561,
      "balance_loss_mlp": 1.02105784,
      "epoch": 0.8141646125172849,
      "flos": 27160568910720.0,
      "grad_norm": 1.9095714791237495,
      "language_loss": 0.71275502,
      "learning_rate": 3.513138828271829e-07,
      "loss": 0.73452991,
      "num_input_tokens_seen": 146215880,
      "step": 6771,
      "time_per_iteration": 3.6501471996307373
    },
    {
      "auxiliary_loss_clip": 0.0110466,
      "auxiliary_loss_mlp": 0.0102352,
      "balance_loss_clip": 1.04289699,
      "balance_loss_mlp": 1.01602221,
      "epoch": 0.8142848554079241,
      "flos": 39673102700160.0,
      "grad_norm": 1.9306066332982483,
      "language_loss": 0.70286191,
      "learning_rate": 3.508730393947179e-07,
      "loss": 0.72414368,
      "num_input_tokens_seen": 146239135,
      "step": 6772,
      "time_per_iteration": 2.9424855709075928
    },
    {
      "auxiliary_loss_clip": 0.01159849,
      "auxiliary_loss_mlp": 0.01028428,
      "balance_loss_clip": 1.05113435,
      "balance_loss_mlp": 1.02046514,
      "epoch": 0.8144050982985631,
      "flos": 22237288197120.0,
      "grad_norm": 1.8656416579696635,
      "language_loss": 0.72249031,
      "learning_rate": 3.504324461335024e-07,
      "loss": 0.74437314,
      "num_input_tokens_seen": 146259245,
      "step": 6773,
      "time_per_iteration": 2.6285529136657715
    },
    {
      "auxiliary_loss_clip": 0.01109646,
      "auxiliary_loss_mlp": 0.00762494,
      "balance_loss_clip": 1.04433894,
      "balance_loss_mlp": 1.00049317,
      "epoch": 0.8145253411892022,
      "flos": 23038239617280.0,
      "grad_norm": 2.222425297692101,
      "language_loss": 0.88518834,
      "learning_rate": 3.499921031103732e-07,
      "loss": 0.90390974,
      "num_input_tokens_seen": 146280015,
      "step": 6774,
      "time_per_iteration": 3.790168285369873
    },
    {
      "auxiliary_loss_clip": 0.01134429,
      "auxiliary_loss_mlp": 0.01031027,
      "balance_loss_clip": 1.04984331,
      "balance_loss_mlp": 1.02329612,
      "epoch": 0.8146455840798413,
      "flos": 24827668387200.0,
      "grad_norm": 1.8411149103085587,
      "language_loss": 0.7861445,
      "learning_rate": 3.4955201039212987e-07,
      "loss": 0.80779904,
      "num_input_tokens_seen": 146300935,
      "step": 6775,
      "time_per_iteration": 2.7784340381622314
    },
    {
      "auxiliary_loss_clip": 0.01177695,
      "auxiliary_loss_mlp": 0.01030191,
      "balance_loss_clip": 1.05469775,
      "balance_loss_mlp": 1.02221,
      "epoch": 0.8147658269704804,
      "flos": 19974520978560.0,
      "grad_norm": 3.9514994818337046,
      "language_loss": 0.65807551,
      "learning_rate": 3.4911216804553465e-07,
      "loss": 0.68015438,
      "num_input_tokens_seen": 146319835,
      "step": 6776,
      "time_per_iteration": 2.621997594833374
    },
    {
      "auxiliary_loss_clip": 0.01168122,
      "auxiliary_loss_mlp": 0.00762832,
      "balance_loss_clip": 1.05128455,
      "balance_loss_mlp": 1.00049448,
      "epoch": 0.8148860698611194,
      "flos": 21178031097600.0,
      "grad_norm": 2.679901459343514,
      "language_loss": 0.70918769,
      "learning_rate": 3.4867257613731017e-07,
      "loss": 0.72849727,
      "num_input_tokens_seen": 146339030,
      "step": 6777,
      "time_per_iteration": 2.68222975730896
    },
    {
      "auxiliary_loss_clip": 0.01106163,
      "auxiliary_loss_mlp": 0.01029674,
      "balance_loss_clip": 1.0423547,
      "balance_loss_mlp": 1.02175212,
      "epoch": 0.8150063127517585,
      "flos": 19606903234560.0,
      "grad_norm": 2.449143994181143,
      "language_loss": 0.8593992,
      "learning_rate": 3.4823323473414343e-07,
      "loss": 0.88075757,
      "num_input_tokens_seen": 146358550,
      "step": 6778,
      "time_per_iteration": 3.676833152770996
    },
    {
      "auxiliary_loss_clip": 0.01162194,
      "auxiliary_loss_mlp": 0.00762375,
      "balance_loss_clip": 1.04945326,
      "balance_loss_mlp": 1.00048292,
      "epoch": 0.8151265556423977,
      "flos": 22638374438400.0,
      "grad_norm": 2.3040332013373406,
      "language_loss": 0.76304424,
      "learning_rate": 3.477941439026812e-07,
      "loss": 0.78228986,
      "num_input_tokens_seen": 146376770,
      "step": 6779,
      "time_per_iteration": 2.6925604343414307
    },
    {
      "auxiliary_loss_clip": 0.01152921,
      "auxiliary_loss_mlp": 0.01031583,
      "balance_loss_clip": 1.0513525,
      "balance_loss_mlp": 1.02431381,
      "epoch": 0.8152467985330367,
      "flos": 17968048277760.0,
      "grad_norm": 1.8741304382781003,
      "language_loss": 0.7333048,
      "learning_rate": 3.473553037095349e-07,
      "loss": 0.75514978,
      "num_input_tokens_seen": 146395795,
      "step": 6780,
      "time_per_iteration": 2.6383180618286133
    },
    {
      "auxiliary_loss_clip": 0.01124496,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.04590487,
      "balance_loss_mlp": 1.01761091,
      "epoch": 0.8153670414236758,
      "flos": 24969012405120.0,
      "grad_norm": 2.1859937878221976,
      "language_loss": 0.8348977,
      "learning_rate": 3.469167142212743e-07,
      "loss": 0.85639292,
      "num_input_tokens_seen": 146417640,
      "step": 6781,
      "time_per_iteration": 2.794163942337036
    },
    {
      "auxiliary_loss_clip": 0.01165035,
      "auxiliary_loss_mlp": 0.0102826,
      "balance_loss_clip": 1.05017495,
      "balance_loss_mlp": 1.02017498,
      "epoch": 0.8154872843143149,
      "flos": 31066069754880.0,
      "grad_norm": 2.8842973581334808,
      "language_loss": 0.63283741,
      "learning_rate": 3.4647837550443337e-07,
      "loss": 0.65477037,
      "num_input_tokens_seen": 146436205,
      "step": 6782,
      "time_per_iteration": 2.7050445079803467
    },
    {
      "auxiliary_loss_clip": 0.01142887,
      "auxiliary_loss_mlp": 0.01028369,
      "balance_loss_clip": 1.04873586,
      "balance_loss_mlp": 1.01995564,
      "epoch": 0.815607527204954,
      "flos": 19391654983680.0,
      "grad_norm": 1.9272948754412162,
      "language_loss": 0.74371201,
      "learning_rate": 3.460402876255086e-07,
      "loss": 0.76542461,
      "num_input_tokens_seen": 146453595,
      "step": 6783,
      "time_per_iteration": 2.6668975353240967
    },
    {
      "auxiliary_loss_clip": 0.01137883,
      "auxiliary_loss_mlp": 0.0102937,
      "balance_loss_clip": 1.04650748,
      "balance_loss_mlp": 1.02159715,
      "epoch": 0.815727770095593,
      "flos": 26140418743680.0,
      "grad_norm": 6.239969121485881,
      "language_loss": 0.71809912,
      "learning_rate": 3.456024506509574e-07,
      "loss": 0.7397716,
      "num_input_tokens_seen": 146474515,
      "step": 6784,
      "time_per_iteration": 3.7921879291534424
    },
    {
      "auxiliary_loss_clip": 0.01113763,
      "auxiliary_loss_mlp": 0.01030837,
      "balance_loss_clip": 1.04858553,
      "balance_loss_mlp": 1.02315426,
      "epoch": 0.8158480129862322,
      "flos": 25337527989120.0,
      "grad_norm": 1.538735882847895,
      "language_loss": 0.73938346,
      "learning_rate": 3.4516486464719873e-07,
      "loss": 0.76082945,
      "num_input_tokens_seen": 146493905,
      "step": 6785,
      "time_per_iteration": 2.795987129211426
    },
    {
      "auxiliary_loss_clip": 0.011493,
      "auxiliary_loss_mlp": 0.01025824,
      "balance_loss_clip": 1.04597461,
      "balance_loss_mlp": 1.01807547,
      "epoch": 0.8159682558768713,
      "flos": 34423645559040.0,
      "grad_norm": 1.711976738427416,
      "language_loss": 0.61993635,
      "learning_rate": 3.4472752968061445e-07,
      "loss": 0.64168757,
      "num_input_tokens_seen": 146518335,
      "step": 6786,
      "time_per_iteration": 2.801778793334961
    },
    {
      "auxiliary_loss_clip": 0.01108462,
      "auxiliary_loss_mlp": 0.01026655,
      "balance_loss_clip": 1.04280484,
      "balance_loss_mlp": 1.01916909,
      "epoch": 0.8160884987675103,
      "flos": 18653223185280.0,
      "grad_norm": 2.286876228287801,
      "language_loss": 0.73899651,
      "learning_rate": 3.442904458175475e-07,
      "loss": 0.7603476,
      "num_input_tokens_seen": 146535655,
      "step": 6787,
      "time_per_iteration": 2.7389256954193115
    },
    {
      "auxiliary_loss_clip": 0.01120872,
      "auxiliary_loss_mlp": 0.01027351,
      "balance_loss_clip": 1.04339361,
      "balance_loss_mlp": 1.01909292,
      "epoch": 0.8162087416581495,
      "flos": 31430527102080.0,
      "grad_norm": 3.1090619801180406,
      "language_loss": 0.76384521,
      "learning_rate": 3.438536131243044e-07,
      "loss": 0.78532749,
      "num_input_tokens_seen": 146556815,
      "step": 6788,
      "time_per_iteration": 2.856572389602661
    },
    {
      "auxiliary_loss_clip": 0.01093883,
      "auxiliary_loss_mlp": 0.01025063,
      "balance_loss_clip": 1.04275179,
      "balance_loss_mlp": 1.01631904,
      "epoch": 0.8163289845487885,
      "flos": 37593910915200.0,
      "grad_norm": 2.8124114930344244,
      "language_loss": 0.62317979,
      "learning_rate": 3.434170316671503e-07,
      "loss": 0.64436924,
      "num_input_tokens_seen": 146581845,
      "step": 6789,
      "time_per_iteration": 2.9699950218200684
    },
    {
      "auxiliary_loss_clip": 0.01092519,
      "auxiliary_loss_mlp": 0.0103673,
      "balance_loss_clip": 1.04275286,
      "balance_loss_mlp": 1.02888608,
      "epoch": 0.8164492274394276,
      "flos": 13953989554560.0,
      "grad_norm": 2.3174916469510247,
      "language_loss": 0.90170527,
      "learning_rate": 3.4298070151231583e-07,
      "loss": 0.92299783,
      "num_input_tokens_seen": 146597245,
      "step": 6790,
      "time_per_iteration": 2.81587815284729
    },
    {
      "auxiliary_loss_clip": 0.01124614,
      "auxiliary_loss_mlp": 0.01029502,
      "balance_loss_clip": 1.04361498,
      "balance_loss_mlp": 1.02174735,
      "epoch": 0.8165694703300668,
      "flos": 28986554747520.0,
      "grad_norm": 2.119469023892629,
      "language_loss": 0.59919035,
      "learning_rate": 3.425446227259916e-07,
      "loss": 0.62073147,
      "num_input_tokens_seen": 146618210,
      "step": 6791,
      "time_per_iteration": 2.8491883277893066
    },
    {
      "auxiliary_loss_clip": 0.01162056,
      "auxiliary_loss_mlp": 0.01025958,
      "balance_loss_clip": 1.05006826,
      "balance_loss_mlp": 1.01856744,
      "epoch": 0.8166897132207058,
      "flos": 25118365155840.0,
      "grad_norm": 1.906075490898517,
      "language_loss": 0.82330716,
      "learning_rate": 3.421087953743296e-07,
      "loss": 0.84518725,
      "num_input_tokens_seen": 146637975,
      "step": 6792,
      "time_per_iteration": 2.6772844791412354
    },
    {
      "auxiliary_loss_clip": 0.01135484,
      "auxiliary_loss_mlp": 0.01023766,
      "balance_loss_clip": 1.04707408,
      "balance_loss_mlp": 1.01613331,
      "epoch": 0.8168099561113449,
      "flos": 23148593176320.0,
      "grad_norm": 1.9950457659312384,
      "language_loss": 0.80145842,
      "learning_rate": 3.416732195234464e-07,
      "loss": 0.82305086,
      "num_input_tokens_seen": 146658030,
      "step": 6793,
      "time_per_iteration": 2.721055746078491
    },
    {
      "auxiliary_loss_clip": 0.01132863,
      "auxiliary_loss_mlp": 0.00761781,
      "balance_loss_clip": 1.04805553,
      "balance_loss_mlp": 1.00051975,
      "epoch": 0.816930199001984,
      "flos": 18407666833920.0,
      "grad_norm": 1.5329475370600216,
      "language_loss": 0.79751724,
      "learning_rate": 3.4123789523941613e-07,
      "loss": 0.81646365,
      "num_input_tokens_seen": 146677855,
      "step": 6794,
      "time_per_iteration": 2.723867177963257
    },
    {
      "auxiliary_loss_clip": 0.01139131,
      "auxiliary_loss_mlp": 0.01029403,
      "balance_loss_clip": 1.04725266,
      "balance_loss_mlp": 1.02144575,
      "epoch": 0.8170504418926231,
      "flos": 21251324799360.0,
      "grad_norm": 2.3000592400912194,
      "language_loss": 0.6339227,
      "learning_rate": 3.4080282258827884e-07,
      "loss": 0.655608,
      "num_input_tokens_seen": 146696230,
      "step": 6795,
      "time_per_iteration": 2.7622509002685547
    },
    {
      "auxiliary_loss_clip": 0.01142616,
      "auxiliary_loss_mlp": 0.01027989,
      "balance_loss_clip": 1.04769671,
      "balance_loss_mlp": 1.02049971,
      "epoch": 0.8171706847832622,
      "flos": 19099234362240.0,
      "grad_norm": 3.205858766142367,
      "language_loss": 0.72409785,
      "learning_rate": 3.403680016360342e-07,
      "loss": 0.74580383,
      "num_input_tokens_seen": 146714835,
      "step": 6796,
      "time_per_iteration": 2.7111659049987793
    },
    {
      "auxiliary_loss_clip": 0.01129499,
      "auxiliary_loss_mlp": 0.01029995,
      "balance_loss_clip": 1.04962921,
      "balance_loss_mlp": 1.0220263,
      "epoch": 0.8172909276739013,
      "flos": 21470128496640.0,
      "grad_norm": 1.6344306685924603,
      "language_loss": 0.67797351,
      "learning_rate": 3.3993343244864403e-07,
      "loss": 0.69956839,
      "num_input_tokens_seen": 146734425,
      "step": 6797,
      "time_per_iteration": 3.616443395614624
    },
    {
      "auxiliary_loss_clip": 0.01132865,
      "auxiliary_loss_mlp": 0.01028054,
      "balance_loss_clip": 1.04869139,
      "balance_loss_mlp": 1.01995957,
      "epoch": 0.8174111705645404,
      "flos": 27599792417280.0,
      "grad_norm": 3.8131216543867015,
      "language_loss": 0.73032165,
      "learning_rate": 3.394991150920323e-07,
      "loss": 0.75193077,
      "num_input_tokens_seen": 146757545,
      "step": 6798,
      "time_per_iteration": 2.808114528656006
    },
    {
      "auxiliary_loss_clip": 0.01166008,
      "auxiliary_loss_mlp": 0.01033013,
      "balance_loss_clip": 1.05308747,
      "balance_loss_mlp": 1.02410197,
      "epoch": 0.8175314134551794,
      "flos": 14064594508800.0,
      "grad_norm": 2.159235819812832,
      "language_loss": 0.74350762,
      "learning_rate": 3.3906504963208396e-07,
      "loss": 0.7654978,
      "num_input_tokens_seen": 146774240,
      "step": 6799,
      "time_per_iteration": 2.5699613094329834
    },
    {
      "auxiliary_loss_clip": 0.01158608,
      "auxiliary_loss_mlp": 0.01030121,
      "balance_loss_clip": 1.04999948,
      "balance_loss_mlp": 1.02242863,
      "epoch": 0.8176516563458186,
      "flos": 22708076780160.0,
      "grad_norm": 1.9272626151605812,
      "language_loss": 0.66501784,
      "learning_rate": 3.3863123613464774e-07,
      "loss": 0.68690515,
      "num_input_tokens_seen": 146793140,
      "step": 6800,
      "time_per_iteration": 3.5874433517456055
    },
    {
      "auxiliary_loss_clip": 0.01140113,
      "auxiliary_loss_mlp": 0.0102981,
      "balance_loss_clip": 1.04690576,
      "balance_loss_mlp": 1.0225141,
      "epoch": 0.8177718992364577,
      "flos": 21945406279680.0,
      "grad_norm": 1.8251626865942299,
      "language_loss": 0.7499432,
      "learning_rate": 3.381976746655317e-07,
      "loss": 0.77164239,
      "num_input_tokens_seen": 146812895,
      "step": 6801,
      "time_per_iteration": 2.698291778564453
    },
    {
      "auxiliary_loss_clip": 0.01128613,
      "auxiliary_loss_mlp": 0.01026624,
      "balance_loss_clip": 1.04754114,
      "balance_loss_mlp": 1.01928663,
      "epoch": 0.8178921421270967,
      "flos": 22017443005440.0,
      "grad_norm": 2.066465082657757,
      "language_loss": 0.67371273,
      "learning_rate": 3.3776436529050756e-07,
      "loss": 0.69526511,
      "num_input_tokens_seen": 146832445,
      "step": 6802,
      "time_per_iteration": 2.681535005569458
    },
    {
      "auxiliary_loss_clip": 0.01172573,
      "auxiliary_loss_mlp": 0.01030279,
      "balance_loss_clip": 1.04933858,
      "balance_loss_mlp": 1.02229166,
      "epoch": 0.8180123850177359,
      "flos": 33183111496320.0,
      "grad_norm": 2.015495452005777,
      "language_loss": 0.72819507,
      "learning_rate": 3.373313080753073e-07,
      "loss": 0.75022358,
      "num_input_tokens_seen": 146856505,
      "step": 6803,
      "time_per_iteration": 2.7772810459136963
    },
    {
      "auxiliary_loss_clip": 0.01175077,
      "auxiliary_loss_mlp": 0.00762065,
      "balance_loss_clip": 1.05193543,
      "balance_loss_mlp": 1.00046325,
      "epoch": 0.8181326279083749,
      "flos": 22091167670400.0,
      "grad_norm": 2.798815984858419,
      "language_loss": 0.77830052,
      "learning_rate": 3.3689850308562527e-07,
      "loss": 0.79767191,
      "num_input_tokens_seen": 146876950,
      "step": 6804,
      "time_per_iteration": 3.579505443572998
    },
    {
      "auxiliary_loss_clip": 0.01146219,
      "auxiliary_loss_mlp": 0.01027772,
      "balance_loss_clip": 1.04816175,
      "balance_loss_mlp": 1.0207628,
      "epoch": 0.818252870799014,
      "flos": 15705747936000.0,
      "grad_norm": 2.2923819352286827,
      "language_loss": 0.7778579,
      "learning_rate": 3.364659503871183e-07,
      "loss": 0.7995978,
      "num_input_tokens_seen": 146894885,
      "step": 6805,
      "time_per_iteration": 2.687673807144165
    },
    {
      "auxiliary_loss_clip": 0.01127161,
      "auxiliary_loss_mlp": 0.01023481,
      "balance_loss_clip": 1.04196322,
      "balance_loss_mlp": 1.0166471,
      "epoch": 0.8183731136896532,
      "flos": 18770687637120.0,
      "grad_norm": 3.4221216448280063,
      "language_loss": 0.84229869,
      "learning_rate": 3.3603365004540417e-07,
      "loss": 0.86380517,
      "num_input_tokens_seen": 146913180,
      "step": 6806,
      "time_per_iteration": 2.7033963203430176
    },
    {
      "auxiliary_loss_clip": 0.01160226,
      "auxiliary_loss_mlp": 0.01025659,
      "balance_loss_clip": 1.05227149,
      "balance_loss_mlp": 1.01763284,
      "epoch": 0.8184933565802922,
      "flos": 26541792293760.0,
      "grad_norm": 1.9218272286733056,
      "language_loss": 0.77153623,
      "learning_rate": 3.356016021260624e-07,
      "loss": 0.79339504,
      "num_input_tokens_seen": 146933510,
      "step": 6807,
      "time_per_iteration": 2.6525566577911377
    },
    {
      "auxiliary_loss_clip": 0.01131948,
      "auxiliary_loss_mlp": 0.01029549,
      "balance_loss_clip": 1.04558039,
      "balance_loss_mlp": 1.02150846,
      "epoch": 0.8186135994709313,
      "flos": 17530117660800.0,
      "grad_norm": 2.6996383200397998,
      "language_loss": 0.66016853,
      "learning_rate": 3.35169806694634e-07,
      "loss": 0.68178356,
      "num_input_tokens_seen": 146951760,
      "step": 6808,
      "time_per_iteration": 2.7411394119262695
    },
    {
      "auxiliary_loss_clip": 0.01043388,
      "auxiliary_loss_mlp": 0.01001846,
      "balance_loss_clip": 1.02058768,
      "balance_loss_mlp": 1.00095236,
      "epoch": 0.8187338423615703,
      "flos": 63480300675840.0,
      "grad_norm": 0.722028764734722,
      "language_loss": 0.60577899,
      "learning_rate": 3.3473826381662186e-07,
      "loss": 0.62623137,
      "num_input_tokens_seen": 147022900,
      "step": 6809,
      "time_per_iteration": 3.3387646675109863
    },
    {
      "auxiliary_loss_clip": 0.01160902,
      "auxiliary_loss_mlp": 0.01030096,
      "balance_loss_clip": 1.05201387,
      "balance_loss_mlp": 1.02271152,
      "epoch": 0.8188540852522095,
      "flos": 17529974006400.0,
      "grad_norm": 2.0390423728851483,
      "language_loss": 0.82046652,
      "learning_rate": 3.3430697355749216e-07,
      "loss": 0.84237647,
      "num_input_tokens_seen": 147040590,
      "step": 6810,
      "time_per_iteration": 3.6185452938079834
    },
    {
      "auxiliary_loss_clip": 0.01163824,
      "auxiliary_loss_mlp": 0.01031845,
      "balance_loss_clip": 1.05206418,
      "balance_loss_mlp": 1.02398348,
      "epoch": 0.8189743281428485,
      "flos": 14392530702720.0,
      "grad_norm": 2.052106115902923,
      "language_loss": 0.7538271,
      "learning_rate": 3.3387593598266907e-07,
      "loss": 0.77578378,
      "num_input_tokens_seen": 147057200,
      "step": 6811,
      "time_per_iteration": 2.6441454887390137
    },
    {
      "auxiliary_loss_clip": 0.01131444,
      "auxiliary_loss_mlp": 0.0102371,
      "balance_loss_clip": 1.04920185,
      "balance_loss_mlp": 1.0155797,
      "epoch": 0.8190945710334876,
      "flos": 25080479285760.0,
      "grad_norm": 3.476009439841939,
      "language_loss": 0.78116894,
      "learning_rate": 3.3344515115754225e-07,
      "loss": 0.80272055,
      "num_input_tokens_seen": 147076180,
      "step": 6812,
      "time_per_iteration": 2.7383594512939453
    },
    {
      "auxiliary_loss_clip": 0.01162486,
      "auxiliary_loss_mlp": 0.01031185,
      "balance_loss_clip": 1.05010271,
      "balance_loss_mlp": 1.02318645,
      "epoch": 0.8192148139241268,
      "flos": 21507152440320.0,
      "grad_norm": 2.11771546186193,
      "language_loss": 0.80047214,
      "learning_rate": 3.33014619147461e-07,
      "loss": 0.82240885,
      "num_input_tokens_seen": 147094205,
      "step": 6813,
      "time_per_iteration": 2.682941436767578
    },
    {
      "auxiliary_loss_clip": 0.01138609,
      "auxiliary_loss_mlp": 0.01030251,
      "balance_loss_clip": 1.05058885,
      "balance_loss_mlp": 1.02222824,
      "epoch": 0.8193350568147658,
      "flos": 23952166289280.0,
      "grad_norm": 2.2291659561890245,
      "language_loss": 0.71598011,
      "learning_rate": 3.325843400177362e-07,
      "loss": 0.73766863,
      "num_input_tokens_seen": 147115545,
      "step": 6814,
      "time_per_iteration": 2.734184741973877
    },
    {
      "auxiliary_loss_clip": 0.01135293,
      "auxiliary_loss_mlp": 0.00761781,
      "balance_loss_clip": 1.04771495,
      "balance_loss_mlp": 1.00057006,
      "epoch": 0.8194552997054049,
      "flos": 20559469962240.0,
      "grad_norm": 1.729967642935272,
      "language_loss": 0.732517,
      "learning_rate": 3.32154313833642e-07,
      "loss": 0.75148785,
      "num_input_tokens_seen": 147135700,
      "step": 6815,
      "time_per_iteration": 2.816411018371582
    },
    {
      "auxiliary_loss_clip": 0.01117827,
      "auxiliary_loss_mlp": 0.00762751,
      "balance_loss_clip": 1.04886103,
      "balance_loss_mlp": 1.00051522,
      "epoch": 0.819575542596044,
      "flos": 26031753123840.0,
      "grad_norm": 13.73007093155316,
      "language_loss": 0.59392774,
      "learning_rate": 3.3172454066041164e-07,
      "loss": 0.61273348,
      "num_input_tokens_seen": 147155205,
      "step": 6816,
      "time_per_iteration": 2.7698044776916504
    },
    {
      "auxiliary_loss_clip": 0.01098604,
      "auxiliary_loss_mlp": 0.01033531,
      "balance_loss_clip": 1.0419234,
      "balance_loss_mlp": 1.02630091,
      "epoch": 0.8196957854866831,
      "flos": 29096944220160.0,
      "grad_norm": 1.8643775359433743,
      "language_loss": 0.76230794,
      "learning_rate": 3.3129502056324234e-07,
      "loss": 0.7836293,
      "num_input_tokens_seen": 147176570,
      "step": 6817,
      "time_per_iteration": 2.8557090759277344
    },
    {
      "auxiliary_loss_clip": 0.0105131,
      "auxiliary_loss_mlp": 0.01001731,
      "balance_loss_clip": 1.00859976,
      "balance_loss_mlp": 1.00090206,
      "epoch": 0.8198160283773221,
      "flos": 69033631898880.0,
      "grad_norm": 0.7996638910691548,
      "language_loss": 0.59638631,
      "learning_rate": 3.3086575360729165e-07,
      "loss": 0.61691672,
      "num_input_tokens_seen": 147234105,
      "step": 6818,
      "time_per_iteration": 3.0817012786865234
    },
    {
      "auxiliary_loss_clip": 0.01086878,
      "auxiliary_loss_mlp": 0.01030506,
      "balance_loss_clip": 1.04005468,
      "balance_loss_mlp": 1.02217925,
      "epoch": 0.8199362712679613,
      "flos": 16618058496000.0,
      "grad_norm": 1.6991802383184995,
      "language_loss": 0.71410161,
      "learning_rate": 3.3043673985767906e-07,
      "loss": 0.73527545,
      "num_input_tokens_seen": 147253170,
      "step": 6819,
      "time_per_iteration": 2.7513654232025146
    },
    {
      "auxiliary_loss_clip": 0.0113298,
      "auxiliary_loss_mlp": 0.01029384,
      "balance_loss_clip": 1.04856431,
      "balance_loss_mlp": 1.02142668,
      "epoch": 0.8200565141586004,
      "flos": 21757664868480.0,
      "grad_norm": 1.6993371048569061,
      "language_loss": 0.78044105,
      "learning_rate": 3.3000797937948564e-07,
      "loss": 0.80206466,
      "num_input_tokens_seen": 147271465,
      "step": 6820,
      "time_per_iteration": 2.745821237564087
    },
    {
      "auxiliary_loss_clip": 0.01055552,
      "auxiliary_loss_mlp": 0.01003408,
      "balance_loss_clip": 1.00940776,
      "balance_loss_mlp": 1.002496,
      "epoch": 0.8201767570492394,
      "flos": 69807112392960.0,
      "grad_norm": 0.9469023589665979,
      "language_loss": 0.64911032,
      "learning_rate": 3.295794722377534e-07,
      "loss": 0.66969991,
      "num_input_tokens_seen": 147335070,
      "step": 6821,
      "time_per_iteration": 3.2344260215759277
    },
    {
      "auxiliary_loss_clip": 0.01140394,
      "auxiliary_loss_mlp": 0.01030494,
      "balance_loss_clip": 1.04869139,
      "balance_loss_mlp": 1.02269828,
      "epoch": 0.8202969999398786,
      "flos": 23111892455040.0,
      "grad_norm": 1.6702377555296934,
      "language_loss": 0.80181825,
      "learning_rate": 3.291512184974876e-07,
      "loss": 0.8235271,
      "num_input_tokens_seen": 147355460,
      "step": 6822,
      "time_per_iteration": 2.7336418628692627
    },
    {
      "auxiliary_loss_clip": 0.01144179,
      "auxiliary_loss_mlp": 0.01030232,
      "balance_loss_clip": 1.04690957,
      "balance_loss_mlp": 1.02211952,
      "epoch": 0.8204172428305176,
      "flos": 28220616109440.0,
      "grad_norm": 2.939695051567114,
      "language_loss": 0.6672768,
      "learning_rate": 3.2872321822365346e-07,
      "loss": 0.68902093,
      "num_input_tokens_seen": 147375675,
      "step": 6823,
      "time_per_iteration": 2.7536439895629883
    },
    {
      "auxiliary_loss_clip": 0.01106997,
      "auxiliary_loss_mlp": 0.01029041,
      "balance_loss_clip": 1.04312027,
      "balance_loss_mlp": 1.0212357,
      "epoch": 0.8205374857211567,
      "flos": 20887011106560.0,
      "grad_norm": 2.6050661764062983,
      "language_loss": 0.7343787,
      "learning_rate": 3.282954714811783e-07,
      "loss": 0.75573897,
      "num_input_tokens_seen": 147394580,
      "step": 6824,
      "time_per_iteration": 3.778371572494507
    },
    {
      "auxiliary_loss_clip": 0.01148902,
      "auxiliary_loss_mlp": 0.01028003,
      "balance_loss_clip": 1.04903913,
      "balance_loss_mlp": 1.02005732,
      "epoch": 0.8206577286117959,
      "flos": 13152140294400.0,
      "grad_norm": 2.0107538756034713,
      "language_loss": 0.7059564,
      "learning_rate": 3.2786797833495093e-07,
      "loss": 0.72772551,
      "num_input_tokens_seen": 147409935,
      "step": 6825,
      "time_per_iteration": 2.70884370803833
    },
    {
      "auxiliary_loss_clip": 0.01143652,
      "auxiliary_loss_mlp": 0.01027823,
      "balance_loss_clip": 1.05039394,
      "balance_loss_mlp": 1.02054238,
      "epoch": 0.8207779715024349,
      "flos": 25265634917760.0,
      "grad_norm": 2.1246674481786387,
      "language_loss": 0.72663873,
      "learning_rate": 3.274407388498213e-07,
      "loss": 0.74835342,
      "num_input_tokens_seen": 147428065,
      "step": 6826,
      "time_per_iteration": 3.694094657897949
    },
    {
      "auxiliary_loss_clip": 0.0116069,
      "auxiliary_loss_mlp": 0.00761617,
      "balance_loss_clip": 1.05122542,
      "balance_loss_mlp": 1.00049973,
      "epoch": 0.820898214393074,
      "flos": 19610243199360.0,
      "grad_norm": 1.8767160076140306,
      "language_loss": 0.74520862,
      "learning_rate": 3.270137530906021e-07,
      "loss": 0.76443172,
      "num_input_tokens_seen": 147447300,
      "step": 6827,
      "time_per_iteration": 2.6933367252349854
    },
    {
      "auxiliary_loss_clip": 0.01110546,
      "auxiliary_loss_mlp": 0.01024976,
      "balance_loss_clip": 1.04284346,
      "balance_loss_mlp": 1.01769495,
      "epoch": 0.8210184572837131,
      "flos": 15596615439360.0,
      "grad_norm": 2.6781493581370537,
      "language_loss": 0.8344754,
      "learning_rate": 3.265870211220665e-07,
      "loss": 0.85583055,
      "num_input_tokens_seen": 147465135,
      "step": 6828,
      "time_per_iteration": 2.7015445232391357
    },
    {
      "auxiliary_loss_clip": 0.01133752,
      "auxiliary_loss_mlp": 0.01035496,
      "balance_loss_clip": 1.04854584,
      "balance_loss_mlp": 1.02721739,
      "epoch": 0.8211387001743522,
      "flos": 20813932886400.0,
      "grad_norm": 2.036079895755093,
      "language_loss": 0.81954557,
      "learning_rate": 3.2616054300894934e-07,
      "loss": 0.84123808,
      "num_input_tokens_seen": 147484585,
      "step": 6829,
      "time_per_iteration": 3.803152322769165
    },
    {
      "auxiliary_loss_clip": 0.01157711,
      "auxiliary_loss_mlp": 0.01036394,
      "balance_loss_clip": 1.04852915,
      "balance_loss_mlp": 1.02825832,
      "epoch": 0.8212589430649913,
      "flos": 27704579368320.0,
      "grad_norm": 2.3267782803032206,
      "language_loss": 0.84701848,
      "learning_rate": 3.2573431881594693e-07,
      "loss": 0.86895949,
      "num_input_tokens_seen": 147504130,
      "step": 6830,
      "time_per_iteration": 2.7243735790252686
    },
    {
      "auxiliary_loss_clip": 0.01117767,
      "auxiliary_loss_mlp": 0.01025769,
      "balance_loss_clip": 1.0434469,
      "balance_loss_mlp": 1.01760316,
      "epoch": 0.8213791859556304,
      "flos": 22455625017600.0,
      "grad_norm": 3.4583119238813103,
      "language_loss": 0.65930855,
      "learning_rate": 3.2530834860771663e-07,
      "loss": 0.68074393,
      "num_input_tokens_seen": 147523510,
      "step": 6831,
      "time_per_iteration": 2.7343387603759766
    },
    {
      "auxiliary_loss_clip": 0.01145186,
      "auxiliary_loss_mlp": 0.01026671,
      "balance_loss_clip": 1.04989052,
      "balance_loss_mlp": 1.01888621,
      "epoch": 0.8214994288462695,
      "flos": 16654471908480.0,
      "grad_norm": 2.0504049580505073,
      "language_loss": 0.74005246,
      "learning_rate": 3.248826324488794e-07,
      "loss": 0.76177102,
      "num_input_tokens_seen": 147540805,
      "step": 6832,
      "time_per_iteration": 2.7120494842529297
    },
    {
      "auxiliary_loss_clip": 0.01176575,
      "auxiliary_loss_mlp": 0.0103094,
      "balance_loss_clip": 1.05527616,
      "balance_loss_mlp": 1.02342939,
      "epoch": 0.8216196717369085,
      "flos": 25221787390080.0,
      "grad_norm": 1.908820917060888,
      "language_loss": 0.87987375,
      "learning_rate": 3.244571704040138e-07,
      "loss": 0.90194893,
      "num_input_tokens_seen": 147560965,
      "step": 6833,
      "time_per_iteration": 2.6280388832092285
    },
    {
      "auxiliary_loss_clip": 0.01175385,
      "auxiliary_loss_mlp": 0.01022223,
      "balance_loss_clip": 1.05023766,
      "balance_loss_mlp": 1.01423573,
      "epoch": 0.8217399146275477,
      "flos": 25371930240000.0,
      "grad_norm": 2.8659481073273816,
      "language_loss": 0.74058759,
      "learning_rate": 3.2403196253766374e-07,
      "loss": 0.76256371,
      "num_input_tokens_seen": 147580045,
      "step": 6834,
      "time_per_iteration": 2.6896488666534424
    },
    {
      "auxiliary_loss_clip": 0.01115484,
      "auxiliary_loss_mlp": 0.01031316,
      "balance_loss_clip": 1.04928124,
      "balance_loss_mlp": 1.02282846,
      "epoch": 0.8218601575181868,
      "flos": 25629625388160.0,
      "grad_norm": 2.5376029941322535,
      "language_loss": 0.793033,
      "learning_rate": 3.2360700891433254e-07,
      "loss": 0.81450093,
      "num_input_tokens_seen": 147599070,
      "step": 6835,
      "time_per_iteration": 2.775149345397949
    },
    {
      "auxiliary_loss_clip": 0.01052729,
      "auxiliary_loss_mlp": 0.01003601,
      "balance_loss_clip": 1.00957501,
      "balance_loss_mlp": 1.00269532,
      "epoch": 0.8219804004088258,
      "flos": 67660229427840.0,
      "grad_norm": 0.7983602774553088,
      "language_loss": 0.57203197,
      "learning_rate": 3.231823095984847e-07,
      "loss": 0.59259534,
      "num_input_tokens_seen": 147653710,
      "step": 6836,
      "time_per_iteration": 3.1621763706207275
    },
    {
      "auxiliary_loss_clip": 0.01125855,
      "auxiliary_loss_mlp": 0.00761667,
      "balance_loss_clip": 1.04786706,
      "balance_loss_mlp": 1.00059414,
      "epoch": 0.822100643299465,
      "flos": 19464266327040.0,
      "grad_norm": 2.1903311656078492,
      "language_loss": 0.76087737,
      "learning_rate": 3.2275786465454814e-07,
      "loss": 0.77975261,
      "num_input_tokens_seen": 147670360,
      "step": 6837,
      "time_per_iteration": 3.671381711959839
    },
    {
      "auxiliary_loss_clip": 0.01145862,
      "auxiliary_loss_mlp": 0.0102821,
      "balance_loss_clip": 1.04839206,
      "balance_loss_mlp": 1.02044344,
      "epoch": 0.822220886190104,
      "flos": 24681368292480.0,
      "grad_norm": 1.9054758708674613,
      "language_loss": 0.7516892,
      "learning_rate": 3.2233367414690917e-07,
      "loss": 0.77342987,
      "num_input_tokens_seen": 147692550,
      "step": 6838,
      "time_per_iteration": 2.69050931930542
    },
    {
      "auxiliary_loss_clip": 0.01140688,
      "auxiliary_loss_mlp": 0.01021455,
      "balance_loss_clip": 1.04833651,
      "balance_loss_mlp": 1.01407588,
      "epoch": 0.8223411290807431,
      "flos": 27819062991360.0,
      "grad_norm": 4.254653409355822,
      "language_loss": 0.8521499,
      "learning_rate": 3.219097381399183e-07,
      "loss": 0.87377131,
      "num_input_tokens_seen": 147709725,
      "step": 6839,
      "time_per_iteration": 2.777493953704834
    },
    {
      "auxiliary_loss_clip": 0.01095179,
      "auxiliary_loss_mlp": 0.00762322,
      "balance_loss_clip": 1.04361343,
      "balance_loss_mlp": 1.00058174,
      "epoch": 0.8224613719713821,
      "flos": 23218546913280.0,
      "grad_norm": 2.8719591159193625,
      "language_loss": 0.81409359,
      "learning_rate": 3.2148605669788584e-07,
      "loss": 0.83266866,
      "num_input_tokens_seen": 147729615,
      "step": 6840,
      "time_per_iteration": 2.8359062671661377
    },
    {
      "auxiliary_loss_clip": 0.01111425,
      "auxiliary_loss_mlp": 0.01028499,
      "balance_loss_clip": 1.04365706,
      "balance_loss_mlp": 1.02038121,
      "epoch": 0.8225816148620213,
      "flos": 15706250726400.0,
      "grad_norm": 3.177346453850323,
      "language_loss": 0.77795631,
      "learning_rate": 3.2106262988508405e-07,
      "loss": 0.79935551,
      "num_input_tokens_seen": 147747665,
      "step": 6841,
      "time_per_iteration": 2.815985918045044
    },
    {
      "auxiliary_loss_clip": 0.0107899,
      "auxiliary_loss_mlp": 0.01025932,
      "balance_loss_clip": 1.03479433,
      "balance_loss_mlp": 1.01786125,
      "epoch": 0.8227018577526604,
      "flos": 18515111391360.0,
      "grad_norm": 3.3478197533729412,
      "language_loss": 0.74383807,
      "learning_rate": 3.206394577657465e-07,
      "loss": 0.76488733,
      "num_input_tokens_seen": 147765445,
      "step": 6842,
      "time_per_iteration": 2.8952829837799072
    },
    {
      "auxiliary_loss_clip": 0.01180178,
      "auxiliary_loss_mlp": 0.01032853,
      "balance_loss_clip": 1.05284572,
      "balance_loss_mlp": 1.02415049,
      "epoch": 0.8228221006432994,
      "flos": 22236785406720.0,
      "grad_norm": 3.952702557405665,
      "language_loss": 0.72403091,
      "learning_rate": 3.202165404040675e-07,
      "loss": 0.74616122,
      "num_input_tokens_seen": 147783365,
      "step": 6843,
      "time_per_iteration": 2.8169987201690674
    },
    {
      "auxiliary_loss_clip": 0.01114636,
      "auxiliary_loss_mlp": 0.01027102,
      "balance_loss_clip": 1.04405284,
      "balance_loss_mlp": 1.01874554,
      "epoch": 0.8229423435339386,
      "flos": 24097532630400.0,
      "grad_norm": 2.330324052382038,
      "language_loss": 0.75012589,
      "learning_rate": 3.1979387786420396e-07,
      "loss": 0.77154326,
      "num_input_tokens_seen": 147803605,
      "step": 6844,
      "time_per_iteration": 2.8159091472625732
    },
    {
      "auxiliary_loss_clip": 0.01160033,
      "auxiliary_loss_mlp": 0.01028112,
      "balance_loss_clip": 1.05122626,
      "balance_loss_mlp": 1.02053952,
      "epoch": 0.8230625864245776,
      "flos": 23878549365120.0,
      "grad_norm": 1.8656721666797693,
      "language_loss": 0.82042813,
      "learning_rate": 3.1937147021027346e-07,
      "loss": 0.84230953,
      "num_input_tokens_seen": 147822060,
      "step": 6845,
      "time_per_iteration": 2.663738250732422
    },
    {
      "auxiliary_loss_clip": 0.01104469,
      "auxiliary_loss_mlp": 0.01026497,
      "balance_loss_clip": 1.0417614,
      "balance_loss_mlp": 1.01914513,
      "epoch": 0.8231828293152167,
      "flos": 16581106379520.0,
      "grad_norm": 2.511996357585965,
      "language_loss": 0.76909912,
      "learning_rate": 3.189493175063547e-07,
      "loss": 0.79040873,
      "num_input_tokens_seen": 147839295,
      "step": 6846,
      "time_per_iteration": 2.7465248107910156
    },
    {
      "auxiliary_loss_clip": 0.01148411,
      "auxiliary_loss_mlp": 0.00762115,
      "balance_loss_clip": 1.04872561,
      "balance_loss_mlp": 1.00040185,
      "epoch": 0.8233030722058559,
      "flos": 18880071528960.0,
      "grad_norm": 2.116796160508061,
      "language_loss": 0.67515457,
      "learning_rate": 3.1852741981648776e-07,
      "loss": 0.69425988,
      "num_input_tokens_seen": 147857945,
      "step": 6847,
      "time_per_iteration": 2.6319403648376465
    },
    {
      "auxiliary_loss_clip": 0.01139695,
      "auxiliary_loss_mlp": 0.01025101,
      "balance_loss_clip": 1.04958034,
      "balance_loss_mlp": 1.01724529,
      "epoch": 0.8234233150964949,
      "flos": 28439024757120.0,
      "grad_norm": 2.0718128516956553,
      "language_loss": 0.70133686,
      "learning_rate": 3.1810577720467404e-07,
      "loss": 0.72298479,
      "num_input_tokens_seen": 147879675,
      "step": 6848,
      "time_per_iteration": 2.8034701347351074
    },
    {
      "auxiliary_loss_clip": 0.01131418,
      "auxiliary_loss_mlp": 0.01026577,
      "balance_loss_clip": 1.04244196,
      "balance_loss_mlp": 1.01818419,
      "epoch": 0.823543557987134,
      "flos": 33765941577600.0,
      "grad_norm": 1.4987473601740917,
      "language_loss": 0.56882,
      "learning_rate": 3.176843897348769e-07,
      "loss": 0.59039992,
      "num_input_tokens_seen": 147902870,
      "step": 6849,
      "time_per_iteration": 2.842970371246338
    },
    {
      "auxiliary_loss_clip": 0.01116916,
      "auxiliary_loss_mlp": 0.01024247,
      "balance_loss_clip": 1.0453403,
      "balance_loss_mlp": 1.01654005,
      "epoch": 0.8236638008777731,
      "flos": 17092366611840.0,
      "grad_norm": 3.8102368964999225,
      "language_loss": 0.76004142,
      "learning_rate": 3.1726325747102034e-07,
      "loss": 0.78145307,
      "num_input_tokens_seen": 147921245,
      "step": 6850,
      "time_per_iteration": 3.7110698223114014
    },
    {
      "auxiliary_loss_clip": 0.01158387,
      "auxiliary_loss_mlp": 0.01025968,
      "balance_loss_clip": 1.05036056,
      "balance_loss_mlp": 1.01795721,
      "epoch": 0.8237840437684122,
      "flos": 61639982334720.0,
      "grad_norm": 1.5554461050519481,
      "language_loss": 0.64111555,
      "learning_rate": 3.1684238047698974e-07,
      "loss": 0.66295904,
      "num_input_tokens_seen": 147949515,
      "step": 6851,
      "time_per_iteration": 3.979942798614502
    },
    {
      "auxiliary_loss_clip": 0.01124634,
      "auxiliary_loss_mlp": 0.01030193,
      "balance_loss_clip": 1.04462838,
      "balance_loss_mlp": 1.02155018,
      "epoch": 0.8239042866590512,
      "flos": 27309023821440.0,
      "grad_norm": 2.4201679633132533,
      "language_loss": 0.53262269,
      "learning_rate": 3.1642175881663155e-07,
      "loss": 0.55417097,
      "num_input_tokens_seen": 147969245,
      "step": 6852,
      "time_per_iteration": 2.823596477508545
    },
    {
      "auxiliary_loss_clip": 0.01120433,
      "auxiliary_loss_mlp": 0.00761644,
      "balance_loss_clip": 1.04355669,
      "balance_loss_mlp": 1.00048327,
      "epoch": 0.8240245295496904,
      "flos": 21726351187200.0,
      "grad_norm": 2.090514431956174,
      "language_loss": 0.83657837,
      "learning_rate": 3.160013925537537e-07,
      "loss": 0.85539913,
      "num_input_tokens_seen": 147990080,
      "step": 6853,
      "time_per_iteration": 2.799957752227783
    },
    {
      "auxiliary_loss_clip": 0.01134854,
      "auxiliary_loss_mlp": 0.01025355,
      "balance_loss_clip": 1.04938769,
      "balance_loss_mlp": 1.01726675,
      "epoch": 0.8241447724403295,
      "flos": 20009318279040.0,
      "grad_norm": 2.3092182376067005,
      "language_loss": 0.75836927,
      "learning_rate": 3.155812817521266e-07,
      "loss": 0.77997136,
      "num_input_tokens_seen": 148010455,
      "step": 6854,
      "time_per_iteration": 2.786851167678833
    },
    {
      "auxiliary_loss_clip": 0.01145422,
      "auxiliary_loss_mlp": 0.01031195,
      "balance_loss_clip": 1.04774237,
      "balance_loss_mlp": 1.02302885,
      "epoch": 0.8242650153309685,
      "flos": 22272983337600.0,
      "grad_norm": 2.2366536309693523,
      "language_loss": 0.78259134,
      "learning_rate": 3.151614264754787e-07,
      "loss": 0.80435753,
      "num_input_tokens_seen": 148028400,
      "step": 6855,
      "time_per_iteration": 3.7093920707702637
    },
    {
      "auxiliary_loss_clip": 0.01132039,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.04297101,
      "balance_loss_mlp": 1.01922274,
      "epoch": 0.8243852582216077,
      "flos": 22309971367680.0,
      "grad_norm": 2.235376834163271,
      "language_loss": 0.79237384,
      "learning_rate": 3.147418267875035e-07,
      "loss": 0.81396449,
      "num_input_tokens_seen": 148046530,
      "step": 6856,
      "time_per_iteration": 2.7451841831207275
    },
    {
      "auxiliary_loss_clip": 0.01174979,
      "auxiliary_loss_mlp": 0.01023666,
      "balance_loss_clip": 1.05170178,
      "balance_loss_mlp": 1.01616788,
      "epoch": 0.8245055011122467,
      "flos": 24645421756800.0,
      "grad_norm": 2.3658936790249245,
      "language_loss": 0.65716314,
      "learning_rate": 3.1432248275185315e-07,
      "loss": 0.67914963,
      "num_input_tokens_seen": 148067040,
      "step": 6857,
      "time_per_iteration": 2.64955735206604
    },
    {
      "auxiliary_loss_clip": 0.01131908,
      "auxiliary_loss_mlp": 0.01027258,
      "balance_loss_clip": 1.04855537,
      "balance_loss_mlp": 1.0196228,
      "epoch": 0.8246257440028858,
      "flos": 17487275713920.0,
      "grad_norm": 2.2071924060775765,
      "language_loss": 0.77376485,
      "learning_rate": 3.139033944321412e-07,
      "loss": 0.79535651,
      "num_input_tokens_seen": 148084400,
      "step": 6858,
      "time_per_iteration": 2.7322309017181396
    },
    {
      "auxiliary_loss_clip": 0.01124991,
      "auxiliary_loss_mlp": 0.00761938,
      "balance_loss_clip": 1.04504728,
      "balance_loss_mlp": 1.00056434,
      "epoch": 0.824745986893525,
      "flos": 25010130499200.0,
      "grad_norm": 2.19351973155367,
      "language_loss": 0.7872175,
      "learning_rate": 3.1348456189194507e-07,
      "loss": 0.80608678,
      "num_input_tokens_seen": 148104860,
      "step": 6859,
      "time_per_iteration": 2.8566272258758545
    },
    {
      "auxiliary_loss_clip": 0.01115638,
      "auxiliary_loss_mlp": 0.01024714,
      "balance_loss_clip": 1.04417253,
      "balance_loss_mlp": 1.01691151,
      "epoch": 0.824866229784164,
      "flos": 18772698798720.0,
      "grad_norm": 2.6180694368144435,
      "language_loss": 0.83058804,
      "learning_rate": 3.1306598519479876e-07,
      "loss": 0.85199153,
      "num_input_tokens_seen": 148124680,
      "step": 6860,
      "time_per_iteration": 2.7655704021453857
    },
    {
      "auxiliary_loss_clip": 0.01150852,
      "auxiliary_loss_mlp": 0.01029066,
      "balance_loss_clip": 1.04887187,
      "balance_loss_mlp": 1.02145493,
      "epoch": 0.8249864726748031,
      "flos": 23842171866240.0,
      "grad_norm": 2.078086045148228,
      "language_loss": 0.78534508,
      "learning_rate": 3.1264766440420177e-07,
      "loss": 0.80714428,
      "num_input_tokens_seen": 148147150,
      "step": 6861,
      "time_per_iteration": 2.7563958168029785
    },
    {
      "auxiliary_loss_clip": 0.01160086,
      "auxiliary_loss_mlp": 0.0102368,
      "balance_loss_clip": 1.04829276,
      "balance_loss_mlp": 1.01608968,
      "epoch": 0.8251067155654422,
      "flos": 20303103617280.0,
      "grad_norm": 2.4988764763347198,
      "language_loss": 0.69594884,
      "learning_rate": 3.122295995836124e-07,
      "loss": 0.71778649,
      "num_input_tokens_seen": 148167020,
      "step": 6862,
      "time_per_iteration": 2.6412456035614014
    },
    {
      "auxiliary_loss_clip": 0.01146184,
      "auxiliary_loss_mlp": 0.01024592,
      "balance_loss_clip": 1.04690325,
      "balance_loss_mlp": 1.01659274,
      "epoch": 0.8252269584560813,
      "flos": 25009699536000.0,
      "grad_norm": 1.8153668880726892,
      "language_loss": 0.77746433,
      "learning_rate": 3.118117907964508e-07,
      "loss": 0.7991721,
      "num_input_tokens_seen": 148188965,
      "step": 6863,
      "time_per_iteration": 3.6750540733337402
    },
    {
      "auxiliary_loss_clip": 0.01145897,
      "auxiliary_loss_mlp": 0.00761958,
      "balance_loss_clip": 1.0513103,
      "balance_loss_mlp": 1.00052845,
      "epoch": 0.8253472013467203,
      "flos": 17128564542720.0,
      "grad_norm": 2.123103241747715,
      "language_loss": 0.80512166,
      "learning_rate": 3.1139423810609856e-07,
      "loss": 0.82420027,
      "num_input_tokens_seen": 148205660,
      "step": 6864,
      "time_per_iteration": 2.664984941482544
    },
    {
      "auxiliary_loss_clip": 0.01137943,
      "auxiliary_loss_mlp": 0.0102638,
      "balance_loss_clip": 1.0435369,
      "balance_loss_mlp": 1.01831591,
      "epoch": 0.8254674442373595,
      "flos": 22414794232320.0,
      "grad_norm": 3.6615049133554876,
      "language_loss": 0.75610292,
      "learning_rate": 3.1097694157589714e-07,
      "loss": 0.77774614,
      "num_input_tokens_seen": 148225545,
      "step": 6865,
      "time_per_iteration": 2.7679102420806885
    },
    {
      "auxiliary_loss_clip": 0.01140037,
      "auxiliary_loss_mlp": 0.01026354,
      "balance_loss_clip": 1.04733706,
      "balance_loss_mlp": 1.0186944,
      "epoch": 0.8255876871279986,
      "flos": 24786765774720.0,
      "grad_norm": 4.156145076648411,
      "language_loss": 0.75946259,
      "learning_rate": 3.105599012691511e-07,
      "loss": 0.78112644,
      "num_input_tokens_seen": 148243975,
      "step": 6866,
      "time_per_iteration": 2.751206159591675
    },
    {
      "auxiliary_loss_clip": 0.01127476,
      "auxiliary_loss_mlp": 0.010255,
      "balance_loss_clip": 1.04305744,
      "balance_loss_mlp": 1.01786733,
      "epoch": 0.8257079300186376,
      "flos": 27455431656960.0,
      "grad_norm": 1.5374519991679785,
      "language_loss": 0.8254047,
      "learning_rate": 3.101431172491249e-07,
      "loss": 0.84693444,
      "num_input_tokens_seen": 148265520,
      "step": 6867,
      "time_per_iteration": 2.754889488220215
    },
    {
      "auxiliary_loss_clip": 0.01175409,
      "auxiliary_loss_mlp": 0.01032343,
      "balance_loss_clip": 1.0510987,
      "balance_loss_mlp": 1.02404034,
      "epoch": 0.8258281729092768,
      "flos": 16471866142080.0,
      "grad_norm": 4.490522138554538,
      "language_loss": 0.72044587,
      "learning_rate": 3.097265895790444e-07,
      "loss": 0.74252343,
      "num_input_tokens_seen": 148283730,
      "step": 6868,
      "time_per_iteration": 2.6287591457366943
    },
    {
      "auxiliary_loss_clip": 0.01122767,
      "auxiliary_loss_mlp": 0.01031769,
      "balance_loss_clip": 1.04520273,
      "balance_loss_mlp": 1.02353191,
      "epoch": 0.8259484157999158,
      "flos": 21433822824960.0,
      "grad_norm": 2.0253423645622615,
      "language_loss": 0.83818829,
      "learning_rate": 3.093103183220962e-07,
      "loss": 0.85973364,
      "num_input_tokens_seen": 148303775,
      "step": 6869,
      "time_per_iteration": 2.7030324935913086
    },
    {
      "auxiliary_loss_clip": 0.01029475,
      "auxiliary_loss_mlp": 0.01000081,
      "balance_loss_clip": 1.01011801,
      "balance_loss_mlp": 0.99923468,
      "epoch": 0.8260686586905549,
      "flos": 58322342453760.0,
      "grad_norm": 0.8177593756175817,
      "language_loss": 0.59248227,
      "learning_rate": 3.0889430354142796e-07,
      "loss": 0.61277783,
      "num_input_tokens_seen": 148365285,
      "step": 6870,
      "time_per_iteration": 3.2700302600860596
    },
    {
      "auxiliary_loss_clip": 0.01129937,
      "auxiliary_loss_mlp": 0.01033033,
      "balance_loss_clip": 1.04166722,
      "balance_loss_mlp": 1.02454543,
      "epoch": 0.826188901581194,
      "flos": 27527288814720.0,
      "grad_norm": 2.429047704952141,
      "language_loss": 0.69841635,
      "learning_rate": 3.084785453001497e-07,
      "loss": 0.72004604,
      "num_input_tokens_seen": 148386200,
      "step": 6871,
      "time_per_iteration": 2.760702610015869
    },
    {
      "auxiliary_loss_clip": 0.01148907,
      "auxiliary_loss_mlp": 0.0103275,
      "balance_loss_clip": 1.05095029,
      "balance_loss_mlp": 1.02427459,
      "epoch": 0.8263091444718331,
      "flos": 23696051339520.0,
      "grad_norm": 2.217744748871906,
      "language_loss": 0.82178241,
      "learning_rate": 3.080630436613314e-07,
      "loss": 0.84359896,
      "num_input_tokens_seen": 148403970,
      "step": 6872,
      "time_per_iteration": 2.740736722946167
    },
    {
      "auxiliary_loss_clip": 0.01157741,
      "auxiliary_loss_mlp": 0.00762404,
      "balance_loss_clip": 1.05025995,
      "balance_loss_mlp": 1.00050306,
      "epoch": 0.8264293873624722,
      "flos": 17165157523200.0,
      "grad_norm": 2.406884065194521,
      "language_loss": 0.86121213,
      "learning_rate": 3.076477986880039e-07,
      "loss": 0.88041353,
      "num_input_tokens_seen": 148421765,
      "step": 6873,
      "time_per_iteration": 2.6609466075897217
    },
    {
      "auxiliary_loss_clip": 0.01097146,
      "auxiliary_loss_mlp": 0.00761832,
      "balance_loss_clip": 1.04148316,
      "balance_loss_mlp": 1.0004158,
      "epoch": 0.8265496302531112,
      "flos": 24098645952000.0,
      "grad_norm": 2.0952598778447946,
      "language_loss": 0.6926049,
      "learning_rate": 3.0723281044315986e-07,
      "loss": 0.71119463,
      "num_input_tokens_seen": 148443720,
      "step": 6874,
      "time_per_iteration": 2.7648072242736816
    },
    {
      "auxiliary_loss_clip": 0.01159514,
      "auxiliary_loss_mlp": 0.00761423,
      "balance_loss_clip": 1.04867029,
      "balance_loss_mlp": 1.00055933,
      "epoch": 0.8266698731437504,
      "flos": 14099894599680.0,
      "grad_norm": 2.0938320505216756,
      "language_loss": 0.76772964,
      "learning_rate": 3.068180789897521e-07,
      "loss": 0.78693902,
      "num_input_tokens_seen": 148462130,
      "step": 6875,
      "time_per_iteration": 2.7363815307617188
    },
    {
      "auxiliary_loss_clip": 0.01134273,
      "auxiliary_loss_mlp": 0.01027061,
      "balance_loss_clip": 1.04707015,
      "balance_loss_mlp": 1.01878834,
      "epoch": 0.8267901160343895,
      "flos": 30777563715840.0,
      "grad_norm": 1.6017167459314043,
      "language_loss": 0.81391627,
      "learning_rate": 3.064036043906966e-07,
      "loss": 0.83552957,
      "num_input_tokens_seen": 148485570,
      "step": 6876,
      "time_per_iteration": 3.724344491958618
    },
    {
      "auxiliary_loss_clip": 0.01163044,
      "auxiliary_loss_mlp": 0.01025301,
      "balance_loss_clip": 1.05312467,
      "balance_loss_mlp": 1.01672411,
      "epoch": 0.8269103589250285,
      "flos": 40624915242240.0,
      "grad_norm": 3.000389670042657,
      "language_loss": 0.6812287,
      "learning_rate": 3.059893867088668e-07,
      "loss": 0.70311219,
      "num_input_tokens_seen": 148509715,
      "step": 6877,
      "time_per_iteration": 3.814547061920166
    },
    {
      "auxiliary_loss_clip": 0.01080087,
      "auxiliary_loss_mlp": 0.01030566,
      "balance_loss_clip": 1.03736663,
      "balance_loss_mlp": 1.02275181,
      "epoch": 0.8270306018156677,
      "flos": 30263645877120.0,
      "grad_norm": 1.8082840276437941,
      "language_loss": 0.66641259,
      "learning_rate": 3.055754260071004e-07,
      "loss": 0.68751907,
      "num_input_tokens_seen": 148532010,
      "step": 6878,
      "time_per_iteration": 3.0426580905914307
    },
    {
      "auxiliary_loss_clip": 0.01169293,
      "auxiliary_loss_mlp": 0.01023859,
      "balance_loss_clip": 1.048087,
      "balance_loss_mlp": 1.01606822,
      "epoch": 0.8271508447063067,
      "flos": 25226599812480.0,
      "grad_norm": 2.359569421675569,
      "language_loss": 0.73673278,
      "learning_rate": 3.051617223481948e-07,
      "loss": 0.75866425,
      "num_input_tokens_seen": 148553330,
      "step": 6879,
      "time_per_iteration": 2.91538667678833
    },
    {
      "auxiliary_loss_clip": 0.01156193,
      "auxiliary_loss_mlp": 0.00762553,
      "balance_loss_clip": 1.05340111,
      "balance_loss_mlp": 1.000476,
      "epoch": 0.8272710875969458,
      "flos": 17566602900480.0,
      "grad_norm": 11.382035207178143,
      "language_loss": 0.75642705,
      "learning_rate": 3.047482757949078e-07,
      "loss": 0.7756145,
      "num_input_tokens_seen": 148570960,
      "step": 6880,
      "time_per_iteration": 2.6848113536834717
    },
    {
      "auxiliary_loss_clip": 0.01155113,
      "auxiliary_loss_mlp": 0.01026885,
      "balance_loss_clip": 1.04554534,
      "balance_loss_mlp": 1.01921368,
      "epoch": 0.827391330487585,
      "flos": 19755465886080.0,
      "grad_norm": 1.9712119746001129,
      "language_loss": 0.85857582,
      "learning_rate": 3.043350864099605e-07,
      "loss": 0.88039577,
      "num_input_tokens_seen": 148589520,
      "step": 6881,
      "time_per_iteration": 3.6792452335357666
    },
    {
      "auxiliary_loss_clip": 0.0114139,
      "auxiliary_loss_mlp": 0.00761864,
      "balance_loss_clip": 1.04877388,
      "balance_loss_mlp": 1.00056696,
      "epoch": 0.827511573378224,
      "flos": 16835174254080.0,
      "grad_norm": 2.13360836479264,
      "language_loss": 0.81131053,
      "learning_rate": 3.039221542560315e-07,
      "loss": 0.83034307,
      "num_input_tokens_seen": 148606085,
      "step": 6882,
      "time_per_iteration": 2.6518311500549316
    },
    {
      "auxiliary_loss_clip": 0.01159713,
      "auxiliary_loss_mlp": 0.01030113,
      "balance_loss_clip": 1.05053532,
      "balance_loss_mlp": 1.02234089,
      "epoch": 0.8276318162688631,
      "flos": 18369242259840.0,
      "grad_norm": 1.855769989040638,
      "language_loss": 0.73738176,
      "learning_rate": 3.0350947939576356e-07,
      "loss": 0.75927997,
      "num_input_tokens_seen": 148625240,
      "step": 6883,
      "time_per_iteration": 2.6615099906921387
    },
    {
      "auxiliary_loss_clip": 0.01145238,
      "auxiliary_loss_mlp": 0.0103066,
      "balance_loss_clip": 1.05281949,
      "balance_loss_mlp": 1.02196372,
      "epoch": 0.8277520591595022,
      "flos": 19352691705600.0,
      "grad_norm": 2.279624188841798,
      "language_loss": 0.72582972,
      "learning_rate": 3.0309706189175876e-07,
      "loss": 0.74758875,
      "num_input_tokens_seen": 148645075,
      "step": 6884,
      "time_per_iteration": 2.6649813652038574
    },
    {
      "auxiliary_loss_clip": 0.01013423,
      "auxiliary_loss_mlp": 0.00752166,
      "balance_loss_clip": 1.01770735,
      "balance_loss_mlp": 1.0002017,
      "epoch": 0.8278723020501413,
      "flos": 67918858329600.0,
      "grad_norm": 0.793566502988179,
      "language_loss": 0.57265955,
      "learning_rate": 3.0268490180658045e-07,
      "loss": 0.5903154,
      "num_input_tokens_seen": 148707855,
      "step": 6885,
      "time_per_iteration": 3.3890576362609863
    },
    {
      "auxiliary_loss_clip": 0.01158029,
      "auxiliary_loss_mlp": 0.00761985,
      "balance_loss_clip": 1.05278182,
      "balance_loss_mlp": 1.00053525,
      "epoch": 0.8279925449407803,
      "flos": 18185738653440.0,
      "grad_norm": 2.688052505878223,
      "language_loss": 0.79396647,
      "learning_rate": 3.0227299920275305e-07,
      "loss": 0.81316662,
      "num_input_tokens_seen": 148724170,
      "step": 6886,
      "time_per_iteration": 2.635944128036499
    },
    {
      "auxiliary_loss_clip": 0.01153338,
      "auxiliary_loss_mlp": 0.01025704,
      "balance_loss_clip": 1.04911828,
      "balance_loss_mlp": 1.01709747,
      "epoch": 0.8281127878314195,
      "flos": 20631434860800.0,
      "grad_norm": 3.03040658638926,
      "language_loss": 0.86139941,
      "learning_rate": 3.018613541427613e-07,
      "loss": 0.8831898,
      "num_input_tokens_seen": 148743690,
      "step": 6887,
      "time_per_iteration": 2.7178590297698975
    },
    {
      "auxiliary_loss_clip": 0.01112905,
      "auxiliary_loss_mlp": 0.01028228,
      "balance_loss_clip": 1.03978276,
      "balance_loss_mlp": 1.0204196,
      "epoch": 0.8282330307220586,
      "flos": 18004282122240.0,
      "grad_norm": 1.7068526880940018,
      "language_loss": 0.73927808,
      "learning_rate": 3.0144996668905243e-07,
      "loss": 0.76068944,
      "num_input_tokens_seen": 148761070,
      "step": 6888,
      "time_per_iteration": 2.707533359527588
    },
    {
      "auxiliary_loss_clip": 0.0115556,
      "auxiliary_loss_mlp": 0.01032747,
      "balance_loss_clip": 1.0518949,
      "balance_loss_mlp": 1.02544546,
      "epoch": 0.8283532736126976,
      "flos": 20084120352000.0,
      "grad_norm": 3.1270333472421217,
      "language_loss": 0.82074422,
      "learning_rate": 3.010388369040331e-07,
      "loss": 0.84262729,
      "num_input_tokens_seen": 148779730,
      "step": 6889,
      "time_per_iteration": 3.620161294937134
    },
    {
      "auxiliary_loss_clip": 0.01161683,
      "auxiliary_loss_mlp": 0.01029157,
      "balance_loss_clip": 1.05052209,
      "balance_loss_mlp": 1.0211283,
      "epoch": 0.8284735165033368,
      "flos": 31868421805440.0,
      "grad_norm": 5.34519357496693,
      "language_loss": 0.83131742,
      "learning_rate": 3.0062796485007156e-07,
      "loss": 0.85322583,
      "num_input_tokens_seen": 148800670,
      "step": 6890,
      "time_per_iteration": 2.824957847595215
    },
    {
      "auxiliary_loss_clip": 0.01154502,
      "auxiliary_loss_mlp": 0.010268,
      "balance_loss_clip": 1.04899502,
      "balance_loss_mlp": 1.01878345,
      "epoch": 0.8285937593939758,
      "flos": 26651319840000.0,
      "grad_norm": 6.299165198444288,
      "language_loss": 0.65469468,
      "learning_rate": 3.002173505894965e-07,
      "loss": 0.67650771,
      "num_input_tokens_seen": 148819820,
      "step": 6891,
      "time_per_iteration": 2.718947410583496
    },
    {
      "auxiliary_loss_clip": 0.01108198,
      "auxiliary_loss_mlp": 0.01028132,
      "balance_loss_clip": 1.03896117,
      "balance_loss_mlp": 1.01891756,
      "epoch": 0.8287140022846149,
      "flos": 20193683811840.0,
      "grad_norm": 3.084401032647132,
      "language_loss": 0.6278913,
      "learning_rate": 2.998069941845973e-07,
      "loss": 0.64925468,
      "num_input_tokens_seen": 148838890,
      "step": 6892,
      "time_per_iteration": 2.785407304763794
    },
    {
      "auxiliary_loss_clip": 0.01021495,
      "auxiliary_loss_mlp": 0.01000488,
      "balance_loss_clip": 1.00825751,
      "balance_loss_mlp": 0.99956459,
      "epoch": 0.8288342451752541,
      "flos": 70755980019840.0,
      "grad_norm": 0.7136153697705149,
      "language_loss": 0.57393938,
      "learning_rate": 2.993968956976258e-07,
      "loss": 0.59415919,
      "num_input_tokens_seen": 148906635,
      "step": 6893,
      "time_per_iteration": 3.463386297225952
    },
    {
      "auxiliary_loss_clip": 0.01136378,
      "auxiliary_loss_mlp": 0.01034489,
      "balance_loss_clip": 1.04616082,
      "balance_loss_mlp": 1.02658582,
      "epoch": 0.8289544880658931,
      "flos": 24572235795840.0,
      "grad_norm": 7.0619882168610175,
      "language_loss": 0.70302445,
      "learning_rate": 2.9898705519079313e-07,
      "loss": 0.72473311,
      "num_input_tokens_seen": 148925740,
      "step": 6894,
      "time_per_iteration": 2.7743749618530273
    },
    {
      "auxiliary_loss_clip": 0.01174397,
      "auxiliary_loss_mlp": 0.01022615,
      "balance_loss_clip": 1.052279,
      "balance_loss_mlp": 1.01531291,
      "epoch": 0.8290747309565322,
      "flos": 22273378387200.0,
      "grad_norm": 2.042527445801408,
      "language_loss": 0.7495349,
      "learning_rate": 2.985774727262715e-07,
      "loss": 0.771505,
      "num_input_tokens_seen": 148944585,
      "step": 6895,
      "time_per_iteration": 2.660050868988037
    },
    {
      "auxiliary_loss_clip": 0.01131822,
      "auxiliary_loss_mlp": 0.01028552,
      "balance_loss_clip": 1.04651117,
      "balance_loss_mlp": 1.02111697,
      "epoch": 0.8291949738471713,
      "flos": 23255570856960.0,
      "grad_norm": 1.8670946445822223,
      "language_loss": 0.81644726,
      "learning_rate": 2.981681483661949e-07,
      "loss": 0.83805108,
      "num_input_tokens_seen": 148964170,
      "step": 6896,
      "time_per_iteration": 2.71545147895813
    },
    {
      "auxiliary_loss_clip": 0.01148388,
      "auxiliary_loss_mlp": 0.01031897,
      "balance_loss_clip": 1.04845238,
      "balance_loss_mlp": 1.0238235,
      "epoch": 0.8293152167378104,
      "flos": 52555768185600.0,
      "grad_norm": 1.5948620576157497,
      "language_loss": 0.71312249,
      "learning_rate": 2.9775908217265633e-07,
      "loss": 0.73492539,
      "num_input_tokens_seen": 148989405,
      "step": 6897,
      "time_per_iteration": 3.017333745956421
    },
    {
      "auxiliary_loss_clip": 0.01029651,
      "auxiliary_loss_mlp": 0.01001851,
      "balance_loss_clip": 1.00797141,
      "balance_loss_mlp": 1.00092697,
      "epoch": 0.8294354596284494,
      "flos": 63356156294400.0,
      "grad_norm": 0.8263708465393172,
      "language_loss": 0.50292337,
      "learning_rate": 2.9735027420771253e-07,
      "loss": 0.52323842,
      "num_input_tokens_seen": 149049740,
      "step": 6898,
      "time_per_iteration": 3.2065184116363525
    },
    {
      "auxiliary_loss_clip": 0.01139028,
      "auxiliary_loss_mlp": 0.01027139,
      "balance_loss_clip": 1.05101252,
      "balance_loss_mlp": 1.02001035,
      "epoch": 0.8295557025190886,
      "flos": 24827021942400.0,
      "grad_norm": 4.191163780666742,
      "language_loss": 0.71536702,
      "learning_rate": 2.969417245333774e-07,
      "loss": 0.73702872,
      "num_input_tokens_seen": 149069120,
      "step": 6899,
      "time_per_iteration": 2.74474835395813
    },
    {
      "auxiliary_loss_clip": 0.01175403,
      "auxiliary_loss_mlp": 0.01023092,
      "balance_loss_clip": 1.05300879,
      "balance_loss_mlp": 1.01568031,
      "epoch": 0.8296759454097277,
      "flos": 25118580637440.0,
      "grad_norm": 2.1100803795157512,
      "language_loss": 0.7857393,
      "learning_rate": 2.9653343321162915e-07,
      "loss": 0.80772424,
      "num_input_tokens_seen": 149088630,
      "step": 6900,
      "time_per_iteration": 2.6878697872161865
    },
    {
      "auxiliary_loss_clip": 0.01137329,
      "auxiliary_loss_mlp": 0.01027227,
      "balance_loss_clip": 1.0472014,
      "balance_loss_mlp": 1.01896,
      "epoch": 0.8297961883003667,
      "flos": 24132581326080.0,
      "grad_norm": 3.235655619691237,
      "language_loss": 0.65154707,
      "learning_rate": 2.9612540030440446e-07,
      "loss": 0.67319262,
      "num_input_tokens_seen": 149109175,
      "step": 6901,
      "time_per_iteration": 2.719761610031128
    },
    {
      "auxiliary_loss_clip": 0.01054888,
      "auxiliary_loss_mlp": 0.01001637,
      "balance_loss_clip": 1.00880587,
      "balance_loss_mlp": 1.00076723,
      "epoch": 0.8299164311910058,
      "flos": 67446561375360.0,
      "grad_norm": 0.8403753739429716,
      "language_loss": 0.64033842,
      "learning_rate": 2.9571762587360206e-07,
      "loss": 0.66090369,
      "num_input_tokens_seen": 149165560,
      "step": 6902,
      "time_per_iteration": 4.391862154006958
    },
    {
      "auxiliary_loss_clip": 0.0108826,
      "auxiliary_loss_mlp": 0.00761722,
      "balance_loss_clip": 1.04472148,
      "balance_loss_mlp": 1.00047874,
      "epoch": 0.8300366740816449,
      "flos": 25228682801280.0,
      "grad_norm": 1.7734362212647696,
      "language_loss": 0.74045259,
      "learning_rate": 2.953101099810806e-07,
      "loss": 0.75895238,
      "num_input_tokens_seen": 149185165,
      "step": 6903,
      "time_per_iteration": 3.878941535949707
    },
    {
      "auxiliary_loss_clip": 0.01133516,
      "auxiliary_loss_mlp": 0.01032039,
      "balance_loss_clip": 1.04723024,
      "balance_loss_mlp": 1.02430797,
      "epoch": 0.830156916972284,
      "flos": 18041018757120.0,
      "grad_norm": 2.1761849298275235,
      "language_loss": 0.83015889,
      "learning_rate": 2.9490285268865965e-07,
      "loss": 0.85181451,
      "num_input_tokens_seen": 149202655,
      "step": 6904,
      "time_per_iteration": 2.742096185684204
    },
    {
      "auxiliary_loss_clip": 0.01150288,
      "auxiliary_loss_mlp": 0.0103327,
      "balance_loss_clip": 1.0481379,
      "balance_loss_mlp": 1.02457976,
      "epoch": 0.830277159862923,
      "flos": 26322485806080.0,
      "grad_norm": 2.0544759985404784,
      "language_loss": 0.80220038,
      "learning_rate": 2.9449585405812085e-07,
      "loss": 0.824036,
      "num_input_tokens_seen": 149220035,
      "step": 6905,
      "time_per_iteration": 2.7939889430999756
    },
    {
      "auxiliary_loss_clip": 0.01134793,
      "auxiliary_loss_mlp": 0.01030493,
      "balance_loss_clip": 1.04655719,
      "balance_loss_mlp": 1.02239263,
      "epoch": 0.8303974027535622,
      "flos": 19938861751680.0,
      "grad_norm": 2.1493738465411605,
      "language_loss": 0.7368387,
      "learning_rate": 2.940891141512043e-07,
      "loss": 0.75849158,
      "num_input_tokens_seen": 149238055,
      "step": 6906,
      "time_per_iteration": 2.713407278060913
    },
    {
      "auxiliary_loss_clip": 0.01158759,
      "auxiliary_loss_mlp": 0.00761704,
      "balance_loss_clip": 1.0508002,
      "balance_loss_mlp": 1.00044429,
      "epoch": 0.8305176456442013,
      "flos": 17165552572800.0,
      "grad_norm": 2.8166325909999523,
      "language_loss": 0.72326905,
      "learning_rate": 2.9368263302961385e-07,
      "loss": 0.74247372,
      "num_input_tokens_seen": 149256755,
      "step": 6907,
      "time_per_iteration": 2.7072384357452393
    },
    {
      "auxiliary_loss_clip": 0.01128696,
      "auxiliary_loss_mlp": 0.01028787,
      "balance_loss_clip": 1.04591918,
      "balance_loss_mlp": 1.02081764,
      "epoch": 0.8306378885348403,
      "flos": 25627614226560.0,
      "grad_norm": 1.7535119712195217,
      "language_loss": 0.80211902,
      "learning_rate": 2.9327641075501075e-07,
      "loss": 0.82369387,
      "num_input_tokens_seen": 149275745,
      "step": 6908,
      "time_per_iteration": 3.7294890880584717
    },
    {
      "auxiliary_loss_clip": 0.0114263,
      "auxiliary_loss_mlp": 0.01030615,
      "balance_loss_clip": 1.04596853,
      "balance_loss_mlp": 1.02250838,
      "epoch": 0.8307581314254795,
      "flos": 33947864985600.0,
      "grad_norm": 4.225441818964784,
      "language_loss": 0.66924238,
      "learning_rate": 2.9287044738901866e-07,
      "loss": 0.69097483,
      "num_input_tokens_seen": 149293730,
      "step": 6909,
      "time_per_iteration": 2.753697395324707
    },
    {
      "auxiliary_loss_clip": 0.01077624,
      "auxiliary_loss_mlp": 0.01027464,
      "balance_loss_clip": 1.04254127,
      "balance_loss_mlp": 1.0195787,
      "epoch": 0.8308783743161186,
      "flos": 17562724231680.0,
      "grad_norm": 2.1101916050490273,
      "language_loss": 0.90899813,
      "learning_rate": 2.9246474299322274e-07,
      "loss": 0.93004906,
      "num_input_tokens_seen": 149309290,
      "step": 6910,
      "time_per_iteration": 2.794275999069214
    },
    {
      "auxiliary_loss_clip": 0.01029297,
      "auxiliary_loss_mlp": 0.01001683,
      "balance_loss_clip": 1.00907922,
      "balance_loss_mlp": 1.00078917,
      "epoch": 0.8309986172067576,
      "flos": 69412885649280.0,
      "grad_norm": 0.9086490736549672,
      "language_loss": 0.63117051,
      "learning_rate": 2.920592976291678e-07,
      "loss": 0.65148032,
      "num_input_tokens_seen": 149366620,
      "step": 6911,
      "time_per_iteration": 3.1565730571746826
    },
    {
      "auxiliary_loss_clip": 0.01137375,
      "auxiliary_loss_mlp": 0.01025625,
      "balance_loss_clip": 1.05222392,
      "balance_loss_mlp": 1.01781082,
      "epoch": 0.8311188600973968,
      "flos": 22309755886080.0,
      "grad_norm": 2.3977759870795,
      "language_loss": 0.81001198,
      "learning_rate": 2.916541113583595e-07,
      "loss": 0.83164203,
      "num_input_tokens_seen": 149385120,
      "step": 6912,
      "time_per_iteration": 2.8190488815307617
    },
    {
      "auxiliary_loss_clip": 0.01067301,
      "auxiliary_loss_mlp": 0.01031707,
      "balance_loss_clip": 1.04080653,
      "balance_loss_mlp": 1.02374649,
      "epoch": 0.8312391029880358,
      "flos": 18770077105920.0,
      "grad_norm": 4.396663940616176,
      "language_loss": 0.66469008,
      "learning_rate": 2.912491842422642e-07,
      "loss": 0.68568015,
      "num_input_tokens_seen": 149402825,
      "step": 6913,
      "time_per_iteration": 2.970665693283081
    },
    {
      "auxiliary_loss_clip": 0.0113726,
      "auxiliary_loss_mlp": 0.01029246,
      "balance_loss_clip": 1.05143666,
      "balance_loss_mlp": 1.02204907,
      "epoch": 0.8313593458786749,
      "flos": 20376648714240.0,
      "grad_norm": 1.7010877858728874,
      "language_loss": 0.71139252,
      "learning_rate": 2.9084451634230857e-07,
      "loss": 0.73305762,
      "num_input_tokens_seen": 149422125,
      "step": 6914,
      "time_per_iteration": 4.068606853485107
    },
    {
      "auxiliary_loss_clip": 0.01163181,
      "auxiliary_loss_mlp": 0.0102951,
      "balance_loss_clip": 1.05047023,
      "balance_loss_mlp": 1.02143407,
      "epoch": 0.831479588769314,
      "flos": 32124069878400.0,
      "grad_norm": 3.5108367349886245,
      "language_loss": 0.7176621,
      "learning_rate": 2.9044010771988125e-07,
      "loss": 0.73958904,
      "num_input_tokens_seen": 149441940,
      "step": 6915,
      "time_per_iteration": 2.773879051208496
    },
    {
      "auxiliary_loss_clip": 0.01106008,
      "auxiliary_loss_mlp": 0.00761829,
      "balance_loss_clip": 1.04040062,
      "balance_loss_mlp": 1.00052571,
      "epoch": 0.8315998316599531,
      "flos": 45185929338240.0,
      "grad_norm": 1.8844364417586348,
      "language_loss": 0.72450954,
      "learning_rate": 2.900359584363303e-07,
      "loss": 0.7431879,
      "num_input_tokens_seen": 149465045,
      "step": 6916,
      "time_per_iteration": 2.9995946884155273
    },
    {
      "auxiliary_loss_clip": 0.01112281,
      "auxiliary_loss_mlp": 0.01038129,
      "balance_loss_clip": 1.04689538,
      "balance_loss_mlp": 1.03002262,
      "epoch": 0.8317200745505922,
      "flos": 18363747479040.0,
      "grad_norm": 4.5752582893674125,
      "language_loss": 0.84782362,
      "learning_rate": 2.8963206855296494e-07,
      "loss": 0.86932772,
      "num_input_tokens_seen": 149481285,
      "step": 6917,
      "time_per_iteration": 2.7876834869384766
    },
    {
      "auxiliary_loss_clip": 0.01127607,
      "auxiliary_loss_mlp": 0.01023857,
      "balance_loss_clip": 1.04697847,
      "balance_loss_mlp": 1.01605487,
      "epoch": 0.8318403174412313,
      "flos": 24206557386240.0,
      "grad_norm": 2.9161171532807537,
      "language_loss": 0.77191675,
      "learning_rate": 2.892284381310548e-07,
      "loss": 0.7934314,
      "num_input_tokens_seen": 149502700,
      "step": 6918,
      "time_per_iteration": 2.746004104614258
    },
    {
      "auxiliary_loss_clip": 0.01173905,
      "auxiliary_loss_mlp": 0.01027445,
      "balance_loss_clip": 1.04948044,
      "balance_loss_mlp": 1.01971447,
      "epoch": 0.8319605603318704,
      "flos": 22418780641920.0,
      "grad_norm": 2.9324834970852023,
      "language_loss": 0.72673833,
      "learning_rate": 2.888250672318302e-07,
      "loss": 0.74875188,
      "num_input_tokens_seen": 149520100,
      "step": 6919,
      "time_per_iteration": 2.6536028385162354
    },
    {
      "auxiliary_loss_clip": 0.01106858,
      "auxiliary_loss_mlp": 0.01027456,
      "balance_loss_clip": 1.0493536,
      "balance_loss_mlp": 1.01894999,
      "epoch": 0.8320808032225094,
      "flos": 37414501459200.0,
      "grad_norm": 1.563922647084587,
      "language_loss": 0.68909138,
      "learning_rate": 2.884219559164831e-07,
      "loss": 0.7104345,
      "num_input_tokens_seen": 149543245,
      "step": 6920,
      "time_per_iteration": 2.9421815872192383
    },
    {
      "auxiliary_loss_clip": 0.01106689,
      "auxiliary_loss_mlp": 0.01036806,
      "balance_loss_clip": 1.04028177,
      "balance_loss_mlp": 1.02853537,
      "epoch": 0.8322010461131486,
      "flos": 12787395638400.0,
      "grad_norm": 1.89926477466444,
      "language_loss": 0.81766111,
      "learning_rate": 2.880191042461635e-07,
      "loss": 0.83909607,
      "num_input_tokens_seen": 149559185,
      "step": 6921,
      "time_per_iteration": 2.7221364974975586
    },
    {
      "auxiliary_loss_clip": 0.01131145,
      "auxiliary_loss_mlp": 0.01025925,
      "balance_loss_clip": 1.04764569,
      "balance_loss_mlp": 1.01830459,
      "epoch": 0.8323212890037877,
      "flos": 15815455050240.0,
      "grad_norm": 1.6907036981868733,
      "language_loss": 0.80282027,
      "learning_rate": 2.876165122819849e-07,
      "loss": 0.82439101,
      "num_input_tokens_seen": 149577165,
      "step": 6922,
      "time_per_iteration": 2.8050537109375
    },
    {
      "auxiliary_loss_clip": 0.01141736,
      "auxiliary_loss_mlp": 0.01025128,
      "balance_loss_clip": 1.04720092,
      "balance_loss_mlp": 1.01788568,
      "epoch": 0.8324415318944267,
      "flos": 21719276208000.0,
      "grad_norm": 3.525499448608447,
      "language_loss": 0.79410875,
      "learning_rate": 2.872141800850201e-07,
      "loss": 0.81577742,
      "num_input_tokens_seen": 149594340,
      "step": 6923,
      "time_per_iteration": 2.7180304527282715
    },
    {
      "auxiliary_loss_clip": 0.01144389,
      "auxiliary_loss_mlp": 0.01028261,
      "balance_loss_clip": 1.04725635,
      "balance_loss_mlp": 1.02072144,
      "epoch": 0.8325617747850659,
      "flos": 34198700636160.0,
      "grad_norm": 1.8512243709261897,
      "language_loss": 0.73421693,
      "learning_rate": 2.868121077163024e-07,
      "loss": 0.75594342,
      "num_input_tokens_seen": 149613895,
      "step": 6924,
      "time_per_iteration": 2.813483476638794
    },
    {
      "auxiliary_loss_clip": 0.0114092,
      "auxiliary_loss_mlp": 0.01036084,
      "balance_loss_clip": 1.04514658,
      "balance_loss_mlp": 1.0279181,
      "epoch": 0.8326820176757049,
      "flos": 18369457741440.0,
      "grad_norm": 1.8717843897267288,
      "language_loss": 0.7237699,
      "learning_rate": 2.864102952368257e-07,
      "loss": 0.7455399,
      "num_input_tokens_seen": 149631820,
      "step": 6925,
      "time_per_iteration": 2.6815478801727295
    },
    {
      "auxiliary_loss_clip": 0.01141256,
      "auxiliary_loss_mlp": 0.01025258,
      "balance_loss_clip": 1.04802513,
      "balance_loss_mlp": 1.01762295,
      "epoch": 0.832802260566344,
      "flos": 35991325716480.0,
      "grad_norm": 1.3591213708414536,
      "language_loss": 0.59388757,
      "learning_rate": 2.860087427075444e-07,
      "loss": 0.61555272,
      "num_input_tokens_seen": 149656070,
      "step": 6926,
      "time_per_iteration": 2.8393478393554688
    },
    {
      "auxiliary_loss_clip": 0.01176259,
      "auxiliary_loss_mlp": 0.01027869,
      "balance_loss_clip": 1.05374455,
      "balance_loss_mlp": 1.02005517,
      "epoch": 0.8329225034569832,
      "flos": 14244434928000.0,
      "grad_norm": 2.7433095974513755,
      "language_loss": 0.86140418,
      "learning_rate": 2.856074501893744e-07,
      "loss": 0.88344544,
      "num_input_tokens_seen": 149671270,
      "step": 6927,
      "time_per_iteration": 3.563286542892456
    },
    {
      "auxiliary_loss_clip": 0.01126709,
      "auxiliary_loss_mlp": 0.01029889,
      "balance_loss_clip": 1.04953504,
      "balance_loss_mlp": 1.02219439,
      "epoch": 0.8330427463476222,
      "flos": 18077468083200.0,
      "grad_norm": 2.373561594136738,
      "language_loss": 0.81536973,
      "learning_rate": 2.8520641774319054e-07,
      "loss": 0.83693576,
      "num_input_tokens_seen": 149689360,
      "step": 6928,
      "time_per_iteration": 2.744297981262207
    },
    {
      "auxiliary_loss_clip": 0.01120155,
      "auxiliary_loss_mlp": 0.01034611,
      "balance_loss_clip": 1.04313862,
      "balance_loss_mlp": 1.02633739,
      "epoch": 0.8331629892382613,
      "flos": 18040839189120.0,
      "grad_norm": 2.982157849469137,
      "language_loss": 0.76052654,
      "learning_rate": 2.848056454298309e-07,
      "loss": 0.78207415,
      "num_input_tokens_seen": 149706685,
      "step": 6929,
      "time_per_iteration": 3.8099002838134766
    },
    {
      "auxiliary_loss_clip": 0.01145225,
      "auxiliary_loss_mlp": 0.01027567,
      "balance_loss_clip": 1.05049372,
      "balance_loss_mlp": 1.01962209,
      "epoch": 0.8332832321289004,
      "flos": 17457398576640.0,
      "grad_norm": 2.045981928171955,
      "language_loss": 0.65885067,
      "learning_rate": 2.844051333100905e-07,
      "loss": 0.68057859,
      "num_input_tokens_seen": 149724230,
      "step": 6930,
      "time_per_iteration": 2.7235260009765625
    },
    {
      "auxiliary_loss_clip": 0.01109965,
      "auxiliary_loss_mlp": 0.01026101,
      "balance_loss_clip": 1.04886103,
      "balance_loss_mlp": 1.01970243,
      "epoch": 0.8334034750195395,
      "flos": 15084852416640.0,
      "grad_norm": 3.199315506697477,
      "language_loss": 0.83856237,
      "learning_rate": 2.840048814447269e-07,
      "loss": 0.85992301,
      "num_input_tokens_seen": 149742395,
      "step": 6931,
      "time_per_iteration": 2.7291157245635986
    },
    {
      "auxiliary_loss_clip": 0.0115365,
      "auxiliary_loss_mlp": 0.01027877,
      "balance_loss_clip": 1.05151486,
      "balance_loss_mlp": 1.01985419,
      "epoch": 0.8335237179101785,
      "flos": 19427170556160.0,
      "grad_norm": 2.6774820774126353,
      "language_loss": 0.74207115,
      "learning_rate": 2.836048898944587e-07,
      "loss": 0.76388645,
      "num_input_tokens_seen": 149760820,
      "step": 6932,
      "time_per_iteration": 2.7281277179718018
    },
    {
      "auxiliary_loss_clip": 0.01109032,
      "auxiliary_loss_mlp": 0.01028471,
      "balance_loss_clip": 1.04345679,
      "balance_loss_mlp": 1.02130091,
      "epoch": 0.8336439608008177,
      "flos": 21762046327680.0,
      "grad_norm": 2.2514959971481794,
      "language_loss": 0.72888815,
      "learning_rate": 2.832051587199642e-07,
      "loss": 0.75026309,
      "num_input_tokens_seen": 149778075,
      "step": 6933,
      "time_per_iteration": 3.8848142623901367
    },
    {
      "auxiliary_loss_clip": 0.01024478,
      "auxiliary_loss_mlp": 0.00752144,
      "balance_loss_clip": 1.0176146,
      "balance_loss_mlp": 1.00021577,
      "epoch": 0.8337642036914568,
      "flos": 59702783990400.0,
      "grad_norm": 0.8083490270922867,
      "language_loss": 0.57604206,
      "learning_rate": 2.828056879818821e-07,
      "loss": 0.59380829,
      "num_input_tokens_seen": 149837150,
      "step": 6934,
      "time_per_iteration": 3.25437331199646
    },
    {
      "auxiliary_loss_clip": 0.01132798,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.04600167,
      "balance_loss_mlp": 1.01732481,
      "epoch": 0.8338844465820958,
      "flos": 27162185022720.0,
      "grad_norm": 3.6967301700556603,
      "language_loss": 0.83569098,
      "learning_rate": 2.824064777408117e-07,
      "loss": 0.85726494,
      "num_input_tokens_seen": 149856940,
      "step": 6935,
      "time_per_iteration": 2.7906973361968994
    },
    {
      "auxiliary_loss_clip": 0.01138351,
      "auxiliary_loss_mlp": 0.01032657,
      "balance_loss_clip": 1.04788649,
      "balance_loss_mlp": 1.02429414,
      "epoch": 0.8340046894727349,
      "flos": 30481264425600.0,
      "grad_norm": 2.2019287066429625,
      "language_loss": 0.76135755,
      "learning_rate": 2.8200752805731263e-07,
      "loss": 0.78306758,
      "num_input_tokens_seen": 149879930,
      "step": 6936,
      "time_per_iteration": 2.822004556655884
    },
    {
      "auxiliary_loss_clip": 0.01175452,
      "auxiliary_loss_mlp": 0.01028083,
      "balance_loss_clip": 1.05277503,
      "balance_loss_mlp": 1.02030134,
      "epoch": 0.834124932363374,
      "flos": 27126166659840.0,
      "grad_norm": 1.8997223786823922,
      "language_loss": 0.81103921,
      "learning_rate": 2.8160883899190625e-07,
      "loss": 0.83307457,
      "num_input_tokens_seen": 149903200,
      "step": 6937,
      "time_per_iteration": 2.7023725509643555
    },
    {
      "auxiliary_loss_clip": 0.01163956,
      "auxiliary_loss_mlp": 0.0102642,
      "balance_loss_clip": 1.05016232,
      "balance_loss_mlp": 1.01815271,
      "epoch": 0.8342451752540131,
      "flos": 24569865498240.0,
      "grad_norm": 2.6721231092747417,
      "language_loss": 0.73358107,
      "learning_rate": 2.8121041060507234e-07,
      "loss": 0.75548482,
      "num_input_tokens_seen": 149922230,
      "step": 6938,
      "time_per_iteration": 2.6522247791290283
    },
    {
      "auxiliary_loss_clip": 0.01176964,
      "auxiliary_loss_mlp": 0.01032533,
      "balance_loss_clip": 1.05381513,
      "balance_loss_mlp": 1.02457917,
      "epoch": 0.8343654181446521,
      "flos": 26615085995520.0,
      "grad_norm": 3.1676664603590168,
      "language_loss": 0.71565175,
      "learning_rate": 2.808122429572528e-07,
      "loss": 0.73774672,
      "num_input_tokens_seen": 149942435,
      "step": 6939,
      "time_per_iteration": 2.7000722885131836
    },
    {
      "auxiliary_loss_clip": 0.01082266,
      "auxiliary_loss_mlp": 0.00762068,
      "balance_loss_clip": 1.03820944,
      "balance_loss_mlp": 1.00051844,
      "epoch": 0.8344856610352913,
      "flos": 20777268078720.0,
      "grad_norm": 2.700586502106263,
      "language_loss": 0.75706273,
      "learning_rate": 2.804143361088489e-07,
      "loss": 0.77550608,
      "num_input_tokens_seen": 149961615,
      "step": 6940,
      "time_per_iteration": 3.8041515350341797
    },
    {
      "auxiliary_loss_clip": 0.01104667,
      "auxiliary_loss_mlp": 0.01028631,
      "balance_loss_clip": 1.04508376,
      "balance_loss_mlp": 1.02082825,
      "epoch": 0.8346059039259304,
      "flos": 26095960684800.0,
      "grad_norm": 5.677690599597057,
      "language_loss": 0.78024954,
      "learning_rate": 2.8001669012022277e-07,
      "loss": 0.80158257,
      "num_input_tokens_seen": 149979585,
      "step": 6941,
      "time_per_iteration": 2.7618319988250732
    },
    {
      "auxiliary_loss_clip": 0.01160177,
      "auxiliary_loss_mlp": 0.01030851,
      "balance_loss_clip": 1.05298042,
      "balance_loss_mlp": 1.02318025,
      "epoch": 0.8347261468165694,
      "flos": 29027708755200.0,
      "grad_norm": 1.5762601969554704,
      "language_loss": 0.69127983,
      "learning_rate": 2.7961930505169795e-07,
      "loss": 0.71319008,
      "num_input_tokens_seen": 150003830,
      "step": 6942,
      "time_per_iteration": 2.782708168029785
    },
    {
      "auxiliary_loss_clip": 0.01164997,
      "auxiliary_loss_mlp": 0.010269,
      "balance_loss_clip": 1.05206227,
      "balance_loss_mlp": 1.01834989,
      "epoch": 0.8348463897072086,
      "flos": 26396461866240.0,
      "grad_norm": 1.8730198328736707,
      "language_loss": 0.76230997,
      "learning_rate": 2.792221809635558e-07,
      "loss": 0.78422892,
      "num_input_tokens_seen": 150024460,
      "step": 6943,
      "time_per_iteration": 2.7084567546844482
    },
    {
      "auxiliary_loss_clip": 0.0113769,
      "auxiliary_loss_mlp": 0.01024369,
      "balance_loss_clip": 1.04775906,
      "balance_loss_mlp": 1.0164237,
      "epoch": 0.8349666325978476,
      "flos": 23367720096000.0,
      "grad_norm": 2.105166443501044,
      "language_loss": 0.74863565,
      "learning_rate": 2.788253179160411e-07,
      "loss": 0.77025628,
      "num_input_tokens_seen": 150045620,
      "step": 6944,
      "time_per_iteration": 2.7497963905334473
    },
    {
      "auxiliary_loss_clip": 0.01173292,
      "auxiliary_loss_mlp": 0.01024671,
      "balance_loss_clip": 1.05343699,
      "balance_loss_mlp": 1.01747632,
      "epoch": 0.8350868754884867,
      "flos": 12896528135040.0,
      "grad_norm": 2.5164425289850665,
      "language_loss": 0.64987457,
      "learning_rate": 2.7842871596935725e-07,
      "loss": 0.67185426,
      "num_input_tokens_seen": 150064135,
      "step": 6945,
      "time_per_iteration": 2.568049669265747
    },
    {
      "auxiliary_loss_clip": 0.01129253,
      "auxiliary_loss_mlp": 0.01030273,
      "balance_loss_clip": 1.04735363,
      "balance_loss_mlp": 1.02220273,
      "epoch": 0.8352071183791259,
      "flos": 26505522535680.0,
      "grad_norm": 1.5543497237842716,
      "language_loss": 0.69355869,
      "learning_rate": 2.780323751836682e-07,
      "loss": 0.71515393,
      "num_input_tokens_seen": 150085350,
      "step": 6946,
      "time_per_iteration": 2.809720754623413
    },
    {
      "auxiliary_loss_clip": 0.01120583,
      "auxiliary_loss_mlp": 0.01034014,
      "balance_loss_clip": 1.04593539,
      "balance_loss_mlp": 1.02581549,
      "epoch": 0.8353273612697649,
      "flos": 20668063754880.0,
      "grad_norm": 1.4061921574705198,
      "language_loss": 0.79150212,
      "learning_rate": 2.7763629561909876e-07,
      "loss": 0.81304812,
      "num_input_tokens_seen": 150106180,
      "step": 6947,
      "time_per_iteration": 2.7817578315734863
    },
    {
      "auxiliary_loss_clip": 0.0113757,
      "auxiliary_loss_mlp": 0.01028566,
      "balance_loss_clip": 1.04326105,
      "balance_loss_mlp": 1.02020979,
      "epoch": 0.835447604160404,
      "flos": 19754137082880.0,
      "grad_norm": 3.1846189187876246,
      "language_loss": 0.77602106,
      "learning_rate": 2.772404773357335e-07,
      "loss": 0.7976824,
      "num_input_tokens_seen": 150125585,
      "step": 6948,
      "time_per_iteration": 2.737513780593872
    },
    {
      "auxiliary_loss_clip": 0.01110892,
      "auxiliary_loss_mlp": 0.010254,
      "balance_loss_clip": 1.04167509,
      "balance_loss_mlp": 1.01749682,
      "epoch": 0.8355678470510431,
      "flos": 23435842239360.0,
      "grad_norm": 2.083640658739509,
      "language_loss": 0.78514498,
      "learning_rate": 2.7684492039361853e-07,
      "loss": 0.80650795,
      "num_input_tokens_seen": 150144810,
      "step": 6949,
      "time_per_iteration": 2.8347785472869873
    },
    {
      "auxiliary_loss_clip": 0.01135807,
      "auxiliary_loss_mlp": 0.01029525,
      "balance_loss_clip": 1.05070627,
      "balance_loss_mlp": 1.02224135,
      "epoch": 0.8356880899416822,
      "flos": 21214588164480.0,
      "grad_norm": 1.8028809799192513,
      "language_loss": 0.83530259,
      "learning_rate": 2.764496248527586e-07,
      "loss": 0.85695589,
      "num_input_tokens_seen": 150163785,
      "step": 6950,
      "time_per_iteration": 2.659792423248291
    },
    {
      "auxiliary_loss_clip": 0.01144164,
      "auxiliary_loss_mlp": 0.01021667,
      "balance_loss_clip": 1.04684842,
      "balance_loss_mlp": 1.01288152,
      "epoch": 0.8358083328323213,
      "flos": 28037543466240.0,
      "grad_norm": 2.2677684005660805,
      "language_loss": 0.78851068,
      "learning_rate": 2.760545907731211e-07,
      "loss": 0.81016898,
      "num_input_tokens_seen": 150184360,
      "step": 6951,
      "time_per_iteration": 2.757957935333252
    },
    {
      "auxiliary_loss_clip": 0.01078508,
      "auxiliary_loss_mlp": 0.01030751,
      "balance_loss_clip": 1.04084384,
      "balance_loss_mlp": 1.0223285,
      "epoch": 0.8359285757229604,
      "flos": 27783655159680.0,
      "grad_norm": 1.7087562719417588,
      "language_loss": 0.67991829,
      "learning_rate": 2.75659818214631e-07,
      "loss": 0.70101088,
      "num_input_tokens_seen": 150205465,
      "step": 6952,
      "time_per_iteration": 2.9009737968444824
    },
    {
      "auxiliary_loss_clip": 0.01160018,
      "auxiliary_loss_mlp": 0.01027987,
      "balance_loss_clip": 1.05035448,
      "balance_loss_mlp": 1.02067697,
      "epoch": 0.8360488186135995,
      "flos": 21435115714560.0,
      "grad_norm": 2.2465575431117752,
      "language_loss": 0.78509355,
      "learning_rate": 2.752653072371749e-07,
      "loss": 0.80697358,
      "num_input_tokens_seen": 150224900,
      "step": 6953,
      "time_per_iteration": 4.352516174316406
    },
    {
      "auxiliary_loss_clip": 0.01157312,
      "auxiliary_loss_mlp": 0.01027496,
      "balance_loss_clip": 1.04962301,
      "balance_loss_mlp": 1.01992583,
      "epoch": 0.8361690615042385,
      "flos": 27632327160960.0,
      "grad_norm": 2.1643630475875066,
      "language_loss": 0.74660265,
      "learning_rate": 2.7487105790060105e-07,
      "loss": 0.76845074,
      "num_input_tokens_seen": 150244310,
      "step": 6954,
      "time_per_iteration": 2.8870954513549805
    },
    {
      "auxiliary_loss_clip": 0.0114112,
      "auxiliary_loss_mlp": 0.00761788,
      "balance_loss_clip": 1.04875553,
      "balance_loss_mlp": 1.00057054,
      "epoch": 0.8362893043948777,
      "flos": 39202529598720.0,
      "grad_norm": 1.8114345785608594,
      "language_loss": 0.69623089,
      "learning_rate": 2.7447707026471587e-07,
      "loss": 0.71526003,
      "num_input_tokens_seen": 150267285,
      "step": 6955,
      "time_per_iteration": 3.8132400512695312
    },
    {
      "auxiliary_loss_clip": 0.01117608,
      "auxiliary_loss_mlp": 0.01028344,
      "balance_loss_clip": 1.04122496,
      "balance_loss_mlp": 1.02087522,
      "epoch": 0.8364095472855168,
      "flos": 24785329230720.0,
      "grad_norm": 2.3092674889572704,
      "language_loss": 0.80182207,
      "learning_rate": 2.740833443892874e-07,
      "loss": 0.82328159,
      "num_input_tokens_seen": 150285455,
      "step": 6956,
      "time_per_iteration": 2.8122920989990234
    },
    {
      "auxiliary_loss_clip": 0.01119092,
      "auxiliary_loss_mlp": 0.01023556,
      "balance_loss_clip": 1.04278255,
      "balance_loss_mlp": 1.01520562,
      "epoch": 0.8365297901761558,
      "flos": 22743412784640.0,
      "grad_norm": 1.9483146341659143,
      "language_loss": 0.80079699,
      "learning_rate": 2.7368988033404327e-07,
      "loss": 0.82222354,
      "num_input_tokens_seen": 150302970,
      "step": 6957,
      "time_per_iteration": 2.7491066455841064
    },
    {
      "auxiliary_loss_clip": 0.01156444,
      "auxiliary_loss_mlp": 0.01023136,
      "balance_loss_clip": 1.05122733,
      "balance_loss_mlp": 1.0157274,
      "epoch": 0.836650033066795,
      "flos": 28396003242240.0,
      "grad_norm": 1.7567016289529744,
      "language_loss": 0.84797823,
      "learning_rate": 2.732966781586712e-07,
      "loss": 0.8697741,
      "num_input_tokens_seen": 150322715,
      "step": 6958,
      "time_per_iteration": 3.690030813217163
    },
    {
      "auxiliary_loss_clip": 0.01158798,
      "auxiliary_loss_mlp": 0.00761673,
      "balance_loss_clip": 1.05011392,
      "balance_loss_mlp": 1.00054407,
      "epoch": 0.836770275957434,
      "flos": 22236857233920.0,
      "grad_norm": 1.704689578326081,
      "language_loss": 0.66855538,
      "learning_rate": 2.729037379228205e-07,
      "loss": 0.68776011,
      "num_input_tokens_seen": 150342900,
      "step": 6959,
      "time_per_iteration": 2.7348198890686035
    },
    {
      "auxiliary_loss_clip": 0.01130111,
      "auxiliary_loss_mlp": 0.01029797,
      "balance_loss_clip": 1.05031502,
      "balance_loss_mlp": 1.02211368,
      "epoch": 0.8368905188480731,
      "flos": 22491930689280.0,
      "grad_norm": 1.4834636231632556,
      "language_loss": 0.80553454,
      "learning_rate": 2.725110596860998e-07,
      "loss": 0.82713366,
      "num_input_tokens_seen": 150363580,
      "step": 6960,
      "time_per_iteration": 2.719184398651123
    },
    {
      "auxiliary_loss_clip": 0.01127768,
      "auxiliary_loss_mlp": 0.0102444,
      "balance_loss_clip": 1.0493722,
      "balance_loss_mlp": 1.01724577,
      "epoch": 0.8370107617387123,
      "flos": 13370405287680.0,
      "grad_norm": 5.631935437832696,
      "language_loss": 0.70351291,
      "learning_rate": 2.7211864350807776e-07,
      "loss": 0.72503495,
      "num_input_tokens_seen": 150381780,
      "step": 6961,
      "time_per_iteration": 2.7713587284088135
    },
    {
      "auxiliary_loss_clip": 0.01130121,
      "auxiliary_loss_mlp": 0.01031158,
      "balance_loss_clip": 1.0450455,
      "balance_loss_mlp": 1.02238441,
      "epoch": 0.8371310046293513,
      "flos": 25261289372160.0,
      "grad_norm": 1.9778138096420592,
      "language_loss": 0.74134648,
      "learning_rate": 2.717264894482836e-07,
      "loss": 0.76295924,
      "num_input_tokens_seen": 150402120,
      "step": 6962,
      "time_per_iteration": 2.7895405292510986
    },
    {
      "auxiliary_loss_clip": 0.0114168,
      "auxiliary_loss_mlp": 0.01025516,
      "balance_loss_clip": 1.04637218,
      "balance_loss_mlp": 1.01735353,
      "epoch": 0.8372512475199904,
      "flos": 19792705311360.0,
      "grad_norm": 1.8407728165684467,
      "language_loss": 0.80978519,
      "learning_rate": 2.7133459756620646e-07,
      "loss": 0.83145714,
      "num_input_tokens_seen": 150419315,
      "step": 6963,
      "time_per_iteration": 2.6932644844055176
    },
    {
      "auxiliary_loss_clip": 0.01176861,
      "auxiliary_loss_mlp": 0.01030853,
      "balance_loss_clip": 1.05389452,
      "balance_loss_mlp": 1.02314627,
      "epoch": 0.8373714904106295,
      "flos": 19391224020480.0,
      "grad_norm": 1.9333230647062372,
      "language_loss": 0.73808527,
      "learning_rate": 2.7094296792129733e-07,
      "loss": 0.76016247,
      "num_input_tokens_seen": 150438915,
      "step": 6964,
      "time_per_iteration": 2.6312813758850098
    },
    {
      "auxiliary_loss_clip": 0.01147257,
      "auxiliary_loss_mlp": 0.01027082,
      "balance_loss_clip": 1.04866815,
      "balance_loss_mlp": 1.01942849,
      "epoch": 0.8374917333012686,
      "flos": 14975935401600.0,
      "grad_norm": 1.8547716420303009,
      "language_loss": 0.75736833,
      "learning_rate": 2.7055160057296424e-07,
      "loss": 0.77911168,
      "num_input_tokens_seen": 150456155,
      "step": 6965,
      "time_per_iteration": 2.637651205062866
    },
    {
      "auxiliary_loss_clip": 0.01147362,
      "auxiliary_loss_mlp": 0.01030621,
      "balance_loss_clip": 1.04781532,
      "balance_loss_mlp": 1.02234244,
      "epoch": 0.8376119761919076,
      "flos": 30331839847680.0,
      "grad_norm": 6.308455660968569,
      "language_loss": 0.72709441,
      "learning_rate": 2.7016049558057896e-07,
      "loss": 0.74887431,
      "num_input_tokens_seen": 150478115,
      "step": 6966,
      "time_per_iteration": 3.6940529346466064
    },
    {
      "auxiliary_loss_clip": 0.01119417,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 1.04642487,
      "balance_loss_mlp": 1.01820159,
      "epoch": 0.8377322190825467,
      "flos": 29423336129280.0,
      "grad_norm": 2.2606032237991744,
      "language_loss": 0.70922768,
      "learning_rate": 2.6976965300347074e-07,
      "loss": 0.73069137,
      "num_input_tokens_seen": 150500725,
      "step": 6967,
      "time_per_iteration": 2.8257603645324707
    },
    {
      "auxiliary_loss_clip": 0.01112029,
      "auxiliary_loss_mlp": 0.01024165,
      "balance_loss_clip": 1.04521084,
      "balance_loss_mlp": 1.01655984,
      "epoch": 0.8378524619731859,
      "flos": 26687086807680.0,
      "grad_norm": 3.1760142814140138,
      "language_loss": 0.69574869,
      "learning_rate": 2.693790729009309e-07,
      "loss": 0.71711063,
      "num_input_tokens_seen": 150522335,
      "step": 6968,
      "time_per_iteration": 2.847468614578247
    },
    {
      "auxiliary_loss_clip": 0.01158192,
      "auxiliary_loss_mlp": 0.0102862,
      "balance_loss_clip": 1.04918122,
      "balance_loss_mlp": 1.02103865,
      "epoch": 0.8379727048638249,
      "flos": 20703866636160.0,
      "grad_norm": 1.9031186697130247,
      "language_loss": 0.88597572,
      "learning_rate": 2.6898875533220946e-07,
      "loss": 0.90784383,
      "num_input_tokens_seen": 150541640,
      "step": 6969,
      "time_per_iteration": 2.689194679260254
    },
    {
      "auxiliary_loss_clip": 0.01138841,
      "auxiliary_loss_mlp": 0.01029077,
      "balance_loss_clip": 1.04519343,
      "balance_loss_mlp": 1.02184653,
      "epoch": 0.838092947754464,
      "flos": 20084084438400.0,
      "grad_norm": 1.836840155295467,
      "language_loss": 0.81784534,
      "learning_rate": 2.685987003565171e-07,
      "loss": 0.83952451,
      "num_input_tokens_seen": 150559680,
      "step": 6970,
      "time_per_iteration": 2.6859171390533447
    },
    {
      "auxiliary_loss_clip": 0.01139245,
      "auxiliary_loss_mlp": 0.01025522,
      "balance_loss_clip": 1.0456636,
      "balance_loss_mlp": 1.01741552,
      "epoch": 0.8382131906451031,
      "flos": 18113270964480.0,
      "grad_norm": 3.0094296464203243,
      "language_loss": 0.75335902,
      "learning_rate": 2.6820890803302566e-07,
      "loss": 0.77500665,
      "num_input_tokens_seen": 150575205,
      "step": 6971,
      "time_per_iteration": 2.7222790718078613
    },
    {
      "auxiliary_loss_clip": 0.01143857,
      "auxiliary_loss_mlp": 0.0102915,
      "balance_loss_clip": 1.04791403,
      "balance_loss_mlp": 1.02162158,
      "epoch": 0.8383334335357422,
      "flos": 17092653920640.0,
      "grad_norm": 2.5012355121777587,
      "language_loss": 0.81561339,
      "learning_rate": 2.6781937842086557e-07,
      "loss": 0.83734345,
      "num_input_tokens_seen": 150593995,
      "step": 6972,
      "time_per_iteration": 2.6301188468933105
    },
    {
      "auxiliary_loss_clip": 0.01129168,
      "auxiliary_loss_mlp": 0.01033027,
      "balance_loss_clip": 1.04433227,
      "balance_loss_mlp": 1.02531385,
      "epoch": 0.8384536764263812,
      "flos": 20704728562560.0,
      "grad_norm": 3.0371629517875873,
      "language_loss": 0.67546982,
      "learning_rate": 2.6743011157912933e-07,
      "loss": 0.69709176,
      "num_input_tokens_seen": 150613715,
      "step": 6973,
      "time_per_iteration": 2.8597958087921143
    },
    {
      "auxiliary_loss_clip": 0.01144709,
      "auxiliary_loss_mlp": 0.01026935,
      "balance_loss_clip": 1.04755294,
      "balance_loss_mlp": 1.0184474,
      "epoch": 0.8385739193170204,
      "flos": 28986842056320.0,
      "grad_norm": 1.6793579267915364,
      "language_loss": 0.6566146,
      "learning_rate": 2.6704110756686725e-07,
      "loss": 0.67833108,
      "num_input_tokens_seen": 150634540,
      "step": 6974,
      "time_per_iteration": 2.7477011680603027
    },
    {
      "auxiliary_loss_clip": 0.01163191,
      "auxiliary_loss_mlp": 0.01034583,
      "balance_loss_clip": 1.04868937,
      "balance_loss_mlp": 1.02656603,
      "epoch": 0.8386941622076595,
      "flos": 23438068882560.0,
      "grad_norm": 2.7872292314528253,
      "language_loss": 0.84136635,
      "learning_rate": 2.6665236644309085e-07,
      "loss": 0.86334407,
      "num_input_tokens_seen": 150654850,
      "step": 6975,
      "time_per_iteration": 2.7416768074035645
    },
    {
      "auxiliary_loss_clip": 0.01156461,
      "auxiliary_loss_mlp": 0.01025966,
      "balance_loss_clip": 1.04950273,
      "balance_loss_mlp": 1.01888537,
      "epoch": 0.8388144050982985,
      "flos": 23002724044800.0,
      "grad_norm": 1.8252729147671072,
      "language_loss": 0.7957921,
      "learning_rate": 2.662638882667727e-07,
      "loss": 0.81761634,
      "num_input_tokens_seen": 150673790,
      "step": 6976,
      "time_per_iteration": 2.6731765270233154
    },
    {
      "auxiliary_loss_clip": 0.01142941,
      "auxiliary_loss_mlp": 0.01034946,
      "balance_loss_clip": 1.04647052,
      "balance_loss_mlp": 1.02662551,
      "epoch": 0.8389346479889377,
      "flos": 24280353878400.0,
      "grad_norm": 2.103973233897491,
      "language_loss": 0.72944856,
      "learning_rate": 2.658756730968443e-07,
      "loss": 0.75122744,
      "num_input_tokens_seen": 150692255,
      "step": 6977,
      "time_per_iteration": 2.7341489791870117
    },
    {
      "auxiliary_loss_clip": 0.01126894,
      "auxiliary_loss_mlp": 0.01032044,
      "balance_loss_clip": 1.04654312,
      "balance_loss_mlp": 1.02370501,
      "epoch": 0.8390548908795767,
      "flos": 21215019127680.0,
      "grad_norm": 2.8855672181001193,
      "language_loss": 0.88341409,
      "learning_rate": 2.654877209921975e-07,
      "loss": 0.90500343,
      "num_input_tokens_seen": 150709790,
      "step": 6978,
      "time_per_iteration": 2.727390766143799
    },
    {
      "auxiliary_loss_clip": 0.01151908,
      "auxiliary_loss_mlp": 0.0103721,
      "balance_loss_clip": 1.04927444,
      "balance_loss_mlp": 1.02851391,
      "epoch": 0.8391751337702158,
      "flos": 35627299332480.0,
      "grad_norm": 2.452833993885455,
      "language_loss": 0.62743622,
      "learning_rate": 2.651000320116843e-07,
      "loss": 0.6493274,
      "num_input_tokens_seen": 150730675,
      "step": 6979,
      "time_per_iteration": 3.7387847900390625
    },
    {
      "auxiliary_loss_clip": 0.0113895,
      "auxiliary_loss_mlp": 0.01032816,
      "balance_loss_clip": 1.0459249,
      "balance_loss_mlp": 1.02472711,
      "epoch": 0.839295376660855,
      "flos": 21325229032320.0,
      "grad_norm": 1.9322027220802334,
      "language_loss": 0.76170129,
      "learning_rate": 2.647126062141163e-07,
      "loss": 0.78341895,
      "num_input_tokens_seen": 150749750,
      "step": 6980,
      "time_per_iteration": 2.711850881576538
    },
    {
      "auxiliary_loss_clip": 0.01163065,
      "auxiliary_loss_mlp": 0.01027975,
      "balance_loss_clip": 1.04921031,
      "balance_loss_mlp": 1.02006531,
      "epoch": 0.839415619551494,
      "flos": 18442535961600.0,
      "grad_norm": 2.3151088616705238,
      "language_loss": 0.83951396,
      "learning_rate": 2.643254436582669e-07,
      "loss": 0.86142439,
      "num_input_tokens_seen": 150769240,
      "step": 6981,
      "time_per_iteration": 3.5925920009613037
    },
    {
      "auxiliary_loss_clip": 0.01175525,
      "auxiliary_loss_mlp": 0.01026953,
      "balance_loss_clip": 1.05149245,
      "balance_loss_mlp": 1.01897764,
      "epoch": 0.8395358624421331,
      "flos": 23221958705280.0,
      "grad_norm": 2.025624127484114,
      "language_loss": 0.82099986,
      "learning_rate": 2.6393854440286743e-07,
      "loss": 0.84302461,
      "num_input_tokens_seen": 150788410,
      "step": 6982,
      "time_per_iteration": 2.750232219696045
    },
    {
      "auxiliary_loss_clip": 0.01129712,
      "auxiliary_loss_mlp": 0.01026362,
      "balance_loss_clip": 1.04838765,
      "balance_loss_mlp": 1.01886392,
      "epoch": 0.8396561053327722,
      "flos": 24381657210240.0,
      "grad_norm": 2.058720496640443,
      "language_loss": 0.70839548,
      "learning_rate": 2.6355190850661045e-07,
      "loss": 0.72995627,
      "num_input_tokens_seen": 150805245,
      "step": 6983,
      "time_per_iteration": 2.72102689743042
    },
    {
      "auxiliary_loss_clip": 0.0115413,
      "auxiliary_loss_mlp": 0.01025484,
      "balance_loss_clip": 1.05165017,
      "balance_loss_mlp": 1.01810777,
      "epoch": 0.8397763482234113,
      "flos": 22237755073920.0,
      "grad_norm": 1.5439397291313064,
      "language_loss": 0.86430073,
      "learning_rate": 2.631655360281486e-07,
      "loss": 0.88609684,
      "num_input_tokens_seen": 150824920,
      "step": 6984,
      "time_per_iteration": 2.6804981231689453
    },
    {
      "auxiliary_loss_clip": 0.01134694,
      "auxiliary_loss_mlp": 0.01025765,
      "balance_loss_clip": 1.04827523,
      "balance_loss_mlp": 1.01734257,
      "epoch": 0.8398965911140504,
      "flos": 22163743100160.0,
      "grad_norm": 2.119234027584123,
      "language_loss": 0.6594367,
      "learning_rate": 2.6277942702609323e-07,
      "loss": 0.68104124,
      "num_input_tokens_seen": 150844400,
      "step": 6985,
      "time_per_iteration": 3.7888808250427246
    },
    {
      "auxiliary_loss_clip": 0.01111046,
      "auxiliary_loss_mlp": 0.0102503,
      "balance_loss_clip": 1.04599059,
      "balance_loss_mlp": 1.01717412,
      "epoch": 0.8400168340046895,
      "flos": 21542775753600.0,
      "grad_norm": 1.9754677898606825,
      "language_loss": 0.87678164,
      "learning_rate": 2.623935815590186e-07,
      "loss": 0.89814234,
      "num_input_tokens_seen": 150862780,
      "step": 6986,
      "time_per_iteration": 2.747955322265625
    },
    {
      "auxiliary_loss_clip": 0.01132959,
      "auxiliary_loss_mlp": 0.01029468,
      "balance_loss_clip": 1.0498997,
      "balance_loss_mlp": 1.02188659,
      "epoch": 0.8401370768953286,
      "flos": 22491966602880.0,
      "grad_norm": 1.8645106256981643,
      "language_loss": 0.81046498,
      "learning_rate": 2.6200799968545516e-07,
      "loss": 0.83208919,
      "num_input_tokens_seen": 150883075,
      "step": 6987,
      "time_per_iteration": 2.8336799144744873
    },
    {
      "auxiliary_loss_clip": 0.01053589,
      "auxiliary_loss_mlp": 0.01001877,
      "balance_loss_clip": 1.00832438,
      "balance_loss_mlp": 1.00098252,
      "epoch": 0.8402573197859676,
      "flos": 59238890818560.0,
      "grad_norm": 0.7887358825847763,
      "language_loss": 0.5640924,
      "learning_rate": 2.616226814638969e-07,
      "loss": 0.58464706,
      "num_input_tokens_seen": 150948180,
      "step": 6988,
      "time_per_iteration": 3.241241693496704
    },
    {
      "auxiliary_loss_clip": 0.01147384,
      "auxiliary_loss_mlp": 0.01027648,
      "balance_loss_clip": 1.04660809,
      "balance_loss_mlp": 1.01986337,
      "epoch": 0.8403775626766068,
      "flos": 22674608282880.0,
      "grad_norm": 3.3450169805852337,
      "language_loss": 0.78092301,
      "learning_rate": 2.612376269527954e-07,
      "loss": 0.80267334,
      "num_input_tokens_seen": 150967885,
      "step": 6989,
      "time_per_iteration": 2.7348010540008545
    },
    {
      "auxiliary_loss_clip": 0.01147028,
      "auxiliary_loss_mlp": 0.01031956,
      "balance_loss_clip": 1.04948688,
      "balance_loss_mlp": 1.0241363,
      "epoch": 0.8404978055672458,
      "flos": 19609704495360.0,
      "grad_norm": 2.219833853683578,
      "language_loss": 0.67616576,
      "learning_rate": 2.608528362105635e-07,
      "loss": 0.69795561,
      "num_input_tokens_seen": 150987255,
      "step": 6990,
      "time_per_iteration": 2.7480785846710205
    },
    {
      "auxiliary_loss_clip": 0.01132806,
      "auxiliary_loss_mlp": 0.01032727,
      "balance_loss_clip": 1.0433321,
      "balance_loss_mlp": 1.02447796,
      "epoch": 0.8406180484578849,
      "flos": 27526929678720.0,
      "grad_norm": 2.6093640528929893,
      "language_loss": 0.73203111,
      "learning_rate": 2.6046830929557374e-07,
      "loss": 0.75368643,
      "num_input_tokens_seen": 151006905,
      "step": 6991,
      "time_per_iteration": 2.7396395206451416
    },
    {
      "auxiliary_loss_clip": 0.01123814,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.04712343,
      "balance_loss_mlp": 1.02078199,
      "epoch": 0.8407382913485241,
      "flos": 22127473342080.0,
      "grad_norm": 1.9289988934743838,
      "language_loss": 0.85201013,
      "learning_rate": 2.6008404626615776e-07,
      "loss": 0.87353647,
      "num_input_tokens_seen": 151025405,
      "step": 6992,
      "time_per_iteration": 3.7186272144317627
    },
    {
      "auxiliary_loss_clip": 0.01135539,
      "auxiliary_loss_mlp": 0.01026304,
      "balance_loss_clip": 1.05066597,
      "balance_loss_mlp": 1.0181644,
      "epoch": 0.8408585342391631,
      "flos": 13918473982080.0,
      "grad_norm": 3.8796011238457275,
      "language_loss": 0.73869646,
      "learning_rate": 2.597000471806092e-07,
      "loss": 0.76031494,
      "num_input_tokens_seen": 151041970,
      "step": 6993,
      "time_per_iteration": 2.725397825241089
    },
    {
      "auxiliary_loss_clip": 0.01130037,
      "auxiliary_loss_mlp": 0.01034488,
      "balance_loss_clip": 1.04894567,
      "balance_loss_mlp": 1.02633965,
      "epoch": 0.8409787771298022,
      "flos": 20187865808640.0,
      "grad_norm": 6.000006530737197,
      "language_loss": 0.73413783,
      "learning_rate": 2.593163120971793e-07,
      "loss": 0.75578308,
      "num_input_tokens_seen": 151060835,
      "step": 6994,
      "time_per_iteration": 2.704432249069214
    },
    {
      "auxiliary_loss_clip": 0.01166312,
      "auxiliary_loss_mlp": 0.01030596,
      "balance_loss_clip": 1.05253184,
      "balance_loss_mlp": 1.02223349,
      "epoch": 0.8410990200204413,
      "flos": 23142523777920.0,
      "grad_norm": 2.235080581387803,
      "language_loss": 0.69006044,
      "learning_rate": 2.5893284107408165e-07,
      "loss": 0.71202958,
      "num_input_tokens_seen": 151078205,
      "step": 6995,
      "time_per_iteration": 2.721067428588867
    },
    {
      "auxiliary_loss_clip": 0.0115205,
      "auxiliary_loss_mlp": 0.00762304,
      "balance_loss_clip": 1.04743004,
      "balance_loss_mlp": 1.0004487,
      "epoch": 0.8412192629110804,
      "flos": 24027219757440.0,
      "grad_norm": 1.7583222089564212,
      "language_loss": 0.78143001,
      "learning_rate": 2.5854963416948726e-07,
      "loss": 0.80057359,
      "num_input_tokens_seen": 151100470,
      "step": 6996,
      "time_per_iteration": 2.71911883354187
    },
    {
      "auxiliary_loss_clip": 0.01060058,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.03847563,
      "balance_loss_mlp": 1.01950526,
      "epoch": 0.8413395058017195,
      "flos": 25591703604480.0,
      "grad_norm": 1.7490256503845183,
      "language_loss": 0.6946913,
      "learning_rate": 2.5816669144152816e-07,
      "loss": 0.71556485,
      "num_input_tokens_seen": 151121650,
      "step": 6997,
      "time_per_iteration": 3.081015110015869
    },
    {
      "auxiliary_loss_clip": 0.0103161,
      "auxiliary_loss_mlp": 0.01000155,
      "balance_loss_clip": 1.00804818,
      "balance_loss_mlp": 0.9992606,
      "epoch": 0.8414597486923585,
      "flos": 63635396624640.0,
      "grad_norm": 0.8484113159927861,
      "language_loss": 0.66270179,
      "learning_rate": 2.5778401294829777e-07,
      "loss": 0.6830194,
      "num_input_tokens_seen": 151180390,
      "step": 6998,
      "time_per_iteration": 3.642000436782837
    },
    {
      "auxiliary_loss_clip": 0.01111231,
      "auxiliary_loss_mlp": 0.01025413,
      "balance_loss_clip": 1.04345036,
      "balance_loss_mlp": 1.01795363,
      "epoch": 0.8415799915829977,
      "flos": 19098731571840.0,
      "grad_norm": 2.017341843319087,
      "language_loss": 0.65076643,
      "learning_rate": 2.574015987478473e-07,
      "loss": 0.67213285,
      "num_input_tokens_seen": 151198520,
      "step": 6999,
      "time_per_iteration": 2.7556161880493164
    },
    {
      "auxiliary_loss_clip": 0.01168575,
      "auxiliary_loss_mlp": 0.01027326,
      "balance_loss_clip": 1.05285311,
      "balance_loss_mlp": 1.0191958,
      "epoch": 0.8417002344736367,
      "flos": 19821612781440.0,
      "grad_norm": 2.1041141925592113,
      "language_loss": 0.87253928,
      "learning_rate": 2.570194488981887e-07,
      "loss": 0.89449835,
      "num_input_tokens_seen": 151215065,
      "step": 7000,
      "time_per_iteration": 2.6995201110839844
    },
    {
      "auxiliary_loss_clip": 0.01035413,
      "auxiliary_loss_mlp": 0.01000372,
      "balance_loss_clip": 1.00820804,
      "balance_loss_mlp": 0.9994418,
      "epoch": 0.8418204773642758,
      "flos": 62161516834560.0,
      "grad_norm": 0.8425443601522733,
      "language_loss": 0.60152209,
      "learning_rate": 2.566375634572939e-07,
      "loss": 0.62187994,
      "num_input_tokens_seen": 151275705,
      "step": 7001,
      "time_per_iteration": 3.230724573135376
    },
    {
      "auxiliary_loss_clip": 0.0115051,
      "auxiliary_loss_mlp": 0.01026067,
      "balance_loss_clip": 1.04661417,
      "balance_loss_mlp": 1.0174396,
      "epoch": 0.841940720254915,
      "flos": 17092905315840.0,
      "grad_norm": 1.8272756063072044,
      "language_loss": 0.7677772,
      "learning_rate": 2.562559424830943e-07,
      "loss": 0.78954303,
      "num_input_tokens_seen": 151293665,
      "step": 7002,
      "time_per_iteration": 2.700181007385254
    },
    {
      "auxiliary_loss_clip": 0.01156507,
      "auxiliary_loss_mlp": 0.01028026,
      "balance_loss_clip": 1.04989982,
      "balance_loss_mlp": 1.02018833,
      "epoch": 0.842060963145554,
      "flos": 16283586026880.0,
      "grad_norm": 2.780120777763186,
      "language_loss": 0.70634389,
      "learning_rate": 2.5587458603348256e-07,
      "loss": 0.72818929,
      "num_input_tokens_seen": 151310955,
      "step": 7003,
      "time_per_iteration": 2.633282423019409
    },
    {
      "auxiliary_loss_clip": 0.01138816,
      "auxiliary_loss_mlp": 0.01027696,
      "balance_loss_clip": 1.04566121,
      "balance_loss_mlp": 1.019804,
      "epoch": 0.8421812060361931,
      "flos": 21908238681600.0,
      "grad_norm": 5.100705507862309,
      "language_loss": 0.84158862,
      "learning_rate": 2.554934941663085e-07,
      "loss": 0.86325371,
      "num_input_tokens_seen": 151328490,
      "step": 7004,
      "time_per_iteration": 2.787278413772583
    },
    {
      "auxiliary_loss_clip": 0.01150622,
      "auxiliary_loss_mlp": 0.01026842,
      "balance_loss_clip": 1.04739213,
      "balance_loss_mlp": 1.01850963,
      "epoch": 0.8423014489268322,
      "flos": 27777693502080.0,
      "grad_norm": 2.3947696539480297,
      "language_loss": 0.73637378,
      "learning_rate": 2.5511266693938484e-07,
      "loss": 0.75814843,
      "num_input_tokens_seen": 151346950,
      "step": 7005,
      "time_per_iteration": 3.6608355045318604
    },
    {
      "auxiliary_loss_clip": 0.01178043,
      "auxiliary_loss_mlp": 0.0102648,
      "balance_loss_clip": 1.05345356,
      "balance_loss_mlp": 1.01765823,
      "epoch": 0.8424216918174713,
      "flos": 25117610970240.0,
      "grad_norm": 1.5148211116031312,
      "language_loss": 0.77752972,
      "learning_rate": 2.547321044104822e-07,
      "loss": 0.79957497,
      "num_input_tokens_seen": 151368445,
      "step": 7006,
      "time_per_iteration": 2.6058173179626465
    },
    {
      "auxiliary_loss_clip": 0.01134161,
      "auxiliary_loss_mlp": 0.01025691,
      "balance_loss_clip": 1.05034757,
      "balance_loss_mlp": 1.01728129,
      "epoch": 0.8425419347081103,
      "flos": 24748448941440.0,
      "grad_norm": 2.4302829956214667,
      "language_loss": 0.7693997,
      "learning_rate": 2.5435180663733113e-07,
      "loss": 0.79099822,
      "num_input_tokens_seen": 151388745,
      "step": 7007,
      "time_per_iteration": 3.758277177810669
    },
    {
      "auxiliary_loss_clip": 0.01107756,
      "auxiliary_loss_mlp": 0.01025359,
      "balance_loss_clip": 1.04224944,
      "balance_loss_mlp": 1.01781559,
      "epoch": 0.8426621775987495,
      "flos": 24820916630400.0,
      "grad_norm": 2.817903176345943,
      "language_loss": 0.71654803,
      "learning_rate": 2.539717736776241e-07,
      "loss": 0.73787916,
      "num_input_tokens_seen": 151404970,
      "step": 7008,
      "time_per_iteration": 2.8220877647399902
    },
    {
      "auxiliary_loss_clip": 0.01117723,
      "auxiliary_loss_mlp": 0.00761672,
      "balance_loss_clip": 1.04853868,
      "balance_loss_mlp": 1.00053096,
      "epoch": 0.8427824204893886,
      "flos": 23550074467200.0,
      "grad_norm": 1.3843039867353448,
      "language_loss": 0.76521695,
      "learning_rate": 2.535920055890097e-07,
      "loss": 0.78401089,
      "num_input_tokens_seen": 151426265,
      "step": 7009,
      "time_per_iteration": 2.8412203788757324
    },
    {
      "auxiliary_loss_clip": 0.01100213,
      "auxiliary_loss_mlp": 0.01024215,
      "balance_loss_clip": 1.04339635,
      "balance_loss_mlp": 1.01516747,
      "epoch": 0.8429026633800276,
      "flos": 16143858120960.0,
      "grad_norm": 1.8759121717503917,
      "language_loss": 0.6487149,
      "learning_rate": 2.5321250242910006e-07,
      "loss": 0.66995919,
      "num_input_tokens_seen": 151444180,
      "step": 7010,
      "time_per_iteration": 2.7950823307037354
    },
    {
      "auxiliary_loss_clip": 0.01158872,
      "auxiliary_loss_mlp": 0.01027152,
      "balance_loss_clip": 1.05107188,
      "balance_loss_mlp": 1.01946938,
      "epoch": 0.8430229062706668,
      "flos": 22198540400640.0,
      "grad_norm": 1.9881501506294814,
      "language_loss": 0.86658645,
      "learning_rate": 2.5283326425546493e-07,
      "loss": 0.88844669,
      "num_input_tokens_seen": 151463290,
      "step": 7011,
      "time_per_iteration": 3.625070095062256
    },
    {
      "auxiliary_loss_clip": 0.01120419,
      "auxiliary_loss_mlp": 0.01025199,
      "balance_loss_clip": 1.04606712,
      "balance_loss_mlp": 1.01743793,
      "epoch": 0.8431431491613058,
      "flos": 35330317683840.0,
      "grad_norm": 1.9068229528594514,
      "language_loss": 0.69722015,
      "learning_rate": 2.5245429112563443e-07,
      "loss": 0.71867633,
      "num_input_tokens_seen": 151483965,
      "step": 7012,
      "time_per_iteration": 2.959548234939575
    },
    {
      "auxiliary_loss_clip": 0.01117725,
      "auxiliary_loss_mlp": 0.00761789,
      "balance_loss_clip": 1.0449183,
      "balance_loss_mlp": 1.00046039,
      "epoch": 0.8432633920519449,
      "flos": 25812374808960.0,
      "grad_norm": 1.880599354935326,
      "language_loss": 0.82497102,
      "learning_rate": 2.5207558309709865e-07,
      "loss": 0.84376621,
      "num_input_tokens_seen": 151503700,
      "step": 7013,
      "time_per_iteration": 2.858185291290283
    },
    {
      "auxiliary_loss_clip": 0.01054325,
      "auxiliary_loss_mlp": 0.01000277,
      "balance_loss_clip": 1.00871515,
      "balance_loss_mlp": 0.99939507,
      "epoch": 0.8433836349425841,
      "flos": 64959531592320.0,
      "grad_norm": 0.6536759536032286,
      "language_loss": 0.56208366,
      "learning_rate": 2.516971402273065e-07,
      "loss": 0.58262968,
      "num_input_tokens_seen": 151569765,
      "step": 7014,
      "time_per_iteration": 3.2058448791503906
    },
    {
      "auxiliary_loss_clip": 0.01121932,
      "auxiliary_loss_mlp": 0.01030702,
      "balance_loss_clip": 1.04452598,
      "balance_loss_mlp": 1.02312577,
      "epoch": 0.8435038778332231,
      "flos": 20229989483520.0,
      "grad_norm": 2.038059964766236,
      "language_loss": 0.6756258,
      "learning_rate": 2.513189625736687e-07,
      "loss": 0.69715214,
      "num_input_tokens_seen": 151586660,
      "step": 7015,
      "time_per_iteration": 3.1241021156311035
    },
    {
      "auxiliary_loss_clip": 0.01175466,
      "auxiliary_loss_mlp": 0.00762619,
      "balance_loss_clip": 1.05074644,
      "balance_loss_mlp": 1.00045276,
      "epoch": 0.8436241207238622,
      "flos": 20992229020800.0,
      "grad_norm": 2.2688511808525527,
      "language_loss": 0.71423864,
      "learning_rate": 2.509410501935534e-07,
      "loss": 0.73361945,
      "num_input_tokens_seen": 151602295,
      "step": 7016,
      "time_per_iteration": 2.6352906227111816
    },
    {
      "auxiliary_loss_clip": 0.01138244,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.05079353,
      "balance_loss_mlp": 1.02289379,
      "epoch": 0.8437443636145013,
      "flos": 14682257804160.0,
      "grad_norm": 3.3697466678177257,
      "language_loss": 0.75493968,
      "learning_rate": 2.5056340314429116e-07,
      "loss": 0.77663457,
      "num_input_tokens_seen": 151619760,
      "step": 7017,
      "time_per_iteration": 2.797161340713501
    },
    {
      "auxiliary_loss_clip": 0.01064532,
      "auxiliary_loss_mlp": 0.01033726,
      "balance_loss_clip": 1.03620076,
      "balance_loss_mlp": 1.02506506,
      "epoch": 0.8438646065051404,
      "flos": 21608814908160.0,
      "grad_norm": 5.755968951059357,
      "language_loss": 0.80163741,
      "learning_rate": 2.5018602148316904e-07,
      "loss": 0.82262003,
      "num_input_tokens_seen": 151635795,
      "step": 7018,
      "time_per_iteration": 4.289059400558472
    },
    {
      "auxiliary_loss_clip": 0.01133865,
      "auxiliary_loss_mlp": 0.01028695,
      "balance_loss_clip": 1.05081153,
      "balance_loss_mlp": 1.02175987,
      "epoch": 0.8439848493957794,
      "flos": 23289937194240.0,
      "grad_norm": 1.7935299377591525,
      "language_loss": 0.80387443,
      "learning_rate": 2.498089052674359e-07,
      "loss": 0.82550001,
      "num_input_tokens_seen": 151653770,
      "step": 7019,
      "time_per_iteration": 2.8766608238220215
    },
    {
      "auxiliary_loss_clip": 0.01150348,
      "auxiliary_loss_mlp": 0.01030224,
      "balance_loss_clip": 1.04905522,
      "balance_loss_mlp": 1.02217746,
      "epoch": 0.8441050922864186,
      "flos": 19719339782400.0,
      "grad_norm": 2.627944268702703,
      "language_loss": 0.75516045,
      "learning_rate": 2.494320545543007e-07,
      "loss": 0.77696621,
      "num_input_tokens_seen": 151673340,
      "step": 7020,
      "time_per_iteration": 2.732607126235962
    },
    {
      "auxiliary_loss_clip": 0.01159847,
      "auxiliary_loss_mlp": 0.01024304,
      "balance_loss_clip": 1.05033231,
      "balance_loss_mlp": 1.01601887,
      "epoch": 0.8442253351770577,
      "flos": 21835268202240.0,
      "grad_norm": 1.7019319746044772,
      "language_loss": 0.66891778,
      "learning_rate": 2.490554694009308e-07,
      "loss": 0.6907593,
      "num_input_tokens_seen": 151694205,
      "step": 7021,
      "time_per_iteration": 2.6178064346313477
    },
    {
      "auxiliary_loss_clip": 0.01158675,
      "auxiliary_loss_mlp": 0.01025796,
      "balance_loss_clip": 1.05072236,
      "balance_loss_mlp": 1.01815176,
      "epoch": 0.8443455780676967,
      "flos": 34346365447680.0,
      "grad_norm": 1.558519989340324,
      "language_loss": 0.7839154,
      "learning_rate": 2.4867914986445426e-07,
      "loss": 0.80576003,
      "num_input_tokens_seen": 151716595,
      "step": 7022,
      "time_per_iteration": 2.7943437099456787
    },
    {
      "auxiliary_loss_clip": 0.01161579,
      "auxiliary_loss_mlp": 0.01029364,
      "balance_loss_clip": 1.05282903,
      "balance_loss_mlp": 1.02175832,
      "epoch": 0.8444658209583359,
      "flos": 48214599281280.0,
      "grad_norm": 5.660663422519114,
      "language_loss": 0.70931566,
      "learning_rate": 2.483030960019581e-07,
      "loss": 0.73122507,
      "num_input_tokens_seen": 151740525,
      "step": 7023,
      "time_per_iteration": 2.8901851177215576
    },
    {
      "auxiliary_loss_clip": 0.01049203,
      "auxiliary_loss_mlp": 0.01000954,
      "balance_loss_clip": 1.00919509,
      "balance_loss_mlp": 1.00015485,
      "epoch": 0.8445860638489749,
      "flos": 68484773105280.0,
      "grad_norm": 0.7471876766612292,
      "language_loss": 0.55367535,
      "learning_rate": 2.479273078704891e-07,
      "loss": 0.57417691,
      "num_input_tokens_seen": 151793890,
      "step": 7024,
      "time_per_iteration": 3.0655558109283447
    },
    {
      "auxiliary_loss_clip": 0.01039901,
      "auxiliary_loss_mlp": 0.01002751,
      "balance_loss_clip": 1.00889492,
      "balance_loss_mlp": 1.00177372,
      "epoch": 0.844706306739614,
      "flos": 62833331882880.0,
      "grad_norm": 0.7803063526367618,
      "language_loss": 0.64610398,
      "learning_rate": 2.475517855270552e-07,
      "loss": 0.66653049,
      "num_input_tokens_seen": 151853970,
      "step": 7025,
      "time_per_iteration": 3.2789158821105957
    },
    {
      "auxiliary_loss_clip": 0.01158965,
      "auxiliary_loss_mlp": 0.01034706,
      "balance_loss_clip": 1.05137587,
      "balance_loss_mlp": 1.02675152,
      "epoch": 0.8448265496302532,
      "flos": 14976114969600.0,
      "grad_norm": 4.475657039853845,
      "language_loss": 0.73107785,
      "learning_rate": 2.4717652902862143e-07,
      "loss": 0.75301456,
      "num_input_tokens_seen": 151872945,
      "step": 7026,
      "time_per_iteration": 2.61625075340271
    },
    {
      "auxiliary_loss_clip": 0.01172171,
      "auxiliary_loss_mlp": 0.01025776,
      "balance_loss_clip": 1.04999948,
      "balance_loss_mlp": 1.01806927,
      "epoch": 0.8449467925208922,
      "flos": 23441265192960.0,
      "grad_norm": 1.6679695325310238,
      "language_loss": 0.81407058,
      "learning_rate": 2.4680153843211495e-07,
      "loss": 0.83605015,
      "num_input_tokens_seen": 151892875,
      "step": 7027,
      "time_per_iteration": 2.708578109741211
    },
    {
      "auxiliary_loss_clip": 0.01117874,
      "auxiliary_loss_mlp": 0.0103626,
      "balance_loss_clip": 1.04841232,
      "balance_loss_mlp": 1.02808845,
      "epoch": 0.8450670354115313,
      "flos": 22748045639040.0,
      "grad_norm": 2.15459554775299,
      "language_loss": 0.72612977,
      "learning_rate": 2.464268137944212e-07,
      "loss": 0.74767113,
      "num_input_tokens_seen": 151914170,
      "step": 7028,
      "time_per_iteration": 2.7431247234344482
    },
    {
      "auxiliary_loss_clip": 0.01115764,
      "auxiliary_loss_mlp": 0.00762144,
      "balance_loss_clip": 1.04641962,
      "balance_loss_mlp": 1.00049996,
      "epoch": 0.8451872783021703,
      "flos": 29825571605760.0,
      "grad_norm": 2.404349335158514,
      "language_loss": 0.78460151,
      "learning_rate": 2.46052355172385e-07,
      "loss": 0.80338061,
      "num_input_tokens_seen": 151932210,
      "step": 7029,
      "time_per_iteration": 2.854541063308716
    },
    {
      "auxiliary_loss_clip": 0.01160009,
      "auxiliary_loss_mlp": 0.01029295,
      "balance_loss_clip": 1.05165553,
      "balance_loss_mlp": 1.02077127,
      "epoch": 0.8453075211928095,
      "flos": 21870029589120.0,
      "grad_norm": 1.7793618314984754,
      "language_loss": 0.74821544,
      "learning_rate": 2.456781626228128e-07,
      "loss": 0.77010846,
      "num_input_tokens_seen": 151951715,
      "step": 7030,
      "time_per_iteration": 2.652421236038208
    },
    {
      "auxiliary_loss_clip": 0.01000011,
      "auxiliary_loss_mlp": 0.00999966,
      "balance_loss_clip": 1.00810945,
      "balance_loss_mlp": 0.999066,
      "epoch": 0.8454277640834486,
      "flos": 58751869288320.0,
      "grad_norm": 0.9228943921947108,
      "language_loss": 0.66254812,
      "learning_rate": 2.453042362024675e-07,
      "loss": 0.68254787,
      "num_input_tokens_seen": 152004960,
      "step": 7031,
      "time_per_iteration": 4.622745990753174
    },
    {
      "auxiliary_loss_clip": 0.01154698,
      "auxiliary_loss_mlp": 0.01028272,
      "balance_loss_clip": 1.04813313,
      "balance_loss_mlp": 1.02066374,
      "epoch": 0.8455480069740876,
      "flos": 27090076469760.0,
      "grad_norm": 2.4036493331740596,
      "language_loss": 0.73299468,
      "learning_rate": 2.449305759680751e-07,
      "loss": 0.7548244,
      "num_input_tokens_seen": 152026285,
      "step": 7032,
      "time_per_iteration": 3.7881205081939697
    },
    {
      "auxiliary_loss_clip": 0.01120205,
      "auxiliary_loss_mlp": 0.01022464,
      "balance_loss_clip": 1.04794931,
      "balance_loss_mlp": 1.01493275,
      "epoch": 0.8456682498647268,
      "flos": 27198670262400.0,
      "grad_norm": 1.4101133290485484,
      "language_loss": 0.7507472,
      "learning_rate": 2.445571819763188e-07,
      "loss": 0.77217388,
      "num_input_tokens_seen": 152048585,
      "step": 7033,
      "time_per_iteration": 2.8273887634277344
    },
    {
      "auxiliary_loss_clip": 0.01135785,
      "auxiliary_loss_mlp": 0.01025935,
      "balance_loss_clip": 1.04413629,
      "balance_loss_mlp": 1.01796579,
      "epoch": 0.8457884927553658,
      "flos": 20631901737600.0,
      "grad_norm": 1.7212077900567528,
      "language_loss": 0.58714736,
      "learning_rate": 2.4418405428384227e-07,
      "loss": 0.60876453,
      "num_input_tokens_seen": 152068795,
      "step": 7034,
      "time_per_iteration": 2.6715590953826904
    },
    {
      "auxiliary_loss_clip": 0.01145896,
      "auxiliary_loss_mlp": 0.01022926,
      "balance_loss_clip": 1.0477922,
      "balance_loss_mlp": 1.01477206,
      "epoch": 0.8459087356460049,
      "flos": 15299023259520.0,
      "grad_norm": 2.254218257820107,
      "language_loss": 0.71491754,
      "learning_rate": 2.4381119294724864e-07,
      "loss": 0.73660576,
      "num_input_tokens_seen": 152086240,
      "step": 7035,
      "time_per_iteration": 2.735727548599243
    },
    {
      "auxiliary_loss_clip": 0.01093873,
      "auxiliary_loss_mlp": 0.01030325,
      "balance_loss_clip": 1.03970766,
      "balance_loss_mlp": 1.02266574,
      "epoch": 0.846028978536644,
      "flos": 18843155326080.0,
      "grad_norm": 2.229378612420519,
      "language_loss": 0.54147935,
      "learning_rate": 2.434385980231004e-07,
      "loss": 0.56272131,
      "num_input_tokens_seen": 152105080,
      "step": 7036,
      "time_per_iteration": 3.806823253631592
    },
    {
      "auxiliary_loss_clip": 0.01155157,
      "auxiliary_loss_mlp": 0.01026853,
      "balance_loss_clip": 1.04912949,
      "balance_loss_mlp": 1.01951599,
      "epoch": 0.8461492214272831,
      "flos": 52661740285440.0,
      "grad_norm": 1.5315867336355877,
      "language_loss": 0.65873647,
      "learning_rate": 2.4306626956792043e-07,
      "loss": 0.68055654,
      "num_input_tokens_seen": 152130025,
      "step": 7037,
      "time_per_iteration": 2.937448024749756
    },
    {
      "auxiliary_loss_clip": 0.01146619,
      "auxiliary_loss_mlp": 0.01028482,
      "balance_loss_clip": 1.04883814,
      "balance_loss_mlp": 1.02059078,
      "epoch": 0.8462694643179222,
      "flos": 18588405093120.0,
      "grad_norm": 2.5790917064959675,
      "language_loss": 0.762694,
      "learning_rate": 2.4269420763819017e-07,
      "loss": 0.78444505,
      "num_input_tokens_seen": 152148070,
      "step": 7038,
      "time_per_iteration": 2.6673598289489746
    },
    {
      "auxiliary_loss_clip": 0.01155331,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.04898894,
      "balance_loss_mlp": 1.0188148,
      "epoch": 0.8463897072085613,
      "flos": 24387080163840.0,
      "grad_norm": 2.940656584920616,
      "language_loss": 0.83763075,
      "learning_rate": 2.4232241229035223e-07,
      "loss": 0.85944492,
      "num_input_tokens_seen": 152165825,
      "step": 7039,
      "time_per_iteration": 2.697549819946289
    },
    {
      "auxiliary_loss_clip": 0.01062043,
      "auxiliary_loss_mlp": 0.00751896,
      "balance_loss_clip": 1.00798786,
      "balance_loss_mlp": 1.00029492,
      "epoch": 0.8465099500992004,
      "flos": 68702140258560.0,
      "grad_norm": 0.7530617347250369,
      "language_loss": 0.56637383,
      "learning_rate": 2.419508835808064e-07,
      "loss": 0.58451319,
      "num_input_tokens_seen": 152222380,
      "step": 7040,
      "time_per_iteration": 3.1510558128356934
    },
    {
      "auxiliary_loss_clip": 0.01139565,
      "auxiliary_loss_mlp": 0.01029599,
      "balance_loss_clip": 1.04691565,
      "balance_loss_mlp": 1.02111733,
      "epoch": 0.8466301929898394,
      "flos": 13735724561280.0,
      "grad_norm": 1.937656824871549,
      "language_loss": 0.62659115,
      "learning_rate": 2.415796215659134e-07,
      "loss": 0.64828277,
      "num_input_tokens_seen": 152239085,
      "step": 7041,
      "time_per_iteration": 2.6375248432159424
    },
    {
      "auxiliary_loss_clip": 0.0113044,
      "auxiliary_loss_mlp": 0.01028638,
      "balance_loss_clip": 1.0440377,
      "balance_loss_mlp": 1.02091312,
      "epoch": 0.8467504358804786,
      "flos": 19241260738560.0,
      "grad_norm": 2.2498395983730743,
      "language_loss": 0.77534336,
      "learning_rate": 2.412086263019939e-07,
      "loss": 0.79693413,
      "num_input_tokens_seen": 152257110,
      "step": 7042,
      "time_per_iteration": 2.7735040187835693
    },
    {
      "auxiliary_loss_clip": 0.01113755,
      "auxiliary_loss_mlp": 0.01031265,
      "balance_loss_clip": 1.05028343,
      "balance_loss_mlp": 1.02384138,
      "epoch": 0.8468706787711177,
      "flos": 21324115710720.0,
      "grad_norm": 1.712648125207247,
      "language_loss": 0.80019695,
      "learning_rate": 2.408378978453276e-07,
      "loss": 0.82164717,
      "num_input_tokens_seen": 152277230,
      "step": 7043,
      "time_per_iteration": 3.7771806716918945
    },
    {
      "auxiliary_loss_clip": 0.0104201,
      "auxiliary_loss_mlp": 0.01001635,
      "balance_loss_clip": 1.00814712,
      "balance_loss_mlp": 1.00069952,
      "epoch": 0.8469909216617567,
      "flos": 64877439058560.0,
      "grad_norm": 0.8346604217619413,
      "language_loss": 0.6389271,
      "learning_rate": 2.404674362521533e-07,
      "loss": 0.65936363,
      "num_input_tokens_seen": 152335725,
      "step": 7044,
      "time_per_iteration": 3.1606271266937256
    },
    {
      "auxiliary_loss_clip": 0.01135141,
      "auxiliary_loss_mlp": 0.01030907,
      "balance_loss_clip": 1.04454875,
      "balance_loss_mlp": 1.02277112,
      "epoch": 0.8471111645523959,
      "flos": 19280583152640.0,
      "grad_norm": 2.0865829864799137,
      "language_loss": 0.74635708,
      "learning_rate": 2.4009724157866997e-07,
      "loss": 0.76801753,
      "num_input_tokens_seen": 152352785,
      "step": 7045,
      "time_per_iteration": 2.7506542205810547
    },
    {
      "auxiliary_loss_clip": 0.01131288,
      "auxiliary_loss_mlp": 0.01029129,
      "balance_loss_clip": 1.04936969,
      "balance_loss_mlp": 1.02196467,
      "epoch": 0.8472314074430349,
      "flos": 22015826893440.0,
      "grad_norm": 2.833059777525382,
      "language_loss": 0.7654292,
      "learning_rate": 2.3972731388103564e-07,
      "loss": 0.78703332,
      "num_input_tokens_seen": 152371265,
      "step": 7046,
      "time_per_iteration": 2.6142046451568604
    },
    {
      "auxiliary_loss_clip": 0.01015085,
      "auxiliary_loss_mlp": 0.01002865,
      "balance_loss_clip": 1.00897622,
      "balance_loss_mlp": 1.0019058,
      "epoch": 0.847351650333674,
      "flos": 57882580243200.0,
      "grad_norm": 0.8015886157677147,
      "language_loss": 0.62282836,
      "learning_rate": 2.393576532153687e-07,
      "loss": 0.64300787,
      "num_input_tokens_seen": 152435050,
      "step": 7047,
      "time_per_iteration": 3.3729960918426514
    },
    {
      "auxiliary_loss_clip": 0.01048717,
      "auxiliary_loss_mlp": 0.01001277,
      "balance_loss_clip": 1.00741005,
      "balance_loss_mlp": 1.00037694,
      "epoch": 0.8474718932243132,
      "flos": 41284238313600.0,
      "grad_norm": 0.9239404629640542,
      "language_loss": 0.57744938,
      "learning_rate": 2.389882596377453e-07,
      "loss": 0.59794927,
      "num_input_tokens_seen": 152489315,
      "step": 7048,
      "time_per_iteration": 3.053088426589966
    },
    {
      "auxiliary_loss_clip": 0.01134333,
      "auxiliary_loss_mlp": 0.01031266,
      "balance_loss_clip": 1.04318893,
      "balance_loss_mlp": 1.02357101,
      "epoch": 0.8475921361149522,
      "flos": 38180906974080.0,
      "grad_norm": 1.8206953642641934,
      "language_loss": 0.7660526,
      "learning_rate": 2.386191332042031e-07,
      "loss": 0.78770852,
      "num_input_tokens_seen": 152511210,
      "step": 7049,
      "time_per_iteration": 2.6754631996154785
    },
    {
      "auxiliary_loss_clip": 0.01119829,
      "auxiliary_loss_mlp": 0.01031765,
      "balance_loss_clip": 1.04912972,
      "balance_loss_mlp": 1.0237956,
      "epoch": 0.8477123790055913,
      "flos": 25375054723200.0,
      "grad_norm": 1.7718125460719956,
      "language_loss": 0.73111522,
      "learning_rate": 2.3825027397073794e-07,
      "loss": 0.75263119,
      "num_input_tokens_seen": 152531685,
      "step": 7050,
      "time_per_iteration": 2.7601988315582275
    },
    {
      "auxiliary_loss_clip": 0.01123336,
      "auxiliary_loss_mlp": 0.0103646,
      "balance_loss_clip": 1.04569185,
      "balance_loss_mlp": 1.02810311,
      "epoch": 0.8478326218962304,
      "flos": 30225185389440.0,
      "grad_norm": 2.850306425022404,
      "language_loss": 0.66495317,
      "learning_rate": 2.3788168199330515e-07,
      "loss": 0.68655109,
      "num_input_tokens_seen": 152553245,
      "step": 7051,
      "time_per_iteration": 2.739987850189209
    },
    {
      "auxiliary_loss_clip": 0.01123215,
      "auxiliary_loss_mlp": 0.01026282,
      "balance_loss_clip": 1.04177761,
      "balance_loss_mlp": 1.01856279,
      "epoch": 0.8479528647868695,
      "flos": 38213800853760.0,
      "grad_norm": 4.06885577842779,
      "language_loss": 0.72525179,
      "learning_rate": 2.3751335732782074e-07,
      "loss": 0.74674678,
      "num_input_tokens_seen": 152574505,
      "step": 7052,
      "time_per_iteration": 2.827738046646118
    },
    {
      "auxiliary_loss_clip": 0.01148521,
      "auxiliary_loss_mlp": 0.01022878,
      "balance_loss_clip": 1.04951406,
      "balance_loss_mlp": 1.01519442,
      "epoch": 0.8480731076775085,
      "flos": 20957790856320.0,
      "grad_norm": 1.7135646793297528,
      "language_loss": 0.79205912,
      "learning_rate": 2.371453000301582e-07,
      "loss": 0.81377316,
      "num_input_tokens_seen": 152593190,
      "step": 7053,
      "time_per_iteration": 2.6846556663513184
    },
    {
      "auxiliary_loss_clip": 0.01084524,
      "auxiliary_loss_mlp": 0.01025568,
      "balance_loss_clip": 1.03861928,
      "balance_loss_mlp": 1.01808774,
      "epoch": 0.8481933505681477,
      "flos": 32596510487040.0,
      "grad_norm": 3.511191178491519,
      "language_loss": 0.74762928,
      "learning_rate": 2.3677751015615222e-07,
      "loss": 0.76873016,
      "num_input_tokens_seen": 152615265,
      "step": 7054,
      "time_per_iteration": 2.8665502071380615
    },
    {
      "auxiliary_loss_clip": 0.01116609,
      "auxiliary_loss_mlp": 0.01028396,
      "balance_loss_clip": 1.04699564,
      "balance_loss_mlp": 1.02042079,
      "epoch": 0.8483135934587868,
      "flos": 20741177888640.0,
      "grad_norm": 1.6905248860360655,
      "language_loss": 0.85585177,
      "learning_rate": 2.3640998776159593e-07,
      "loss": 0.87730181,
      "num_input_tokens_seen": 152632770,
      "step": 7055,
      "time_per_iteration": 2.8095569610595703
    },
    {
      "auxiliary_loss_clip": 0.01083731,
      "auxiliary_loss_mlp": 0.01023891,
      "balance_loss_clip": 1.04606795,
      "balance_loss_mlp": 1.01713157,
      "epoch": 0.8484338363494258,
      "flos": 21653057485440.0,
      "grad_norm": 1.6389130791387008,
      "language_loss": 0.81199664,
      "learning_rate": 2.3604273290224253e-07,
      "loss": 0.83307284,
      "num_input_tokens_seen": 152653485,
      "step": 7056,
      "time_per_iteration": 2.882606267929077
    },
    {
      "auxiliary_loss_clip": 0.01161729,
      "auxiliary_loss_mlp": 0.01033842,
      "balance_loss_clip": 1.05243874,
      "balance_loss_mlp": 1.02589655,
      "epoch": 0.848554079240065,
      "flos": 15013964926080.0,
      "grad_norm": 1.9242500883100964,
      "language_loss": 0.74848533,
      "learning_rate": 2.356757456338039e-07,
      "loss": 0.77044106,
      "num_input_tokens_seen": 152670970,
      "step": 7057,
      "time_per_iteration": 4.835287570953369
    },
    {
      "auxiliary_loss_clip": 0.01038478,
      "auxiliary_loss_mlp": 0.0100101,
      "balance_loss_clip": 1.00959659,
      "balance_loss_mlp": 1.00013971,
      "epoch": 0.848674322130704,
      "flos": 68060453742720.0,
      "grad_norm": 0.760131794532207,
      "language_loss": 0.58950847,
      "learning_rate": 2.3530902601195147e-07,
      "loss": 0.6099034,
      "num_input_tokens_seen": 152739460,
      "step": 7058,
      "time_per_iteration": 3.390300750732422
    },
    {
      "auxiliary_loss_clip": 0.01091714,
      "auxiliary_loss_mlp": 0.01026178,
      "balance_loss_clip": 1.0408349,
      "balance_loss_mlp": 1.01820314,
      "epoch": 0.8487945650213431,
      "flos": 18475788977280.0,
      "grad_norm": 2.695033340915003,
      "language_loss": 0.78715682,
      "learning_rate": 2.34942574092317e-07,
      "loss": 0.80833572,
      "num_input_tokens_seen": 152754710,
      "step": 7059,
      "time_per_iteration": 2.813154697418213
    },
    {
      "auxiliary_loss_clip": 0.01163137,
      "auxiliary_loss_mlp": 0.01034649,
      "balance_loss_clip": 1.05366457,
      "balance_loss_mlp": 1.02656651,
      "epoch": 0.8489148079119821,
      "flos": 23473189405440.0,
      "grad_norm": 1.8393244088128287,
      "language_loss": 0.76620191,
      "learning_rate": 2.3457638993049045e-07,
      "loss": 0.78817976,
      "num_input_tokens_seen": 152772700,
      "step": 7060,
      "time_per_iteration": 2.695782423019409
    },
    {
      "auxiliary_loss_clip": 0.01145668,
      "auxiliary_loss_mlp": 0.00762356,
      "balance_loss_clip": 1.04877484,
      "balance_loss_mlp": 1.00050282,
      "epoch": 0.8490350508026213,
      "flos": 19937604775680.0,
      "grad_norm": 2.2425735253930648,
      "language_loss": 0.6444959,
      "learning_rate": 2.3421047358202252e-07,
      "loss": 0.66357619,
      "num_input_tokens_seen": 152791550,
      "step": 7061,
      "time_per_iteration": 3.6763927936553955
    },
    {
      "auxiliary_loss_clip": 0.01100417,
      "auxiliary_loss_mlp": 0.0102507,
      "balance_loss_clip": 1.04186928,
      "balance_loss_mlp": 1.01688886,
      "epoch": 0.8491552936932604,
      "flos": 24279958828800.0,
      "grad_norm": 3.318083233609173,
      "language_loss": 0.8330093,
      "learning_rate": 2.3384482510242144e-07,
      "loss": 0.8542642,
      "num_input_tokens_seen": 152809410,
      "step": 7062,
      "time_per_iteration": 2.756166696548462
    },
    {
      "auxiliary_loss_clip": 0.0111109,
      "auxiliary_loss_mlp": 0.01022769,
      "balance_loss_clip": 1.04175949,
      "balance_loss_mlp": 1.01453781,
      "epoch": 0.8492755365838994,
      "flos": 22522526098560.0,
      "grad_norm": 2.074592161247839,
      "language_loss": 0.77522027,
      "learning_rate": 2.3347944454715575e-07,
      "loss": 0.79655886,
      "num_input_tokens_seen": 152825800,
      "step": 7063,
      "time_per_iteration": 2.7419960498809814
    },
    {
      "auxiliary_loss_clip": 0.01102412,
      "auxiliary_loss_mlp": 0.0102544,
      "balance_loss_clip": 1.04520833,
      "balance_loss_mlp": 1.01596308,
      "epoch": 0.8493957794745386,
      "flos": 26980441182720.0,
      "grad_norm": 1.8061061762774526,
      "language_loss": 0.67590529,
      "learning_rate": 2.331143319716542e-07,
      "loss": 0.69718379,
      "num_input_tokens_seen": 152845330,
      "step": 7064,
      "time_per_iteration": 2.859386920928955
    },
    {
      "auxiliary_loss_clip": 0.01130267,
      "auxiliary_loss_mlp": 0.01025837,
      "balance_loss_clip": 1.04556298,
      "balance_loss_mlp": 1.01790392,
      "epoch": 0.8495160223651776,
      "flos": 29861985018240.0,
      "grad_norm": 2.5317281992416554,
      "language_loss": 0.65822822,
      "learning_rate": 2.3274948743130363e-07,
      "loss": 0.67978925,
      "num_input_tokens_seen": 152865165,
      "step": 7065,
      "time_per_iteration": 2.769469738006592
    },
    {
      "auxiliary_loss_clip": 0.01149839,
      "auxiliary_loss_mlp": 0.01027097,
      "balance_loss_clip": 1.04542327,
      "balance_loss_mlp": 1.01944637,
      "epoch": 0.8496362652558167,
      "flos": 23075443128960.0,
      "grad_norm": 1.7575320945892197,
      "language_loss": 0.79559851,
      "learning_rate": 2.3238491098145085e-07,
      "loss": 0.81736779,
      "num_input_tokens_seen": 152884695,
      "step": 7066,
      "time_per_iteration": 2.647017240524292
    },
    {
      "auxiliary_loss_clip": 0.01130025,
      "auxiliary_loss_mlp": 0.01026655,
      "balance_loss_clip": 1.04500949,
      "balance_loss_mlp": 1.01858509,
      "epoch": 0.8497565081464559,
      "flos": 14609107756800.0,
      "grad_norm": 3.128222276052259,
      "language_loss": 0.73478651,
      "learning_rate": 2.3202060267740141e-07,
      "loss": 0.75635326,
      "num_input_tokens_seen": 152902220,
      "step": 7067,
      "time_per_iteration": 2.7543492317199707
    },
    {
      "auxiliary_loss_clip": 0.01158617,
      "auxiliary_loss_mlp": 0.01019997,
      "balance_loss_clip": 1.04779792,
      "balance_loss_mlp": 1.01229024,
      "epoch": 0.8498767510370949,
      "flos": 21136446126720.0,
      "grad_norm": 3.668383265645338,
      "language_loss": 0.77037144,
      "learning_rate": 2.3165656257442044e-07,
      "loss": 0.79215759,
      "num_input_tokens_seen": 152920740,
      "step": 7068,
      "time_per_iteration": 2.662646770477295
    },
    {
      "auxiliary_loss_clip": 0.0114345,
      "auxiliary_loss_mlp": 0.01026042,
      "balance_loss_clip": 1.04757023,
      "balance_loss_mlp": 1.01895201,
      "epoch": 0.849996993927734,
      "flos": 23654538195840.0,
      "grad_norm": 4.195170491681043,
      "language_loss": 0.90122509,
      "learning_rate": 2.31292790727734e-07,
      "loss": 0.92292005,
      "num_input_tokens_seen": 152938305,
      "step": 7069,
      "time_per_iteration": 2.714071273803711
    },
    {
      "auxiliary_loss_clip": 0.01149571,
      "auxiliary_loss_mlp": 0.01033355,
      "balance_loss_clip": 1.04580438,
      "balance_loss_mlp": 1.02615476,
      "epoch": 0.8501172368183731,
      "flos": 20558069331840.0,
      "grad_norm": 3.509076513509706,
      "language_loss": 0.80793226,
      "learning_rate": 2.3092928719252392e-07,
      "loss": 0.82976151,
      "num_input_tokens_seen": 152956705,
      "step": 7070,
      "time_per_iteration": 3.686666250228882
    },
    {
      "auxiliary_loss_clip": 0.01127657,
      "auxiliary_loss_mlp": 0.01021275,
      "balance_loss_clip": 1.04801393,
      "balance_loss_mlp": 1.01338589,
      "epoch": 0.8502374797090122,
      "flos": 22272624201600.0,
      "grad_norm": 2.517437137385937,
      "language_loss": 0.78565764,
      "learning_rate": 2.3056605202393475e-07,
      "loss": 0.80714691,
      "num_input_tokens_seen": 152974265,
      "step": 7071,
      "time_per_iteration": 2.681704521179199
    },
    {
      "auxiliary_loss_clip": 0.01145569,
      "auxiliary_loss_mlp": 0.01026789,
      "balance_loss_clip": 1.04640162,
      "balance_loss_mlp": 1.01852179,
      "epoch": 0.8503577225996513,
      "flos": 23659817495040.0,
      "grad_norm": 4.502478897746161,
      "language_loss": 0.67124856,
      "learning_rate": 2.3020308527706888e-07,
      "loss": 0.69297206,
      "num_input_tokens_seen": 152993680,
      "step": 7072,
      "time_per_iteration": 2.7209079265594482
    },
    {
      "auxiliary_loss_clip": 0.01086317,
      "auxiliary_loss_mlp": 0.01030629,
      "balance_loss_clip": 1.03957272,
      "balance_loss_mlp": 1.02269006,
      "epoch": 0.8504779654902904,
      "flos": 26758513002240.0,
      "grad_norm": 2.114827634073283,
      "language_loss": 0.88901579,
      "learning_rate": 2.2984038700698715e-07,
      "loss": 0.91018534,
      "num_input_tokens_seen": 153012990,
      "step": 7073,
      "time_per_iteration": 2.803039789199829
    },
    {
      "auxiliary_loss_clip": 0.01160178,
      "auxiliary_loss_mlp": 0.00761214,
      "balance_loss_clip": 1.05122054,
      "balance_loss_mlp": 1.00046968,
      "epoch": 0.8505982083809295,
      "flos": 26468247196800.0,
      "grad_norm": 1.7325042596473168,
      "language_loss": 0.79179382,
      "learning_rate": 2.2947795726871222e-07,
      "loss": 0.81100768,
      "num_input_tokens_seen": 153034015,
      "step": 7074,
      "time_per_iteration": 2.7226972579956055
    },
    {
      "auxiliary_loss_clip": 0.01146177,
      "auxiliary_loss_mlp": 0.01027549,
      "balance_loss_clip": 1.05193758,
      "balance_loss_mlp": 1.01988995,
      "epoch": 0.8507184512715685,
      "flos": 20303390926080.0,
      "grad_norm": 1.7840627103896491,
      "language_loss": 0.85629928,
      "learning_rate": 2.2911579611722253e-07,
      "loss": 0.8780365,
      "num_input_tokens_seen": 153053160,
      "step": 7075,
      "time_per_iteration": 2.6910157203674316
    },
    {
      "auxiliary_loss_clip": 0.01083733,
      "auxiliary_loss_mlp": 0.01028734,
      "balance_loss_clip": 1.04230022,
      "balance_loss_mlp": 1.02124476,
      "epoch": 0.8508386941622077,
      "flos": 19025186474880.0,
      "grad_norm": 2.135657893877779,
      "language_loss": 0.874654,
      "learning_rate": 2.2875390360745905e-07,
      "loss": 0.89577866,
      "num_input_tokens_seen": 153072565,
      "step": 7076,
      "time_per_iteration": 2.7859444618225098
    },
    {
      "auxiliary_loss_clip": 0.01138256,
      "auxiliary_loss_mlp": 0.0102711,
      "balance_loss_clip": 1.04357433,
      "balance_loss_mlp": 1.01877093,
      "epoch": 0.8509589370528468,
      "flos": 16433405654400.0,
      "grad_norm": 2.45133215099437,
      "language_loss": 0.77653211,
      "learning_rate": 2.2839227979432008e-07,
      "loss": 0.79818577,
      "num_input_tokens_seen": 153090215,
      "step": 7077,
      "time_per_iteration": 2.6540563106536865
    },
    {
      "auxiliary_loss_clip": 0.01149168,
      "auxiliary_loss_mlp": 0.01024716,
      "balance_loss_clip": 1.04893625,
      "balance_loss_mlp": 1.01705098,
      "epoch": 0.8510791799434858,
      "flos": 18259714713600.0,
      "grad_norm": 2.068365941726933,
      "language_loss": 0.85013282,
      "learning_rate": 2.2803092473266373e-07,
      "loss": 0.87187171,
      "num_input_tokens_seen": 153107740,
      "step": 7078,
      "time_per_iteration": 2.668808937072754
    },
    {
      "auxiliary_loss_clip": 0.01174248,
      "auxiliary_loss_mlp": 0.0076204,
      "balance_loss_clip": 1.05222785,
      "balance_loss_mlp": 1.00057364,
      "epoch": 0.851199422834125,
      "flos": 23441372933760.0,
      "grad_norm": 4.40164416413635,
      "language_loss": 0.87505561,
      "learning_rate": 2.2766983847730724e-07,
      "loss": 0.89441848,
      "num_input_tokens_seen": 153127410,
      "step": 7079,
      "time_per_iteration": 2.693969964981079
    },
    {
      "auxiliary_loss_clip": 0.01163408,
      "auxiliary_loss_mlp": 0.01030047,
      "balance_loss_clip": 1.05116904,
      "balance_loss_mlp": 1.02097499,
      "epoch": 0.851319665724764,
      "flos": 16289404030080.0,
      "grad_norm": 1.9213686096089473,
      "language_loss": 0.66703451,
      "learning_rate": 2.2730902108302663e-07,
      "loss": 0.68896908,
      "num_input_tokens_seen": 153144325,
      "step": 7080,
      "time_per_iteration": 2.6153087615966797
    },
    {
      "auxiliary_loss_clip": 0.01108189,
      "auxiliary_loss_mlp": 0.01032946,
      "balance_loss_clip": 1.04019737,
      "balance_loss_mlp": 1.0250957,
      "epoch": 0.8514399086154031,
      "flos": 18989347680000.0,
      "grad_norm": 3.010168380042905,
      "language_loss": 0.68545377,
      "learning_rate": 2.269484726045583e-07,
      "loss": 0.70686519,
      "num_input_tokens_seen": 153163240,
      "step": 7081,
      "time_per_iteration": 2.9633800983428955
    },
    {
      "auxiliary_loss_clip": 0.01104641,
      "auxiliary_loss_mlp": 0.010296,
      "balance_loss_clip": 1.04537964,
      "balance_loss_mlp": 1.02229881,
      "epoch": 0.8515601515060423,
      "flos": 24571194301440.0,
      "grad_norm": 1.7713157445652643,
      "language_loss": 0.79493642,
      "learning_rate": 2.2658819309659672e-07,
      "loss": 0.81627882,
      "num_input_tokens_seen": 153183440,
      "step": 7082,
      "time_per_iteration": 2.793976068496704
    },
    {
      "auxiliary_loss_clip": 0.01100498,
      "auxiliary_loss_mlp": 0.0102485,
      "balance_loss_clip": 1.03988385,
      "balance_loss_mlp": 1.0176084,
      "epoch": 0.8516803943966813,
      "flos": 19529443555200.0,
      "grad_norm": 1.9120454709241435,
      "language_loss": 0.84906673,
      "learning_rate": 2.2622818261379706e-07,
      "loss": 0.8703202,
      "num_input_tokens_seen": 153200460,
      "step": 7083,
      "time_per_iteration": 4.653203964233398
    },
    {
      "auxiliary_loss_clip": 0.01112375,
      "auxiliary_loss_mlp": 0.01026804,
      "balance_loss_clip": 1.0456109,
      "balance_loss_mlp": 1.01810741,
      "epoch": 0.8518006372873204,
      "flos": 20265792364800.0,
      "grad_norm": 1.7570338069368512,
      "language_loss": 0.74968541,
      "learning_rate": 2.2586844121077142e-07,
      "loss": 0.77107722,
      "num_input_tokens_seen": 153218970,
      "step": 7084,
      "time_per_iteration": 2.7752280235290527
    },
    {
      "auxiliary_loss_clip": 0.01135695,
      "auxiliary_loss_mlp": 0.00762018,
      "balance_loss_clip": 1.0457449,
      "balance_loss_mlp": 1.00043523,
      "epoch": 0.8519208801779595,
      "flos": 24133227770880.0,
      "grad_norm": 1.8885860179165328,
      "language_loss": 0.72138715,
      "learning_rate": 2.2550896894209215e-07,
      "loss": 0.74036431,
      "num_input_tokens_seen": 153238485,
      "step": 7085,
      "time_per_iteration": 2.7245211601257324
    },
    {
      "auxiliary_loss_clip": 0.01015364,
      "auxiliary_loss_mlp": 0.0100233,
      "balance_loss_clip": 1.00867665,
      "balance_loss_mlp": 1.00129914,
      "epoch": 0.8520411230685986,
      "flos": 63035223252480.0,
      "grad_norm": 0.6834767115288802,
      "language_loss": 0.56538594,
      "learning_rate": 2.2514976586229184e-07,
      "loss": 0.58556283,
      "num_input_tokens_seen": 153306430,
      "step": 7086,
      "time_per_iteration": 3.4523205757141113
    },
    {
      "auxiliary_loss_clip": 0.01045444,
      "auxiliary_loss_mlp": 0.00998077,
      "balance_loss_clip": 1.01785254,
      "balance_loss_mlp": 0.99712974,
      "epoch": 0.8521613659592376,
      "flos": 65836865283840.0,
      "grad_norm": 0.7661864787669116,
      "language_loss": 0.54540724,
      "learning_rate": 2.247908320258609e-07,
      "loss": 0.56584245,
      "num_input_tokens_seen": 153366520,
      "step": 7087,
      "time_per_iteration": 4.195552825927734
    },
    {
      "auxiliary_loss_clip": 0.01132165,
      "auxiliary_loss_mlp": 0.01027518,
      "balance_loss_clip": 1.04351354,
      "balance_loss_mlp": 1.01860118,
      "epoch": 0.8522816088498768,
      "flos": 23112323418240.0,
      "grad_norm": 3.051734229397607,
      "language_loss": 0.79909253,
      "learning_rate": 2.2443216748724914e-07,
      "loss": 0.82068932,
      "num_input_tokens_seen": 153387230,
      "step": 7088,
      "time_per_iteration": 2.690138101577759
    },
    {
      "auxiliary_loss_clip": 0.01114097,
      "auxiliary_loss_mlp": 0.01033313,
      "balance_loss_clip": 1.04664886,
      "balance_loss_mlp": 1.02512944,
      "epoch": 0.8524018517405159,
      "flos": 31758140073600.0,
      "grad_norm": 2.015307426304272,
      "language_loss": 0.74548805,
      "learning_rate": 2.2407377230086588e-07,
      "loss": 0.76696223,
      "num_input_tokens_seen": 153409585,
      "step": 7089,
      "time_per_iteration": 2.9046976566314697
    },
    {
      "auxiliary_loss_clip": 0.01177081,
      "auxiliary_loss_mlp": 0.01030397,
      "balance_loss_clip": 1.0540309,
      "balance_loss_mlp": 1.02222526,
      "epoch": 0.8525220946311549,
      "flos": 18690318956160.0,
      "grad_norm": 1.9464547945658177,
      "language_loss": 0.83550835,
      "learning_rate": 2.23715646521079e-07,
      "loss": 0.85758311,
      "num_input_tokens_seen": 153427105,
      "step": 7090,
      "time_per_iteration": 2.637272357940674
    },
    {
      "auxiliary_loss_clip": 0.01124345,
      "auxiliary_loss_mlp": 0.01030082,
      "balance_loss_clip": 1.0433116,
      "balance_loss_mlp": 1.02135623,
      "epoch": 0.852642337521794,
      "flos": 21793216354560.0,
      "grad_norm": 2.1384837258315614,
      "language_loss": 0.84384096,
      "learning_rate": 2.2335779020221724e-07,
      "loss": 0.86538523,
      "num_input_tokens_seen": 153443725,
      "step": 7091,
      "time_per_iteration": 2.7128140926361084
    },
    {
      "auxiliary_loss_clip": 0.01032245,
      "auxiliary_loss_mlp": 0.0100131,
      "balance_loss_clip": 1.00816441,
      "balance_loss_mlp": 1.00030899,
      "epoch": 0.8527625804124331,
      "flos": 69040132260480.0,
      "grad_norm": 0.8023706740298416,
      "language_loss": 0.5634656,
      "learning_rate": 2.2300020339856497e-07,
      "loss": 0.58380115,
      "num_input_tokens_seen": 153506410,
      "step": 7092,
      "time_per_iteration": 3.3202946186065674
    },
    {
      "auxiliary_loss_clip": 0.01094802,
      "auxiliary_loss_mlp": 0.01029759,
      "balance_loss_clip": 1.03908658,
      "balance_loss_mlp": 1.02217162,
      "epoch": 0.8528828233030722,
      "flos": 26979399688320.0,
      "grad_norm": 3.5063829183303636,
      "language_loss": 0.77632189,
      "learning_rate": 2.2264288616436966e-07,
      "loss": 0.79756749,
      "num_input_tokens_seen": 153526665,
      "step": 7093,
      "time_per_iteration": 2.835383653640747
    },
    {
      "auxiliary_loss_clip": 0.01121731,
      "auxiliary_loss_mlp": 0.01033812,
      "balance_loss_clip": 1.04398751,
      "balance_loss_mlp": 1.02530062,
      "epoch": 0.8530030661937112,
      "flos": 17487598936320.0,
      "grad_norm": 2.1635313525455304,
      "language_loss": 0.7279371,
      "learning_rate": 2.222858385538351e-07,
      "loss": 0.74949259,
      "num_input_tokens_seen": 153543465,
      "step": 7094,
      "time_per_iteration": 2.740115165710449
    },
    {
      "auxiliary_loss_clip": 0.01140164,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.0516082,
      "balance_loss_mlp": 1.01800025,
      "epoch": 0.8531233090843504,
      "flos": 22160798184960.0,
      "grad_norm": 1.9241334604872844,
      "language_loss": 0.68224955,
      "learning_rate": 2.2192906062112527e-07,
      "loss": 0.70390308,
      "num_input_tokens_seen": 153563340,
      "step": 7095,
      "time_per_iteration": 2.6740217208862305
    },
    {
      "auxiliary_loss_clip": 0.01139683,
      "auxiliary_loss_mlp": 0.01030999,
      "balance_loss_clip": 1.04685509,
      "balance_loss_mlp": 1.02261269,
      "epoch": 0.8532435519749895,
      "flos": 37635388145280.0,
      "grad_norm": 1.9364162899606923,
      "language_loss": 0.70771605,
      "learning_rate": 2.2157255242036377e-07,
      "loss": 0.72942287,
      "num_input_tokens_seen": 153587005,
      "step": 7096,
      "time_per_iteration": 3.711686849594116
    },
    {
      "auxiliary_loss_clip": 0.01117852,
      "auxiliary_loss_mlp": 0.01033062,
      "balance_loss_clip": 1.04526258,
      "balance_loss_mlp": 1.02516437,
      "epoch": 0.8533637948656285,
      "flos": 21398163598080.0,
      "grad_norm": 1.6666929604646856,
      "language_loss": 0.74312907,
      "learning_rate": 2.2121631400563135e-07,
      "loss": 0.76463819,
      "num_input_tokens_seen": 153606835,
      "step": 7097,
      "time_per_iteration": 2.7917864322662354
    },
    {
      "auxiliary_loss_clip": 0.01040106,
      "auxiliary_loss_mlp": 0.00999631,
      "balance_loss_clip": 1.01548433,
      "balance_loss_mlp": 0.99868947,
      "epoch": 0.8534840377562677,
      "flos": 53345122490880.0,
      "grad_norm": 0.7718055985067808,
      "language_loss": 0.52896869,
      "learning_rate": 2.208603454309701e-07,
      "loss": 0.54936612,
      "num_input_tokens_seen": 153664925,
      "step": 7098,
      "time_per_iteration": 3.193992853164673
    },
    {
      "auxiliary_loss_clip": 0.0117828,
      "auxiliary_loss_mlp": 0.01027873,
      "balance_loss_clip": 1.05456543,
      "balance_loss_mlp": 1.01969504,
      "epoch": 0.8536042806469067,
      "flos": 20814148368000.0,
      "grad_norm": 2.7495523656925926,
      "language_loss": 0.70573568,
      "learning_rate": 2.2050464675037994e-07,
      "loss": 0.72779715,
      "num_input_tokens_seen": 153683550,
      "step": 7099,
      "time_per_iteration": 2.6414804458618164
    },
    {
      "auxiliary_loss_clip": 0.01130488,
      "auxiliary_loss_mlp": 0.01029849,
      "balance_loss_clip": 1.04719615,
      "balance_loss_mlp": 1.02166486,
      "epoch": 0.8537245235375458,
      "flos": 24681368292480.0,
      "grad_norm": 2.223531547435041,
      "language_loss": 0.73280609,
      "learning_rate": 2.2014921801782016e-07,
      "loss": 0.75440943,
      "num_input_tokens_seen": 153703040,
      "step": 7100,
      "time_per_iteration": 2.7215511798858643
    },
    {
      "auxiliary_loss_clip": 0.01121757,
      "auxiliary_loss_mlp": 0.01023028,
      "balance_loss_clip": 1.04558849,
      "balance_loss_mlp": 1.01549447,
      "epoch": 0.853844766428185,
      "flos": 24384817607040.0,
      "grad_norm": 2.607005096614737,
      "language_loss": 0.74278414,
      "learning_rate": 2.1979405928720872e-07,
      "loss": 0.76423198,
      "num_input_tokens_seen": 153722695,
      "step": 7101,
      "time_per_iteration": 2.731565237045288
    },
    {
      "auxiliary_loss_clip": 0.01160002,
      "auxiliary_loss_mlp": 0.01026881,
      "balance_loss_clip": 1.05126786,
      "balance_loss_mlp": 1.01932883,
      "epoch": 0.853965009318824,
      "flos": 20955707867520.0,
      "grad_norm": 1.4868179088507218,
      "language_loss": 0.79704744,
      "learning_rate": 2.1943917061242257e-07,
      "loss": 0.81891626,
      "num_input_tokens_seen": 153742550,
      "step": 7102,
      "time_per_iteration": 2.6413683891296387
    },
    {
      "auxiliary_loss_clip": 0.01149298,
      "auxiliary_loss_mlp": 0.01033327,
      "balance_loss_clip": 1.04730332,
      "balance_loss_mlp": 1.02413023,
      "epoch": 0.8540852522094631,
      "flos": 24201816791040.0,
      "grad_norm": 2.2271154702316025,
      "language_loss": 0.66659403,
      "learning_rate": 2.1908455204729903e-07,
      "loss": 0.6884203,
      "num_input_tokens_seen": 153761700,
      "step": 7103,
      "time_per_iteration": 2.7590932846069336
    },
    {
      "auxiliary_loss_clip": 0.01128692,
      "auxiliary_loss_mlp": 0.01032002,
      "balance_loss_clip": 1.04405808,
      "balance_loss_mlp": 1.02383614,
      "epoch": 0.8542054951001022,
      "flos": 25082921410560.0,
      "grad_norm": 2.8738062677146026,
      "language_loss": 0.78646523,
      "learning_rate": 2.1873020364563265e-07,
      "loss": 0.80807221,
      "num_input_tokens_seen": 153780765,
      "step": 7104,
      "time_per_iteration": 2.7471067905426025
    },
    {
      "auxiliary_loss_clip": 0.01153533,
      "auxiliary_loss_mlp": 0.01030404,
      "balance_loss_clip": 1.04589927,
      "balance_loss_mlp": 1.02291191,
      "epoch": 0.8543257379907413,
      "flos": 24316551809280.0,
      "grad_norm": 7.258031097636671,
      "language_loss": 0.76297957,
      "learning_rate": 2.183761254611789e-07,
      "loss": 0.78481889,
      "num_input_tokens_seen": 153801090,
      "step": 7105,
      "time_per_iteration": 2.697658061981201
    },
    {
      "auxiliary_loss_clip": 0.01133512,
      "auxiliary_loss_mlp": 0.01027907,
      "balance_loss_clip": 1.05256629,
      "balance_loss_mlp": 1.02042627,
      "epoch": 0.8544459808813804,
      "flos": 55286630467200.0,
      "grad_norm": 1.8615222067418669,
      "language_loss": 0.70618546,
      "learning_rate": 2.1802231754764987e-07,
      "loss": 0.72779965,
      "num_input_tokens_seen": 153826530,
      "step": 7106,
      "time_per_iteration": 3.062591552734375
    },
    {
      "auxiliary_loss_clip": 0.01144968,
      "auxiliary_loss_mlp": 0.0102993,
      "balance_loss_clip": 1.04829991,
      "balance_loss_mlp": 1.02147233,
      "epoch": 0.8545662237720195,
      "flos": 25776248705280.0,
      "grad_norm": 2.0652837264766584,
      "language_loss": 0.76462489,
      "learning_rate": 2.17668779958718e-07,
      "loss": 0.78637385,
      "num_input_tokens_seen": 153849110,
      "step": 7107,
      "time_per_iteration": 2.8341593742370605
    },
    {
      "auxiliary_loss_clip": 0.0115042,
      "auxiliary_loss_mlp": 0.01023565,
      "balance_loss_clip": 1.05117989,
      "balance_loss_mlp": 1.01568508,
      "epoch": 0.8546864666626586,
      "flos": 11108320427520.0,
      "grad_norm": 2.8923129332155044,
      "language_loss": 0.81021035,
      "learning_rate": 2.1731551274801553e-07,
      "loss": 0.83195019,
      "num_input_tokens_seen": 153865550,
      "step": 7108,
      "time_per_iteration": 2.69692325592041
    },
    {
      "auxiliary_loss_clip": 0.0113681,
      "auxiliary_loss_mlp": 0.01030279,
      "balance_loss_clip": 1.0468384,
      "balance_loss_mlp": 1.0225836,
      "epoch": 0.8548067095532976,
      "flos": 25520169669120.0,
      "grad_norm": 2.5998659655660727,
      "language_loss": 0.62303674,
      "learning_rate": 2.169625159691324e-07,
      "loss": 0.64470756,
      "num_input_tokens_seen": 153885425,
      "step": 7109,
      "time_per_iteration": 4.632241487503052
    },
    {
      "auxiliary_loss_clip": 0.01173511,
      "auxiliary_loss_mlp": 0.01032675,
      "balance_loss_clip": 1.04918385,
      "balance_loss_mlp": 1.02487874,
      "epoch": 0.8549269524439368,
      "flos": 24717853532160.0,
      "grad_norm": 2.0775140607064895,
      "language_loss": 0.74047351,
      "learning_rate": 2.1660978967561784e-07,
      "loss": 0.76253533,
      "num_input_tokens_seen": 153904760,
      "step": 7110,
      "time_per_iteration": 2.680384635925293
    },
    {
      "auxiliary_loss_clip": 0.01159793,
      "auxiliary_loss_mlp": 0.00762211,
      "balance_loss_clip": 1.04798853,
      "balance_loss_mlp": 1.00050545,
      "epoch": 0.8550471953345758,
      "flos": 19825599191040.0,
      "grad_norm": 2.335571454063963,
      "language_loss": 0.78836572,
      "learning_rate": 2.1625733392098035e-07,
      "loss": 0.80758572,
      "num_input_tokens_seen": 153920370,
      "step": 7111,
      "time_per_iteration": 2.8064188957214355
    },
    {
      "auxiliary_loss_clip": 0.01156468,
      "auxiliary_loss_mlp": 0.0103054,
      "balance_loss_clip": 1.04838753,
      "balance_loss_mlp": 1.02298164,
      "epoch": 0.8551674382252149,
      "flos": 22820441500800.0,
      "grad_norm": 1.8696854993385588,
      "language_loss": 0.79886895,
      "learning_rate": 2.159051487586867e-07,
      "loss": 0.82073903,
      "num_input_tokens_seen": 153940500,
      "step": 7112,
      "time_per_iteration": 2.628584861755371
    },
    {
      "auxiliary_loss_clip": 0.01137008,
      "auxiliary_loss_mlp": 0.01029195,
      "balance_loss_clip": 1.04659629,
      "balance_loss_mlp": 1.02129698,
      "epoch": 0.8552876811158541,
      "flos": 20631255292800.0,
      "grad_norm": 2.2610822683968586,
      "language_loss": 0.72274637,
      "learning_rate": 2.155532342421642e-07,
      "loss": 0.74440843,
      "num_input_tokens_seen": 153958500,
      "step": 7113,
      "time_per_iteration": 3.644943952560425
    },
    {
      "auxiliary_loss_clip": 0.01134554,
      "auxiliary_loss_mlp": 0.01026589,
      "balance_loss_clip": 1.04734349,
      "balance_loss_mlp": 1.01868868,
      "epoch": 0.8554079240064931,
      "flos": 23112359331840.0,
      "grad_norm": 1.8446890238940383,
      "language_loss": 0.78619009,
      "learning_rate": 2.1520159042479636e-07,
      "loss": 0.80780154,
      "num_input_tokens_seen": 153976790,
      "step": 7114,
      "time_per_iteration": 2.7332019805908203
    },
    {
      "auxiliary_loss_clip": 0.0113167,
      "auxiliary_loss_mlp": 0.01026377,
      "balance_loss_clip": 1.0502677,
      "balance_loss_mlp": 1.01859272,
      "epoch": 0.8555281668971322,
      "flos": 22128047959680.0,
      "grad_norm": 2.182836263345274,
      "language_loss": 0.70943677,
      "learning_rate": 2.148502173599287e-07,
      "loss": 0.73101723,
      "num_input_tokens_seen": 153994930,
      "step": 7115,
      "time_per_iteration": 2.7198777198791504
    },
    {
      "auxiliary_loss_clip": 0.01174586,
      "auxiliary_loss_mlp": 0.01026739,
      "balance_loss_clip": 1.05226862,
      "balance_loss_mlp": 1.01888347,
      "epoch": 0.8556484097877713,
      "flos": 31139040234240.0,
      "grad_norm": 3.8527634113146765,
      "language_loss": 0.65940559,
      "learning_rate": 2.1449911510086372e-07,
      "loss": 0.6814189,
      "num_input_tokens_seen": 154014400,
      "step": 7116,
      "time_per_iteration": 2.719533681869507
    },
    {
      "auxiliary_loss_clip": 0.01087903,
      "auxiliary_loss_mlp": 0.01026859,
      "balance_loss_clip": 1.04261851,
      "balance_loss_mlp": 1.01924157,
      "epoch": 0.8557686526784104,
      "flos": 24316551809280.0,
      "grad_norm": 2.2665057690710357,
      "language_loss": 0.76714969,
      "learning_rate": 2.141482837008628e-07,
      "loss": 0.7882973,
      "num_input_tokens_seen": 154034940,
      "step": 7117,
      "time_per_iteration": 2.8241868019104004
    },
    {
      "auxiliary_loss_clip": 0.01057357,
      "auxiliary_loss_mlp": 0.01032091,
      "balance_loss_clip": 1.03731275,
      "balance_loss_mlp": 1.02464366,
      "epoch": 0.8558888955690495,
      "flos": 17712723427200.0,
      "grad_norm": 1.9399306524905933,
      "language_loss": 0.72130716,
      "learning_rate": 2.1379772321314826e-07,
      "loss": 0.74220169,
      "num_input_tokens_seen": 154052985,
      "step": 7118,
      "time_per_iteration": 2.9715209007263184
    },
    {
      "auxiliary_loss_clip": 0.01142759,
      "auxiliary_loss_mlp": 0.01037109,
      "balance_loss_clip": 1.05097842,
      "balance_loss_mlp": 1.02903867,
      "epoch": 0.8560091384596886,
      "flos": 19171702051200.0,
      "grad_norm": 3.0285372143351332,
      "language_loss": 0.81649303,
      "learning_rate": 2.1344743369089802e-07,
      "loss": 0.8382917,
      "num_input_tokens_seen": 154068765,
      "step": 7119,
      "time_per_iteration": 2.8757646083831787
    },
    {
      "auxiliary_loss_clip": 0.01106119,
      "auxiliary_loss_mlp": 0.01028169,
      "balance_loss_clip": 1.04552603,
      "balance_loss_mlp": 1.02028966,
      "epoch": 0.8561293813503277,
      "flos": 23914855036800.0,
      "grad_norm": 2.1698020135806373,
      "language_loss": 0.82347536,
      "learning_rate": 2.130974151872522e-07,
      "loss": 0.84481823,
      "num_input_tokens_seen": 154089100,
      "step": 7120,
      "time_per_iteration": 2.8880443572998047
    },
    {
      "auxiliary_loss_clip": 0.01175126,
      "auxiliary_loss_mlp": 0.01028146,
      "balance_loss_clip": 1.05087972,
      "balance_loss_mlp": 1.02084136,
      "epoch": 0.8562496242409667,
      "flos": 22529206028160.0,
      "grad_norm": 1.7372302554903258,
      "language_loss": 0.78759021,
      "learning_rate": 2.1274766775530773e-07,
      "loss": 0.80962294,
      "num_input_tokens_seen": 154108965,
      "step": 7121,
      "time_per_iteration": 2.7016634941101074
    },
    {
      "auxiliary_loss_clip": 0.01136842,
      "auxiliary_loss_mlp": 0.00762236,
      "balance_loss_clip": 1.0485456,
      "balance_loss_mlp": 1.00051165,
      "epoch": 0.8563698671316058,
      "flos": 14712745472640.0,
      "grad_norm": 2.0452814683464036,
      "language_loss": 0.80049896,
      "learning_rate": 2.1239819144812077e-07,
      "loss": 0.81948972,
      "num_input_tokens_seen": 154123425,
      "step": 7122,
      "time_per_iteration": 3.668081760406494
    },
    {
      "auxiliary_loss_clip": 0.01144049,
      "auxiliary_loss_mlp": 0.01026732,
      "balance_loss_clip": 1.04843509,
      "balance_loss_mlp": 1.01878726,
      "epoch": 0.856490110022245,
      "flos": 39167768211840.0,
      "grad_norm": 3.1814577607204697,
      "language_loss": 0.69729328,
      "learning_rate": 2.1204898631870716e-07,
      "loss": 0.71900105,
      "num_input_tokens_seen": 154148315,
      "step": 7123,
      "time_per_iteration": 2.8494253158569336
    },
    {
      "auxiliary_loss_clip": 0.01150184,
      "auxiliary_loss_mlp": 0.0076139,
      "balance_loss_clip": 1.04786634,
      "balance_loss_mlp": 1.0004667,
      "epoch": 0.856610352912884,
      "flos": 29059345658880.0,
      "grad_norm": 2.5871081334082247,
      "language_loss": 0.76125908,
      "learning_rate": 2.1170005242004006e-07,
      "loss": 0.78037483,
      "num_input_tokens_seen": 154169665,
      "step": 7124,
      "time_per_iteration": 2.755722999572754
    },
    {
      "auxiliary_loss_clip": 0.01106686,
      "auxiliary_loss_mlp": 0.01024298,
      "balance_loss_clip": 1.04392076,
      "balance_loss_mlp": 1.01692486,
      "epoch": 0.8567305958035231,
      "flos": 23878333883520.0,
      "grad_norm": 1.7674234081233642,
      "language_loss": 0.78193903,
      "learning_rate": 2.1135138980505384e-07,
      "loss": 0.80324888,
      "num_input_tokens_seen": 154190335,
      "step": 7125,
      "time_per_iteration": 2.820012092590332
    },
    {
      "auxiliary_loss_clip": 0.0112421,
      "auxiliary_loss_mlp": 0.01027517,
      "balance_loss_clip": 1.04625154,
      "balance_loss_mlp": 1.02009058,
      "epoch": 0.8568508386941622,
      "flos": 22200120599040.0,
      "grad_norm": 1.8358802070982907,
      "language_loss": 0.72274327,
      "learning_rate": 2.110029985266395e-07,
      "loss": 0.74426055,
      "num_input_tokens_seen": 154210040,
      "step": 7126,
      "time_per_iteration": 2.7902019023895264
    },
    {
      "auxiliary_loss_clip": 0.01132266,
      "auxiliary_loss_mlp": 0.01025362,
      "balance_loss_clip": 1.0456841,
      "balance_loss_mlp": 1.01756573,
      "epoch": 0.8569710815848013,
      "flos": 17307507121920.0,
      "grad_norm": 1.6380497475353932,
      "language_loss": 0.74071312,
      "learning_rate": 2.1065487863764787e-07,
      "loss": 0.7622894,
      "num_input_tokens_seen": 154228385,
      "step": 7127,
      "time_per_iteration": 2.73722767829895
    },
    {
      "auxiliary_loss_clip": 0.01157449,
      "auxiliary_loss_mlp": 0.01031043,
      "balance_loss_clip": 1.04894829,
      "balance_loss_mlp": 1.02310395,
      "epoch": 0.8570913244754403,
      "flos": 23732285184000.0,
      "grad_norm": 1.6271584924821578,
      "language_loss": 0.86006552,
      "learning_rate": 2.1030703019088846e-07,
      "loss": 0.88195038,
      "num_input_tokens_seen": 154249015,
      "step": 7128,
      "time_per_iteration": 2.705728054046631
    },
    {
      "auxiliary_loss_clip": 0.01143968,
      "auxiliary_loss_mlp": 0.01029178,
      "balance_loss_clip": 1.04615021,
      "balance_loss_mlp": 1.02153635,
      "epoch": 0.8572115673660795,
      "flos": 20048748433920.0,
      "grad_norm": 2.042671760334399,
      "language_loss": 0.71031159,
      "learning_rate": 2.099594532391291e-07,
      "loss": 0.73204303,
      "num_input_tokens_seen": 154267700,
      "step": 7129,
      "time_per_iteration": 2.715683937072754
    },
    {
      "auxiliary_loss_clip": 0.0113516,
      "auxiliary_loss_mlp": 0.01026871,
      "balance_loss_clip": 1.0454886,
      "balance_loss_mlp": 1.01930141,
      "epoch": 0.8573318102567186,
      "flos": 27160389342720.0,
      "grad_norm": 7.140661442943989,
      "language_loss": 0.78865397,
      "learning_rate": 2.0961214783509806e-07,
      "loss": 0.81027424,
      "num_input_tokens_seen": 154290580,
      "step": 7130,
      "time_per_iteration": 2.792890787124634
    },
    {
      "auxiliary_loss_clip": 0.01173556,
      "auxiliary_loss_mlp": 0.01026241,
      "balance_loss_clip": 1.05151606,
      "balance_loss_mlp": 1.01868916,
      "epoch": 0.8574520531473576,
      "flos": 24936585402240.0,
      "grad_norm": 1.9104154743763908,
      "language_loss": 0.74711519,
      "learning_rate": 2.0926511403148051e-07,
      "loss": 0.76911318,
      "num_input_tokens_seen": 154309545,
      "step": 7131,
      "time_per_iteration": 2.596660614013672
    },
    {
      "auxiliary_loss_clip": 0.01093211,
      "auxiliary_loss_mlp": 0.01026309,
      "balance_loss_clip": 1.04421616,
      "balance_loss_mlp": 1.01857221,
      "epoch": 0.8575722960379968,
      "flos": 18771154513920.0,
      "grad_norm": 2.067710020571136,
      "language_loss": 0.75913942,
      "learning_rate": 2.0891835188092143e-07,
      "loss": 0.78033459,
      "num_input_tokens_seen": 154326545,
      "step": 7132,
      "time_per_iteration": 2.803356170654297
    },
    {
      "auxiliary_loss_clip": 0.01110831,
      "auxiliary_loss_mlp": 0.01028478,
      "balance_loss_clip": 1.04176807,
      "balance_loss_mlp": 1.02010393,
      "epoch": 0.8576925389286358,
      "flos": 22200300167040.0,
      "grad_norm": 1.763374523918207,
      "language_loss": 0.8155095,
      "learning_rate": 2.0857186143602434e-07,
      "loss": 0.83690262,
      "num_input_tokens_seen": 154345190,
      "step": 7133,
      "time_per_iteration": 2.720330238342285
    },
    {
      "auxiliary_loss_clip": 0.01147285,
      "auxiliary_loss_mlp": 0.01030384,
      "balance_loss_clip": 1.04798925,
      "balance_loss_mlp": 1.02244496,
      "epoch": 0.8578127818192749,
      "flos": 22894345733760.0,
      "grad_norm": 5.238445290630189,
      "language_loss": 0.67531943,
      "learning_rate": 2.0822564274935094e-07,
      "loss": 0.69709611,
      "num_input_tokens_seen": 154364615,
      "step": 7134,
      "time_per_iteration": 3.868539571762085
    },
    {
      "auxiliary_loss_clip": 0.01143051,
      "auxiliary_loss_mlp": 0.01031471,
      "balance_loss_clip": 1.04822898,
      "balance_loss_mlp": 1.02325141,
      "epoch": 0.8579330247099141,
      "flos": 34824839541120.0,
      "grad_norm": 1.7681984533619322,
      "language_loss": 0.67303491,
      "learning_rate": 2.078796958734239e-07,
      "loss": 0.69478011,
      "num_input_tokens_seen": 154387335,
      "step": 7135,
      "time_per_iteration": 3.7155771255493164
    },
    {
      "auxiliary_loss_clip": 0.01136566,
      "auxiliary_loss_mlp": 0.01027818,
      "balance_loss_clip": 1.04830933,
      "balance_loss_mlp": 1.01973557,
      "epoch": 0.8580532676005531,
      "flos": 19755681367680.0,
      "grad_norm": 2.4696002095694003,
      "language_loss": 0.75119865,
      "learning_rate": 2.0753402086072124e-07,
      "loss": 0.77284253,
      "num_input_tokens_seen": 154405965,
      "step": 7136,
      "time_per_iteration": 2.780000925064087
    },
    {
      "auxiliary_loss_clip": 0.01162764,
      "auxiliary_loss_mlp": 0.01026706,
      "balance_loss_clip": 1.05005896,
      "balance_loss_mlp": 1.01906466,
      "epoch": 0.8581735104911922,
      "flos": 22739318634240.0,
      "grad_norm": 5.509698639317992,
      "language_loss": 0.75200427,
      "learning_rate": 2.071886177636828e-07,
      "loss": 0.77389902,
      "num_input_tokens_seen": 154422750,
      "step": 7137,
      "time_per_iteration": 2.648289918899536
    },
    {
      "auxiliary_loss_clip": 0.0110133,
      "auxiliary_loss_mlp": 0.01024846,
      "balance_loss_clip": 1.03880894,
      "balance_loss_mlp": 1.01675773,
      "epoch": 0.8582937533818313,
      "flos": 23149131880320.0,
      "grad_norm": 1.8301129933510547,
      "language_loss": 0.83836854,
      "learning_rate": 2.0684348663470575e-07,
      "loss": 0.85963029,
      "num_input_tokens_seen": 154442930,
      "step": 7138,
      "time_per_iteration": 3.7219247817993164
    },
    {
      "auxiliary_loss_clip": 0.01092795,
      "auxiliary_loss_mlp": 0.01029992,
      "balance_loss_clip": 1.03634572,
      "balance_loss_mlp": 1.0212481,
      "epoch": 0.8584139962724704,
      "flos": 19498668577920.0,
      "grad_norm": 4.392148130179691,
      "language_loss": 0.61856079,
      "learning_rate": 2.0649862752614555e-07,
      "loss": 0.63978869,
      "num_input_tokens_seen": 154461640,
      "step": 7139,
      "time_per_iteration": 2.9089934825897217
    },
    {
      "auxiliary_loss_clip": 0.01061687,
      "auxiliary_loss_mlp": 0.01000882,
      "balance_loss_clip": 1.00798678,
      "balance_loss_mlp": 0.9999876,
      "epoch": 0.8585342391631094,
      "flos": 71276577788160.0,
      "grad_norm": 1.5487516995751394,
      "language_loss": 0.57052588,
      "learning_rate": 2.0615404049031838e-07,
      "loss": 0.59115154,
      "num_input_tokens_seen": 154518610,
      "step": 7140,
      "time_per_iteration": 3.2022757530212402
    },
    {
      "auxiliary_loss_clip": 0.01087051,
      "auxiliary_loss_mlp": 0.01035328,
      "balance_loss_clip": 1.04499424,
      "balance_loss_mlp": 1.02701283,
      "epoch": 0.8586544820537486,
      "flos": 10815432929280.0,
      "grad_norm": 2.199276942019613,
      "language_loss": 0.78114045,
      "learning_rate": 2.0580972557949616e-07,
      "loss": 0.80236435,
      "num_input_tokens_seen": 154533700,
      "step": 7141,
      "time_per_iteration": 2.844691753387451
    },
    {
      "auxiliary_loss_clip": 0.01005972,
      "auxiliary_loss_mlp": 0.01003789,
      "balance_loss_clip": 1.00892901,
      "balance_loss_mlp": 1.00276363,
      "epoch": 0.8587747249443877,
      "flos": 64811184422400.0,
      "grad_norm": 0.7924792203319945,
      "language_loss": 0.54162443,
      "learning_rate": 2.054656828459125e-07,
      "loss": 0.56172204,
      "num_input_tokens_seen": 154597810,
      "step": 7142,
      "time_per_iteration": 3.457653522491455
    },
    {
      "auxiliary_loss_clip": 0.01144314,
      "auxiliary_loss_mlp": 0.01024787,
      "balance_loss_clip": 1.05057895,
      "balance_loss_mlp": 1.01682663,
      "epoch": 0.8588949678350267,
      "flos": 26834607964800.0,
      "grad_norm": 1.7980067133199866,
      "language_loss": 0.77400988,
      "learning_rate": 2.051219123417578e-07,
      "loss": 0.79570091,
      "num_input_tokens_seen": 154617870,
      "step": 7143,
      "time_per_iteration": 3.3540356159210205
    },
    {
      "auxiliary_loss_clip": 0.01134981,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.04686141,
      "balance_loss_mlp": 1.02020669,
      "epoch": 0.8590152107256659,
      "flos": 26104256726400.0,
      "grad_norm": 2.0747341082831774,
      "language_loss": 0.60123038,
      "learning_rate": 2.0477841411918196e-07,
      "loss": 0.62285972,
      "num_input_tokens_seen": 154637395,
      "step": 7144,
      "time_per_iteration": 2.78983211517334
    },
    {
      "auxiliary_loss_clip": 0.01124571,
      "auxiliary_loss_mlp": 0.01025141,
      "balance_loss_clip": 1.04343855,
      "balance_loss_mlp": 1.01771164,
      "epoch": 0.859135453616305,
      "flos": 26140885620480.0,
      "grad_norm": 2.314380481048608,
      "language_loss": 0.75083452,
      "learning_rate": 2.0443518823029326e-07,
      "loss": 0.77233166,
      "num_input_tokens_seen": 154657935,
      "step": 7145,
      "time_per_iteration": 2.7986836433410645
    },
    {
      "auxiliary_loss_clip": 0.01145997,
      "auxiliary_loss_mlp": 0.01031955,
      "balance_loss_clip": 1.05144286,
      "balance_loss_mlp": 1.02432859,
      "epoch": 0.859255696506944,
      "flos": 12969319046400.0,
      "grad_norm": 2.427031753790481,
      "language_loss": 0.77014959,
      "learning_rate": 2.0409223472715854e-07,
      "loss": 0.79192913,
      "num_input_tokens_seen": 154675080,
      "step": 7146,
      "time_per_iteration": 2.703488826751709
    },
    {
      "auxiliary_loss_clip": 0.01142399,
      "auxiliary_loss_mlp": 0.01024854,
      "balance_loss_clip": 1.05045938,
      "balance_loss_mlp": 1.01741815,
      "epoch": 0.8593759393975832,
      "flos": 18475753063680.0,
      "grad_norm": 2.265262021798428,
      "language_loss": 0.75301588,
      "learning_rate": 2.0374955366180434e-07,
      "loss": 0.77468842,
      "num_input_tokens_seen": 154692720,
      "step": 7147,
      "time_per_iteration": 3.544557809829712
    },
    {
      "auxiliary_loss_clip": 0.01159556,
      "auxiliary_loss_mlp": 0.01025357,
      "balance_loss_clip": 1.04730976,
      "balance_loss_mlp": 1.0172627,
      "epoch": 0.8594961822882222,
      "flos": 22200156512640.0,
      "grad_norm": 1.8114987145514891,
      "language_loss": 0.72587103,
      "learning_rate": 2.034071450862147e-07,
      "loss": 0.74772018,
      "num_input_tokens_seen": 154710190,
      "step": 7148,
      "time_per_iteration": 2.6695773601531982
    },
    {
      "auxiliary_loss_clip": 0.01140439,
      "auxiliary_loss_mlp": 0.0076236,
      "balance_loss_clip": 1.04999197,
      "balance_loss_mlp": 1.00048757,
      "epoch": 0.8596164251788613,
      "flos": 23294749616640.0,
      "grad_norm": 1.8434379109317902,
      "language_loss": 0.76938093,
      "learning_rate": 2.030650090523327e-07,
      "loss": 0.78840888,
      "num_input_tokens_seen": 154729380,
      "step": 7149,
      "time_per_iteration": 2.8108153343200684
    },
    {
      "auxiliary_loss_clip": 0.01140678,
      "auxiliary_loss_mlp": 0.01029484,
      "balance_loss_clip": 1.04886222,
      "balance_loss_mlp": 1.02159262,
      "epoch": 0.8597366680695004,
      "flos": 31649905416960.0,
      "grad_norm": 1.6284905450838176,
      "language_loss": 0.59593922,
      "learning_rate": 2.0272314561205995e-07,
      "loss": 0.61764085,
      "num_input_tokens_seen": 154749775,
      "step": 7150,
      "time_per_iteration": 2.792695999145508
    },
    {
      "auxiliary_loss_clip": 0.01173603,
      "auxiliary_loss_mlp": 0.0102485,
      "balance_loss_clip": 1.05057287,
      "balance_loss_mlp": 1.01772988,
      "epoch": 0.8598569109601395,
      "flos": 21287738211840.0,
      "grad_norm": 1.8053987510458236,
      "language_loss": 0.72570932,
      "learning_rate": 2.023815548172567e-07,
      "loss": 0.74769378,
      "num_input_tokens_seen": 154769845,
      "step": 7151,
      "time_per_iteration": 2.6715242862701416
    },
    {
      "auxiliary_loss_clip": 0.01162009,
      "auxiliary_loss_mlp": 0.01024921,
      "balance_loss_clip": 1.05277586,
      "balance_loss_mlp": 1.0172503,
      "epoch": 0.8599771538507786,
      "flos": 25447809720960.0,
      "grad_norm": 1.6030349788772615,
      "language_loss": 0.6619637,
      "learning_rate": 2.0204023671974267e-07,
      "loss": 0.683833,
      "num_input_tokens_seen": 154789230,
      "step": 7152,
      "time_per_iteration": 2.6870296001434326
    },
    {
      "auxiliary_loss_clip": 0.01113704,
      "auxiliary_loss_mlp": 0.01030966,
      "balance_loss_clip": 1.04236007,
      "balance_loss_mlp": 1.02311563,
      "epoch": 0.8600973967414177,
      "flos": 16723958768640.0,
      "grad_norm": 2.336605395613481,
      "language_loss": 0.81182373,
      "learning_rate": 2.0169919137129532e-07,
      "loss": 0.83327037,
      "num_input_tokens_seen": 154807670,
      "step": 7153,
      "time_per_iteration": 2.782122850418091
    },
    {
      "auxiliary_loss_clip": 0.01147731,
      "auxiliary_loss_mlp": 0.01032408,
      "balance_loss_clip": 1.04780626,
      "balance_loss_mlp": 1.02367628,
      "epoch": 0.8602176396320568,
      "flos": 25227928615680.0,
      "grad_norm": 5.3129601211741,
      "language_loss": 0.70977211,
      "learning_rate": 2.013584188236508e-07,
      "loss": 0.73157346,
      "num_input_tokens_seen": 154825575,
      "step": 7154,
      "time_per_iteration": 2.7104334831237793
    },
    {
      "auxiliary_loss_clip": 0.01143607,
      "auxiliary_loss_mlp": 0.01029066,
      "balance_loss_clip": 1.04977489,
      "balance_loss_mlp": 1.02094185,
      "epoch": 0.8603378825226958,
      "flos": 20412236113920.0,
      "grad_norm": 1.7471383479180285,
      "language_loss": 0.79703593,
      "learning_rate": 2.0101791912850396e-07,
      "loss": 0.81876266,
      "num_input_tokens_seen": 154845115,
      "step": 7155,
      "time_per_iteration": 2.706249713897705
    },
    {
      "auxiliary_loss_clip": 0.01143001,
      "auxiliary_loss_mlp": 0.01025174,
      "balance_loss_clip": 1.04995441,
      "balance_loss_mlp": 1.01757419,
      "epoch": 0.8604581254133349,
      "flos": 34930201109760.0,
      "grad_norm": 2.7272145862065256,
      "language_loss": 0.64337027,
      "learning_rate": 2.006776923375082e-07,
      "loss": 0.66505206,
      "num_input_tokens_seen": 154866770,
      "step": 7156,
      "time_per_iteration": 2.797316312789917
    },
    {
      "auxiliary_loss_clip": 0.01125696,
      "auxiliary_loss_mlp": 0.01028449,
      "balance_loss_clip": 1.04376101,
      "balance_loss_mlp": 1.02094531,
      "epoch": 0.860578368303974,
      "flos": 22596538072320.0,
      "grad_norm": 1.731088237337847,
      "language_loss": 0.71278977,
      "learning_rate": 2.003377385022764e-07,
      "loss": 0.73433125,
      "num_input_tokens_seen": 154885595,
      "step": 7157,
      "time_per_iteration": 2.7722175121307373
    },
    {
      "auxiliary_loss_clip": 0.01145975,
      "auxiliary_loss_mlp": 0.01026196,
      "balance_loss_clip": 1.05022359,
      "balance_loss_mlp": 1.01838827,
      "epoch": 0.8606986111946131,
      "flos": 21324331192320.0,
      "grad_norm": 2.2714963975543503,
      "language_loss": 0.77682734,
      "learning_rate": 1.9999805767437826e-07,
      "loss": 0.79854906,
      "num_input_tokens_seen": 154904485,
      "step": 7158,
      "time_per_iteration": 2.6677658557891846
    },
    {
      "auxiliary_loss_clip": 0.01170247,
      "auxiliary_loss_mlp": 0.00761944,
      "balance_loss_clip": 1.04982936,
      "balance_loss_mlp": 1.00054598,
      "epoch": 0.8608188540852522,
      "flos": 28877206769280.0,
      "grad_norm": 1.9253739630189461,
      "language_loss": 0.71937656,
      "learning_rate": 1.9965864990534386e-07,
      "loss": 0.73869848,
      "num_input_tokens_seen": 154925010,
      "step": 7159,
      "time_per_iteration": 2.685868263244629
    },
    {
      "auxiliary_loss_clip": 0.01146648,
      "auxiliary_loss_mlp": 0.01031514,
      "balance_loss_clip": 1.04862309,
      "balance_loss_mlp": 1.02434647,
      "epoch": 0.8609390969758913,
      "flos": 29716187713920.0,
      "grad_norm": 1.7517271606774667,
      "language_loss": 0.77879202,
      "learning_rate": 1.9931951524666092e-07,
      "loss": 0.80057359,
      "num_input_tokens_seen": 154946100,
      "step": 7160,
      "time_per_iteration": 3.8336257934570312
    },
    {
      "auxiliary_loss_clip": 0.01162185,
      "auxiliary_loss_mlp": 0.01025167,
      "balance_loss_clip": 1.04970443,
      "balance_loss_mlp": 1.01760364,
      "epoch": 0.8610593398665304,
      "flos": 21249349551360.0,
      "grad_norm": 1.661769656590237,
      "language_loss": 0.8131423,
      "learning_rate": 1.9898065374977534e-07,
      "loss": 0.83501577,
      "num_input_tokens_seen": 154966305,
      "step": 7161,
      "time_per_iteration": 4.317176818847656
    },
    {
      "auxiliary_loss_clip": 0.0112919,
      "auxiliary_loss_mlp": 0.00760936,
      "balance_loss_clip": 1.04819882,
      "balance_loss_mlp": 1.00059795,
      "epoch": 0.8611795827571694,
      "flos": 14830102183680.0,
      "grad_norm": 2.8175467262817624,
      "language_loss": 0.73149252,
      "learning_rate": 1.9864206546609342e-07,
      "loss": 0.75039381,
      "num_input_tokens_seen": 154985145,
      "step": 7162,
      "time_per_iteration": 2.767098903656006
    },
    {
      "auxiliary_loss_clip": 0.01160006,
      "auxiliary_loss_mlp": 0.01026862,
      "balance_loss_clip": 1.04891169,
      "balance_loss_mlp": 1.01890469,
      "epoch": 0.8612998256478086,
      "flos": 24243258107520.0,
      "grad_norm": 2.6883117759793005,
      "language_loss": 0.84219944,
      "learning_rate": 1.983037504469771e-07,
      "loss": 0.86406815,
      "num_input_tokens_seen": 155003855,
      "step": 7163,
      "time_per_iteration": 2.718294143676758
    },
    {
      "auxiliary_loss_clip": 0.01136571,
      "auxiliary_loss_mlp": 0.01028388,
      "balance_loss_clip": 1.04833603,
      "balance_loss_mlp": 1.02095819,
      "epoch": 0.8614200685384477,
      "flos": 21252653602560.0,
      "grad_norm": 1.857007363895373,
      "language_loss": 0.66829127,
      "learning_rate": 1.9796570874374984e-07,
      "loss": 0.68994087,
      "num_input_tokens_seen": 155023960,
      "step": 7164,
      "time_per_iteration": 2.7571475505828857
    },
    {
      "auxiliary_loss_clip": 0.01108458,
      "auxiliary_loss_mlp": 0.01028391,
      "balance_loss_clip": 1.04516101,
      "balance_loss_mlp": 1.02070248,
      "epoch": 0.8615403114290867,
      "flos": 20007738080640.0,
      "grad_norm": 2.4645673158699246,
      "language_loss": 0.77587336,
      "learning_rate": 1.976279404076917e-07,
      "loss": 0.79724187,
      "num_input_tokens_seen": 155043360,
      "step": 7165,
      "time_per_iteration": 3.7735376358032227
    },
    {
      "auxiliary_loss_clip": 0.01172694,
      "auxiliary_loss_mlp": 0.01025687,
      "balance_loss_clip": 1.05037308,
      "balance_loss_mlp": 1.01771152,
      "epoch": 0.8616605543197259,
      "flos": 29789373674880.0,
      "grad_norm": 3.1163663071474534,
      "language_loss": 0.76434869,
      "learning_rate": 1.9729044549004193e-07,
      "loss": 0.78633249,
      "num_input_tokens_seen": 155064745,
      "step": 7166,
      "time_per_iteration": 2.6713972091674805
    },
    {
      "auxiliary_loss_clip": 0.01149413,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 1.04733896,
      "balance_loss_mlp": 1.01954937,
      "epoch": 0.8617807972103649,
      "flos": 28911609020160.0,
      "grad_norm": 1.536318634303914,
      "language_loss": 0.70159048,
      "learning_rate": 1.9695322404199822e-07,
      "loss": 0.72335565,
      "num_input_tokens_seen": 155086790,
      "step": 7167,
      "time_per_iteration": 2.8195672035217285
    },
    {
      "auxiliary_loss_clip": 0.01145334,
      "auxiliary_loss_mlp": 0.00762284,
      "balance_loss_clip": 1.0491271,
      "balance_loss_mlp": 1.00038433,
      "epoch": 0.861901040101004,
      "flos": 27673804391040.0,
      "grad_norm": 2.118066593337538,
      "language_loss": 0.8205446,
      "learning_rate": 1.9661627611471654e-07,
      "loss": 0.83962077,
      "num_input_tokens_seen": 155106585,
      "step": 7168,
      "time_per_iteration": 2.7694733142852783
    },
    {
      "auxiliary_loss_clip": 0.01116276,
      "auxiliary_loss_mlp": 0.01033565,
      "balance_loss_clip": 1.04000604,
      "balance_loss_mlp": 1.02507746,
      "epoch": 0.8620212829916432,
      "flos": 49748056755840.0,
      "grad_norm": 3.9303234173282884,
      "language_loss": 0.70032477,
      "learning_rate": 1.9627960175931246e-07,
      "loss": 0.72182322,
      "num_input_tokens_seen": 155131285,
      "step": 7169,
      "time_per_iteration": 2.9810428619384766
    },
    {
      "auxiliary_loss_clip": 0.01113606,
      "auxiliary_loss_mlp": 0.01026849,
      "balance_loss_clip": 1.04645741,
      "balance_loss_mlp": 1.01933289,
      "epoch": 0.8621415258822822,
      "flos": 21138672769920.0,
      "grad_norm": 1.9683299778344878,
      "language_loss": 0.74529415,
      "learning_rate": 1.9594320102685847e-07,
      "loss": 0.76669872,
      "num_input_tokens_seen": 155150555,
      "step": 7170,
      "time_per_iteration": 2.774574041366577
    },
    {
      "auxiliary_loss_clip": 0.01139818,
      "auxiliary_loss_mlp": 0.01032208,
      "balance_loss_clip": 1.05042481,
      "balance_loss_mlp": 1.02482629,
      "epoch": 0.8622617687729213,
      "flos": 21689039934720.0,
      "grad_norm": 2.577107329252987,
      "language_loss": 0.64554799,
      "learning_rate": 1.956070739683864e-07,
      "loss": 0.66726822,
      "num_input_tokens_seen": 155169890,
      "step": 7171,
      "time_per_iteration": 2.7725601196289062
    },
    {
      "auxiliary_loss_clip": 0.01158368,
      "auxiliary_loss_mlp": 0.01027044,
      "balance_loss_clip": 1.04955447,
      "balance_loss_mlp": 1.0195961,
      "epoch": 0.8623820116635604,
      "flos": 26250592734720.0,
      "grad_norm": 1.5783219387484497,
      "language_loss": 0.74112737,
      "learning_rate": 1.9527122063488678e-07,
      "loss": 0.76298153,
      "num_input_tokens_seen": 155191005,
      "step": 7172,
      "time_per_iteration": 2.7022273540496826
    },
    {
      "auxiliary_loss_clip": 0.01107962,
      "auxiliary_loss_mlp": 0.01023993,
      "balance_loss_clip": 1.03917766,
      "balance_loss_mlp": 1.0162499,
      "epoch": 0.8625022545541995,
      "flos": 19647554451840.0,
      "grad_norm": 1.862090195871872,
      "language_loss": 0.80514145,
      "learning_rate": 1.9493564107730755e-07,
      "loss": 0.82646108,
      "num_input_tokens_seen": 155211005,
      "step": 7173,
      "time_per_iteration": 2.7279157638549805
    },
    {
      "auxiliary_loss_clip": 0.01126411,
      "auxiliary_loss_mlp": 0.00761494,
      "balance_loss_clip": 1.04473507,
      "balance_loss_mlp": 1.00051093,
      "epoch": 0.8626224974448385,
      "flos": 21908382336000.0,
      "grad_norm": 3.0027241273882677,
      "language_loss": 0.614021,
      "learning_rate": 1.9460033534655684e-07,
      "loss": 0.63290012,
      "num_input_tokens_seen": 155230365,
      "step": 7174,
      "time_per_iteration": 3.6668732166290283
    },
    {
      "auxiliary_loss_clip": 0.01142322,
      "auxiliary_loss_mlp": 0.01028328,
      "balance_loss_clip": 1.04809666,
      "balance_loss_mlp": 1.02090442,
      "epoch": 0.8627427403354777,
      "flos": 23331198942720.0,
      "grad_norm": 1.5904859153667694,
      "language_loss": 0.84456849,
      "learning_rate": 1.9426530349349978e-07,
      "loss": 0.86627501,
      "num_input_tokens_seen": 155250815,
      "step": 7175,
      "time_per_iteration": 2.766364812850952
    },
    {
      "auxiliary_loss_clip": 0.01127227,
      "auxiliary_loss_mlp": 0.010288,
      "balance_loss_clip": 1.04502559,
      "balance_loss_mlp": 1.0215168,
      "epoch": 0.8628629832261168,
      "flos": 16362877299840.0,
      "grad_norm": 1.9635457486409258,
      "language_loss": 0.65044218,
      "learning_rate": 1.9393054556896038e-07,
      "loss": 0.67200249,
      "num_input_tokens_seen": 155268515,
      "step": 7176,
      "time_per_iteration": 2.7472918033599854
    },
    {
      "auxiliary_loss_clip": 0.01174724,
      "auxiliary_loss_mlp": 0.01026409,
      "balance_loss_clip": 1.05060291,
      "balance_loss_mlp": 1.01883292,
      "epoch": 0.8629832261167558,
      "flos": 28103941756800.0,
      "grad_norm": 2.320994348970641,
      "language_loss": 0.69545925,
      "learning_rate": 1.9359606162372133e-07,
      "loss": 0.71747053,
      "num_input_tokens_seen": 155290120,
      "step": 7177,
      "time_per_iteration": 2.741365909576416
    },
    {
      "auxiliary_loss_clip": 0.01131768,
      "auxiliary_loss_mlp": 0.01024874,
      "balance_loss_clip": 1.0478313,
      "balance_loss_mlp": 1.01724768,
      "epoch": 0.863103469007395,
      "flos": 20230061310720.0,
      "grad_norm": 1.8575611216670784,
      "language_loss": 0.70738274,
      "learning_rate": 1.9326185170852293e-07,
      "loss": 0.72894913,
      "num_input_tokens_seen": 155309085,
      "step": 7178,
      "time_per_iteration": 2.6801187992095947
    },
    {
      "auxiliary_loss_clip": 0.01138709,
      "auxiliary_loss_mlp": 0.01027445,
      "balance_loss_clip": 1.04940844,
      "balance_loss_mlp": 1.01946104,
      "epoch": 0.863223711898034,
      "flos": 24498547044480.0,
      "grad_norm": 2.4842235595431714,
      "language_loss": 0.72426492,
      "learning_rate": 1.9292791587406598e-07,
      "loss": 0.74592644,
      "num_input_tokens_seen": 155327945,
      "step": 7179,
      "time_per_iteration": 2.783778429031372
    },
    {
      "auxiliary_loss_clip": 0.0111172,
      "auxiliary_loss_mlp": 0.01027131,
      "balance_loss_clip": 1.04236054,
      "balance_loss_mlp": 1.01906085,
      "epoch": 0.8633439547886731,
      "flos": 17675376261120.0,
      "grad_norm": 7.0605325836533925,
      "language_loss": 0.86988997,
      "learning_rate": 1.9259425417100661e-07,
      "loss": 0.89127851,
      "num_input_tokens_seen": 155344060,
      "step": 7180,
      "time_per_iteration": 2.7203943729400635
    },
    {
      "auxiliary_loss_clip": 0.01126657,
      "auxiliary_loss_mlp": 0.01025587,
      "balance_loss_clip": 1.04438543,
      "balance_loss_mlp": 1.01770103,
      "epoch": 0.8634641976793123,
      "flos": 12895055677440.0,
      "grad_norm": 2.275812973196516,
      "language_loss": 0.75285912,
      "learning_rate": 1.9226086664996234e-07,
      "loss": 0.77438158,
      "num_input_tokens_seen": 155362305,
      "step": 7181,
      "time_per_iteration": 2.741023302078247
    },
    {
      "auxiliary_loss_clip": 0.01127048,
      "auxiliary_loss_mlp": 0.01028291,
      "balance_loss_clip": 1.04395723,
      "balance_loss_mlp": 1.02049494,
      "epoch": 0.8635844405699513,
      "flos": 23878980328320.0,
      "grad_norm": 2.3297645284751884,
      "language_loss": 0.74591541,
      "learning_rate": 1.9192775336150712e-07,
      "loss": 0.76746881,
      "num_input_tokens_seen": 155382605,
      "step": 7182,
      "time_per_iteration": 2.7800166606903076
    },
    {
      "auxiliary_loss_clip": 0.01061795,
      "auxiliary_loss_mlp": 0.01001672,
      "balance_loss_clip": 1.00784707,
      "balance_loss_mlp": 1.00079012,
      "epoch": 0.8637046834605904,
      "flos": 60453387521280.0,
      "grad_norm": 0.7668570159998007,
      "language_loss": 0.56209326,
      "learning_rate": 1.915949143561739e-07,
      "loss": 0.58272797,
      "num_input_tokens_seen": 155437280,
      "step": 7183,
      "time_per_iteration": 3.1892635822296143
    },
    {
      "auxiliary_loss_clip": 0.0113251,
      "auxiliary_loss_mlp": 0.01030958,
      "balance_loss_clip": 1.05004048,
      "balance_loss_mlp": 1.0230093,
      "epoch": 0.8638249263512295,
      "flos": 20558751690240.0,
      "grad_norm": 1.6526403691296734,
      "language_loss": 0.77856231,
      "learning_rate": 1.9126234968445498e-07,
      "loss": 0.80019701,
      "num_input_tokens_seen": 155456970,
      "step": 7184,
      "time_per_iteration": 2.7356581687927246
    },
    {
      "auxiliary_loss_clip": 0.01146843,
      "auxiliary_loss_mlp": 0.01029146,
      "balance_loss_clip": 1.0498445,
      "balance_loss_mlp": 1.02124798,
      "epoch": 0.8639451692418686,
      "flos": 26615768353920.0,
      "grad_norm": 1.4783265464479283,
      "language_loss": 0.67766917,
      "learning_rate": 1.9093005939679884e-07,
      "loss": 0.69942909,
      "num_input_tokens_seen": 155478925,
      "step": 7185,
      "time_per_iteration": 2.7249555587768555
    },
    {
      "auxiliary_loss_clip": 0.01160888,
      "auxiliary_loss_mlp": 0.01028322,
      "balance_loss_clip": 1.04868329,
      "balance_loss_mlp": 1.02077007,
      "epoch": 0.8640654121325076,
      "flos": 15122450977920.0,
      "grad_norm": 1.9001368995673178,
      "language_loss": 0.76946592,
      "learning_rate": 1.9059804354361452e-07,
      "loss": 0.79135799,
      "num_input_tokens_seen": 155496700,
      "step": 7186,
      "time_per_iteration": 3.578303575515747
    },
    {
      "auxiliary_loss_clip": 0.01161696,
      "auxiliary_loss_mlp": 0.00761477,
      "balance_loss_clip": 1.05011702,
      "balance_loss_mlp": 1.00050664,
      "epoch": 0.8641856550231467,
      "flos": 31869068250240.0,
      "grad_norm": 1.8297906744133705,
      "language_loss": 0.70802271,
      "learning_rate": 1.902663021752684e-07,
      "loss": 0.72725445,
      "num_input_tokens_seen": 155518130,
      "step": 7187,
      "time_per_iteration": 3.6914682388305664
    },
    {
      "auxiliary_loss_clip": 0.01147844,
      "auxiliary_loss_mlp": 0.01030469,
      "balance_loss_clip": 1.04839253,
      "balance_loss_mlp": 1.02258372,
      "epoch": 0.8643058979137859,
      "flos": 14976545932800.0,
      "grad_norm": 2.2081251660256704,
      "language_loss": 0.82646841,
      "learning_rate": 1.8993483534208556e-07,
      "loss": 0.84825158,
      "num_input_tokens_seen": 155537040,
      "step": 7188,
      "time_per_iteration": 2.68918776512146
    },
    {
      "auxiliary_loss_clip": 0.01133177,
      "auxiliary_loss_mlp": 0.0102799,
      "balance_loss_clip": 1.04704952,
      "balance_loss_mlp": 1.01994073,
      "epoch": 0.8644261408044249,
      "flos": 13115726881920.0,
      "grad_norm": 3.420567127222973,
      "language_loss": 0.7544564,
      "learning_rate": 1.8960364309434884e-07,
      "loss": 0.77606809,
      "num_input_tokens_seen": 155554535,
      "step": 7189,
      "time_per_iteration": 2.811237335205078
    },
    {
      "auxiliary_loss_clip": 0.01145257,
      "auxiliary_loss_mlp": 0.01027829,
      "balance_loss_clip": 1.04740167,
      "balance_loss_mlp": 1.02046478,
      "epoch": 0.864546383695064,
      "flos": 20850920916480.0,
      "grad_norm": 1.7816431440109217,
      "language_loss": 0.78379798,
      "learning_rate": 1.8927272548229967e-07,
      "loss": 0.80552882,
      "num_input_tokens_seen": 155574225,
      "step": 7190,
      "time_per_iteration": 2.6973021030426025
    },
    {
      "auxiliary_loss_clip": 0.01157581,
      "auxiliary_loss_mlp": 0.01033418,
      "balance_loss_clip": 1.04950404,
      "balance_loss_mlp": 1.02595603,
      "epoch": 0.8646666265857031,
      "flos": 21324582587520.0,
      "grad_norm": 1.8212021105972875,
      "language_loss": 0.83055186,
      "learning_rate": 1.8894208255613876e-07,
      "loss": 0.85246193,
      "num_input_tokens_seen": 155593540,
      "step": 7191,
      "time_per_iteration": 3.664295196533203
    },
    {
      "auxiliary_loss_clip": 0.01151376,
      "auxiliary_loss_mlp": 0.01025722,
      "balance_loss_clip": 1.04670715,
      "balance_loss_mlp": 1.01885247,
      "epoch": 0.8647868694763422,
      "flos": 19750833031680.0,
      "grad_norm": 2.3075518293285353,
      "language_loss": 0.78041911,
      "learning_rate": 1.8861171436602397e-07,
      "loss": 0.80219007,
      "num_input_tokens_seen": 155610655,
      "step": 7192,
      "time_per_iteration": 2.695409059524536
    },
    {
      "auxiliary_loss_clip": 0.01159125,
      "auxiliary_loss_mlp": 0.0103042,
      "balance_loss_clip": 1.05091822,
      "balance_loss_mlp": 1.02239692,
      "epoch": 0.8649071123669813,
      "flos": 26176760328960.0,
      "grad_norm": 2.112856009626497,
      "language_loss": 0.80522394,
      "learning_rate": 1.882816209620719e-07,
      "loss": 0.82711941,
      "num_input_tokens_seen": 155627365,
      "step": 7193,
      "time_per_iteration": 2.6675949096679688
    },
    {
      "auxiliary_loss_clip": 0.01155629,
      "auxiliary_loss_mlp": 0.01029445,
      "balance_loss_clip": 1.0526011,
      "balance_loss_mlp": 1.02120733,
      "epoch": 0.8650273552576204,
      "flos": 20302888135680.0,
      "grad_norm": 1.9242748405154075,
      "language_loss": 0.7681452,
      "learning_rate": 1.8795180239435738e-07,
      "loss": 0.78999591,
      "num_input_tokens_seen": 155646220,
      "step": 7194,
      "time_per_iteration": 2.7223997116088867
    },
    {
      "auxiliary_loss_clip": 0.011459,
      "auxiliary_loss_mlp": 0.01027919,
      "balance_loss_clip": 1.05049479,
      "balance_loss_mlp": 1.02019453,
      "epoch": 0.8651475981482595,
      "flos": 23951088881280.0,
      "grad_norm": 5.243862632486379,
      "language_loss": 0.75745094,
      "learning_rate": 1.8762225871291348e-07,
      "loss": 0.77918917,
      "num_input_tokens_seen": 155662095,
      "step": 7195,
      "time_per_iteration": 2.6750082969665527
    },
    {
      "auxiliary_loss_clip": 0.01133168,
      "auxiliary_loss_mlp": 0.01028579,
      "balance_loss_clip": 1.04735911,
      "balance_loss_mlp": 1.02086604,
      "epoch": 0.8652678410388985,
      "flos": 21684622561920.0,
      "grad_norm": 1.5584676980389587,
      "language_loss": 0.80960774,
      "learning_rate": 1.8729298996773201e-07,
      "loss": 0.83122516,
      "num_input_tokens_seen": 155680845,
      "step": 7196,
      "time_per_iteration": 2.761399507522583
    },
    {
      "auxiliary_loss_clip": 0.01022341,
      "auxiliary_loss_mlp": 0.01002596,
      "balance_loss_clip": 1.00904739,
      "balance_loss_mlp": 1.00168443,
      "epoch": 0.8653880839295377,
      "flos": 65224660855680.0,
      "grad_norm": 0.8389911267477391,
      "language_loss": 0.60797042,
      "learning_rate": 1.8696399620876301e-07,
      "loss": 0.62821978,
      "num_input_tokens_seen": 155737875,
      "step": 7197,
      "time_per_iteration": 3.2276766300201416
    },
    {
      "auxiliary_loss_clip": 0.01133225,
      "auxiliary_loss_mlp": 0.01023987,
      "balance_loss_clip": 1.04581106,
      "balance_loss_mlp": 1.01597595,
      "epoch": 0.8655083268201768,
      "flos": 17749172753280.0,
      "grad_norm": 2.2316004140922874,
      "language_loss": 0.79267526,
      "learning_rate": 1.866352774859141e-07,
      "loss": 0.81424737,
      "num_input_tokens_seen": 155753100,
      "step": 7198,
      "time_per_iteration": 2.68402361869812
    },
    {
      "auxiliary_loss_clip": 0.0116004,
      "auxiliary_loss_mlp": 0.01024809,
      "balance_loss_clip": 1.05126619,
      "balance_loss_mlp": 1.01733816,
      "epoch": 0.8656285697108158,
      "flos": 20703974376960.0,
      "grad_norm": 2.324042769897978,
      "language_loss": 0.69704884,
      "learning_rate": 1.8630683384905188e-07,
      "loss": 0.71889734,
      "num_input_tokens_seen": 155772430,
      "step": 7199,
      "time_per_iteration": 3.632012128829956
    },
    {
      "auxiliary_loss_clip": 0.01133897,
      "auxiliary_loss_mlp": 0.00762272,
      "balance_loss_clip": 1.04886818,
      "balance_loss_mlp": 1.00042915,
      "epoch": 0.865748812601455,
      "flos": 18653833716480.0,
      "grad_norm": 1.9321666094974008,
      "language_loss": 0.88359427,
      "learning_rate": 1.8597866534800045e-07,
      "loss": 0.90255588,
      "num_input_tokens_seen": 155787545,
      "step": 7200,
      "time_per_iteration": 2.6250290870666504
    },
    {
      "auxiliary_loss_clip": 0.0112125,
      "auxiliary_loss_mlp": 0.0103088,
      "balance_loss_clip": 1.04769635,
      "balance_loss_mlp": 1.0227437,
      "epoch": 0.865869055492094,
      "flos": 70652554807680.0,
      "grad_norm": 1.750698089282917,
      "language_loss": 0.74211711,
      "learning_rate": 1.8565077203254398e-07,
      "loss": 0.76363838,
      "num_input_tokens_seen": 155813005,
      "step": 7201,
      "time_per_iteration": 3.131660223007202
    },
    {
      "auxiliary_loss_clip": 0.01112771,
      "auxiliary_loss_mlp": 0.01028717,
      "balance_loss_clip": 1.0445168,
      "balance_loss_mlp": 1.02014005,
      "epoch": 0.8659892983827331,
      "flos": 17383961220480.0,
      "grad_norm": 2.293345784239635,
      "language_loss": 0.72890472,
      "learning_rate": 1.8532315395242203e-07,
      "loss": 0.7503196,
      "num_input_tokens_seen": 155829455,
      "step": 7202,
      "time_per_iteration": 2.715543031692505
    },
    {
      "auxiliary_loss_clip": 0.01137144,
      "auxiliary_loss_mlp": 0.0102165,
      "balance_loss_clip": 1.04413414,
      "balance_loss_mlp": 1.01440167,
      "epoch": 0.8661095412733723,
      "flos": 17895221452800.0,
      "grad_norm": 3.8481570163406476,
      "language_loss": 0.71871769,
      "learning_rate": 1.849958111573353e-07,
      "loss": 0.74030566,
      "num_input_tokens_seen": 155848060,
      "step": 7203,
      "time_per_iteration": 2.741084337234497
    },
    {
      "auxiliary_loss_clip": 0.01137136,
      "auxiliary_loss_mlp": 0.01023172,
      "balance_loss_clip": 1.04621887,
      "balance_loss_mlp": 1.01565909,
      "epoch": 0.8662297841640113,
      "flos": 18224163227520.0,
      "grad_norm": 2.4251998592673254,
      "language_loss": 0.64028502,
      "learning_rate": 1.8466874369694074e-07,
      "loss": 0.66188806,
      "num_input_tokens_seen": 155865755,
      "step": 7204,
      "time_per_iteration": 2.820615291595459
    },
    {
      "auxiliary_loss_clip": 0.0112154,
      "auxiliary_loss_mlp": 0.01027273,
      "balance_loss_clip": 1.04526138,
      "balance_loss_mlp": 1.01996827,
      "epoch": 0.8663500270546504,
      "flos": 16362159027840.0,
      "grad_norm": 2.6511287307709144,
      "language_loss": 0.70303154,
      "learning_rate": 1.843419516208542e-07,
      "loss": 0.72451961,
      "num_input_tokens_seen": 155882680,
      "step": 7205,
      "time_per_iteration": 2.718536615371704
    },
    {
      "auxiliary_loss_clip": 0.01129029,
      "auxiliary_loss_mlp": 0.0102985,
      "balance_loss_clip": 1.04654491,
      "balance_loss_mlp": 1.02064681,
      "epoch": 0.8664702699452895,
      "flos": 17894431353600.0,
      "grad_norm": 2.46142470898649,
      "language_loss": 0.79955781,
      "learning_rate": 1.8401543497865047e-07,
      "loss": 0.82114661,
      "num_input_tokens_seen": 155900680,
      "step": 7206,
      "time_per_iteration": 2.7451727390289307
    },
    {
      "auxiliary_loss_clip": 0.01080202,
      "auxiliary_loss_mlp": 0.00761504,
      "balance_loss_clip": 1.03661323,
      "balance_loss_mlp": 1.00056589,
      "epoch": 0.8665905128359286,
      "flos": 30736373794560.0,
      "grad_norm": 2.4942884717814255,
      "language_loss": 0.64609587,
      "learning_rate": 1.836891938198608e-07,
      "loss": 0.66451299,
      "num_input_tokens_seen": 155921105,
      "step": 7207,
      "time_per_iteration": 2.8572256565093994
    },
    {
      "auxiliary_loss_clip": 0.01175308,
      "auxiliary_loss_mlp": 0.01027886,
      "balance_loss_clip": 1.05312824,
      "balance_loss_mlp": 1.02026224,
      "epoch": 0.8667107557265676,
      "flos": 18656419495680.0,
      "grad_norm": 2.1946351521164007,
      "language_loss": 0.71309626,
      "learning_rate": 1.8336322819397677e-07,
      "loss": 0.73512816,
      "num_input_tokens_seen": 155938640,
      "step": 7208,
      "time_per_iteration": 2.6207222938537598
    },
    {
      "auxiliary_loss_clip": 0.0114981,
      "auxiliary_loss_mlp": 0.01027488,
      "balance_loss_clip": 1.04804921,
      "balance_loss_mlp": 1.01948345,
      "epoch": 0.8668309986172068,
      "flos": 20083725302400.0,
      "grad_norm": 2.2741510169974073,
      "language_loss": 0.6278336,
      "learning_rate": 1.8303753815044654e-07,
      "loss": 0.64960659,
      "num_input_tokens_seen": 155957945,
      "step": 7209,
      "time_per_iteration": 2.6873247623443604
    },
    {
      "auxiliary_loss_clip": 0.0112116,
      "auxiliary_loss_mlp": 0.01028263,
      "balance_loss_clip": 1.04577935,
      "balance_loss_mlp": 1.01975107,
      "epoch": 0.8669512415078459,
      "flos": 21615099788160.0,
      "grad_norm": 7.087768398062984,
      "language_loss": 0.71309316,
      "learning_rate": 1.827121237386773e-07,
      "loss": 0.73458737,
      "num_input_tokens_seen": 155975390,
      "step": 7210,
      "time_per_iteration": 2.7806408405303955
    },
    {
      "auxiliary_loss_clip": 0.01175434,
      "auxiliary_loss_mlp": 0.01027474,
      "balance_loss_clip": 1.05027735,
      "balance_loss_mlp": 1.01905751,
      "epoch": 0.8670714843984849,
      "flos": 17703601372800.0,
      "grad_norm": 2.8021417636367483,
      "language_loss": 0.75394893,
      "learning_rate": 1.8238698500803374e-07,
      "loss": 0.77597797,
      "num_input_tokens_seen": 155988155,
      "step": 7211,
      "time_per_iteration": 2.53387451171875
    },
    {
      "auxiliary_loss_clip": 0.01031273,
      "auxiliary_loss_mlp": 0.01002267,
      "balance_loss_clip": 1.00764382,
      "balance_loss_mlp": 1.00130105,
      "epoch": 0.8671917272891241,
      "flos": 60705483125760.0,
      "grad_norm": 0.7165710968621318,
      "language_loss": 0.56239158,
      "learning_rate": 1.820621220078391e-07,
      "loss": 0.58272702,
      "num_input_tokens_seen": 156052065,
      "step": 7212,
      "time_per_iteration": 4.250887632369995
    },
    {
      "auxiliary_loss_clip": 0.01143299,
      "auxiliary_loss_mlp": 0.01028892,
      "balance_loss_clip": 1.05068946,
      "balance_loss_mlp": 1.0208869,
      "epoch": 0.8673119701797631,
      "flos": 20451881750400.0,
      "grad_norm": 1.6298300357334685,
      "language_loss": 0.67726803,
      "learning_rate": 1.8173753478737553e-07,
      "loss": 0.69898993,
      "num_input_tokens_seen": 156072500,
      "step": 7213,
      "time_per_iteration": 3.621330499649048
    },
    {
      "auxiliary_loss_clip": 0.0115955,
      "auxiliary_loss_mlp": 0.01030644,
      "balance_loss_clip": 1.05007815,
      "balance_loss_mlp": 1.02241302,
      "epoch": 0.8674322130704022,
      "flos": 19647410797440.0,
      "grad_norm": 2.189841101979527,
      "language_loss": 0.80239677,
      "learning_rate": 1.8141322339588205e-07,
      "loss": 0.82429874,
      "num_input_tokens_seen": 156089840,
      "step": 7214,
      "time_per_iteration": 2.6601734161376953
    },
    {
      "auxiliary_loss_clip": 0.01149148,
      "auxiliary_loss_mlp": 0.00761932,
      "balance_loss_clip": 1.04679394,
      "balance_loss_mlp": 1.0005126,
      "epoch": 0.8675524559610414,
      "flos": 26025001367040.0,
      "grad_norm": 2.302660817657439,
      "language_loss": 0.70527411,
      "learning_rate": 1.810891878825569e-07,
      "loss": 0.7243849,
      "num_input_tokens_seen": 156109815,
      "step": 7215,
      "time_per_iteration": 2.7525365352630615
    },
    {
      "auxiliary_loss_clip": 0.01081966,
      "auxiliary_loss_mlp": 0.01028339,
      "balance_loss_clip": 1.03602481,
      "balance_loss_mlp": 1.02069807,
      "epoch": 0.8676726988516804,
      "flos": 15049444584960.0,
      "grad_norm": 2.0694044674214225,
      "language_loss": 0.7218194,
      "learning_rate": 1.8076542829655561e-07,
      "loss": 0.74292248,
      "num_input_tokens_seen": 156128620,
      "step": 7216,
      "time_per_iteration": 2.7666752338409424
    },
    {
      "auxiliary_loss_clip": 0.01125304,
      "auxiliary_loss_mlp": 0.01029684,
      "balance_loss_clip": 1.0444653,
      "balance_loss_mlp": 1.02153635,
      "epoch": 0.8677929417423195,
      "flos": 16288111140480.0,
      "grad_norm": 2.0459697588760326,
      "language_loss": 0.79540122,
      "learning_rate": 1.8044194468699203e-07,
      "loss": 0.8169511,
      "num_input_tokens_seen": 156145930,
      "step": 7217,
      "time_per_iteration": 3.652149200439453
    },
    {
      "auxiliary_loss_clip": 0.01144517,
      "auxiliary_loss_mlp": 0.01027218,
      "balance_loss_clip": 1.04917979,
      "balance_loss_mlp": 1.01956487,
      "epoch": 0.8679131846329585,
      "flos": 18844160906880.0,
      "grad_norm": 6.1408280277292375,
      "language_loss": 0.7580409,
      "learning_rate": 1.8011873710293912e-07,
      "loss": 0.77975827,
      "num_input_tokens_seen": 156164435,
      "step": 7218,
      "time_per_iteration": 2.8684582710266113
    },
    {
      "auxiliary_loss_clip": 0.01161411,
      "auxiliary_loss_mlp": 0.01021891,
      "balance_loss_clip": 1.04990363,
      "balance_loss_mlp": 1.01382101,
      "epoch": 0.8680334275235977,
      "flos": 33620718890880.0,
      "grad_norm": 2.5466890380917055,
      "language_loss": 0.6962136,
      "learning_rate": 1.7979580559342677e-07,
      "loss": 0.71804667,
      "num_input_tokens_seen": 156185165,
      "step": 7219,
      "time_per_iteration": 2.7572641372680664
    },
    {
      "auxiliary_loss_clip": 0.01130582,
      "auxiliary_loss_mlp": 0.01031405,
      "balance_loss_clip": 1.04638743,
      "balance_loss_mlp": 1.02335811,
      "epoch": 0.8681536704142367,
      "flos": 24681152810880.0,
      "grad_norm": 1.6970147671137763,
      "language_loss": 0.67174602,
      "learning_rate": 1.7947315020744358e-07,
      "loss": 0.69336593,
      "num_input_tokens_seen": 156206260,
      "step": 7220,
      "time_per_iteration": 2.795926809310913
    },
    {
      "auxiliary_loss_clip": 0.01124369,
      "auxiliary_loss_mlp": 0.01027378,
      "balance_loss_clip": 1.04125738,
      "balance_loss_mlp": 1.02014828,
      "epoch": 0.8682739133048758,
      "flos": 20011042131840.0,
      "grad_norm": 2.0086357508920067,
      "language_loss": 0.80576313,
      "learning_rate": 1.7915077099393594e-07,
      "loss": 0.82728064,
      "num_input_tokens_seen": 156222860,
      "step": 7221,
      "time_per_iteration": 2.697087049484253
    },
    {
      "auxiliary_loss_clip": 0.01175068,
      "auxiliary_loss_mlp": 0.01026583,
      "balance_loss_clip": 1.05123138,
      "balance_loss_mlp": 1.01787531,
      "epoch": 0.868394156195515,
      "flos": 16654759217280.0,
      "grad_norm": 2.0302802971327467,
      "language_loss": 0.73495638,
      "learning_rate": 1.788286680018083e-07,
      "loss": 0.75697291,
      "num_input_tokens_seen": 156241570,
      "step": 7222,
      "time_per_iteration": 2.656416893005371
    },
    {
      "auxiliary_loss_clip": 0.01150713,
      "auxiliary_loss_mlp": 0.01021233,
      "balance_loss_clip": 1.04774976,
      "balance_loss_mlp": 1.01358008,
      "epoch": 0.868514399086154,
      "flos": 28001381448960.0,
      "grad_norm": 12.395145093256655,
      "language_loss": 0.72667223,
      "learning_rate": 1.7850684127992443e-07,
      "loss": 0.74839169,
      "num_input_tokens_seen": 156261315,
      "step": 7223,
      "time_per_iteration": 2.763974905014038
    },
    {
      "auxiliary_loss_clip": 0.01146894,
      "auxiliary_loss_mlp": 0.01028686,
      "balance_loss_clip": 1.05248702,
      "balance_loss_mlp": 1.02152753,
      "epoch": 0.8686346419767931,
      "flos": 20084587228800.0,
      "grad_norm": 1.5830976077199843,
      "language_loss": 0.7050668,
      "learning_rate": 1.7818529087710378e-07,
      "loss": 0.72682261,
      "num_input_tokens_seen": 156281670,
      "step": 7224,
      "time_per_iteration": 2.684678316116333
    },
    {
      "auxiliary_loss_clip": 0.01162134,
      "auxiliary_loss_mlp": 0.0102678,
      "balance_loss_clip": 1.05008233,
      "balance_loss_mlp": 1.01905799,
      "epoch": 0.8687548848674322,
      "flos": 18223516782720.0,
      "grad_norm": 2.009994267357446,
      "language_loss": 0.84257251,
      "learning_rate": 1.7786401684212637e-07,
      "loss": 0.86446166,
      "num_input_tokens_seen": 156300500,
      "step": 7225,
      "time_per_iteration": 3.5272700786590576
    },
    {
      "auxiliary_loss_clip": 0.01049998,
      "auxiliary_loss_mlp": 0.0100099,
      "balance_loss_clip": 1.00841165,
      "balance_loss_mlp": 1.00013208,
      "epoch": 0.8688751277580713,
      "flos": 70457885049600.0,
      "grad_norm": 0.7309983353531984,
      "language_loss": 0.5583927,
      "learning_rate": 1.7754301922372883e-07,
      "loss": 0.57890254,
      "num_input_tokens_seen": 156350145,
      "step": 7226,
      "time_per_iteration": 3.0990705490112305
    },
    {
      "auxiliary_loss_clip": 0.01143806,
      "auxiliary_loss_mlp": 0.01026748,
      "balance_loss_clip": 1.0444181,
      "balance_loss_mlp": 1.01865935,
      "epoch": 0.8689953706487104,
      "flos": 26906788344960.0,
      "grad_norm": 1.9510748024012512,
      "language_loss": 0.81191915,
      "learning_rate": 1.7722229807060617e-07,
      "loss": 0.83362472,
      "num_input_tokens_seen": 156368725,
      "step": 7227,
      "time_per_iteration": 2.7192599773406982
    },
    {
      "auxiliary_loss_clip": 0.01095208,
      "auxiliary_loss_mlp": 0.0102632,
      "balance_loss_clip": 1.04091084,
      "balance_loss_mlp": 1.01869631,
      "epoch": 0.8691156135393495,
      "flos": 34637385438720.0,
      "grad_norm": 2.0923292275755294,
      "language_loss": 0.81882763,
      "learning_rate": 1.7690185343141172e-07,
      "loss": 0.84004283,
      "num_input_tokens_seen": 156388640,
      "step": 7228,
      "time_per_iteration": 2.8846259117126465
    },
    {
      "auxiliary_loss_clip": 0.011463,
      "auxiliary_loss_mlp": 0.00761189,
      "balance_loss_clip": 1.04677594,
      "balance_loss_mlp": 1.00054622,
      "epoch": 0.8692358564299886,
      "flos": 18989814556800.0,
      "grad_norm": 2.2641311833573625,
      "language_loss": 0.69899917,
      "learning_rate": 1.7658168535475615e-07,
      "loss": 0.71807408,
      "num_input_tokens_seen": 156406425,
      "step": 7229,
      "time_per_iteration": 2.658102035522461
    },
    {
      "auxiliary_loss_clip": 0.01116583,
      "auxiliary_loss_mlp": 0.0076201,
      "balance_loss_clip": 1.04970026,
      "balance_loss_mlp": 1.00041473,
      "epoch": 0.8693560993206276,
      "flos": 30370839039360.0,
      "grad_norm": 1.7580389102512555,
      "language_loss": 0.64544195,
      "learning_rate": 1.7626179388920948e-07,
      "loss": 0.6642279,
      "num_input_tokens_seen": 156427705,
      "step": 7230,
      "time_per_iteration": 2.887739419937134
    },
    {
      "auxiliary_loss_clip": 0.01148289,
      "auxiliary_loss_mlp": 0.0102611,
      "balance_loss_clip": 1.04887283,
      "balance_loss_mlp": 1.01828957,
      "epoch": 0.8694763422112668,
      "flos": 27200430028800.0,
      "grad_norm": 1.7924851327854714,
      "language_loss": 0.80527925,
      "learning_rate": 1.7594217908329866e-07,
      "loss": 0.82702321,
      "num_input_tokens_seen": 156449890,
      "step": 7231,
      "time_per_iteration": 2.781475067138672
    },
    {
      "auxiliary_loss_clip": 0.01096222,
      "auxiliary_loss_mlp": 0.01026265,
      "balance_loss_clip": 1.04315412,
      "balance_loss_mlp": 1.0187192,
      "epoch": 0.8695965851019059,
      "flos": 26139161767680.0,
      "grad_norm": 2.781316555576869,
      "language_loss": 0.74171931,
      "learning_rate": 1.7562284098550895e-07,
      "loss": 0.76294416,
      "num_input_tokens_seen": 156469600,
      "step": 7232,
      "time_per_iteration": 2.877072334289551
    },
    {
      "auxiliary_loss_clip": 0.01024941,
      "auxiliary_loss_mlp": 0.00998814,
      "balance_loss_clip": 1.01659393,
      "balance_loss_mlp": 0.99788409,
      "epoch": 0.8697168279925449,
      "flos": 67332616456320.0,
      "grad_norm": 0.8356497581572125,
      "language_loss": 0.62207133,
      "learning_rate": 1.753037796442838e-07,
      "loss": 0.64230889,
      "num_input_tokens_seen": 156529040,
      "step": 7233,
      "time_per_iteration": 3.2689809799194336
    },
    {
      "auxiliary_loss_clip": 0.01086231,
      "auxiliary_loss_mlp": 0.01024443,
      "balance_loss_clip": 1.04051709,
      "balance_loss_mlp": 1.01637256,
      "epoch": 0.8698370708831841,
      "flos": 19718693337600.0,
      "grad_norm": 2.2803828226299574,
      "language_loss": 0.75426096,
      "learning_rate": 1.74984995108024e-07,
      "loss": 0.77536774,
      "num_input_tokens_seen": 156546970,
      "step": 7234,
      "time_per_iteration": 2.81916880607605
    },
    {
      "auxiliary_loss_clip": 0.01134865,
      "auxiliary_loss_mlp": 0.01025139,
      "balance_loss_clip": 1.04933953,
      "balance_loss_mlp": 1.01697278,
      "epoch": 0.8699573137738231,
      "flos": 12859971068160.0,
      "grad_norm": 9.555506753258593,
      "language_loss": 0.83404386,
      "learning_rate": 1.7466648742508981e-07,
      "loss": 0.85564393,
      "num_input_tokens_seen": 156563155,
      "step": 7235,
      "time_per_iteration": 2.691868305206299
    },
    {
      "auxiliary_loss_clip": 0.01124223,
      "auxiliary_loss_mlp": 0.01023762,
      "balance_loss_clip": 1.0450604,
      "balance_loss_mlp": 1.01656199,
      "epoch": 0.8700775566644622,
      "flos": 17420733768960.0,
      "grad_norm": 1.941828160628713,
      "language_loss": 0.84876382,
      "learning_rate": 1.7434825664379837e-07,
      "loss": 0.87024373,
      "num_input_tokens_seen": 156581660,
      "step": 7236,
      "time_per_iteration": 2.788149118423462
    },
    {
      "auxiliary_loss_clip": 0.01160466,
      "auxiliary_loss_mlp": 0.01028742,
      "balance_loss_clip": 1.04981136,
      "balance_loss_mlp": 1.02071619,
      "epoch": 0.8701977995551013,
      "flos": 13735221770880.0,
      "grad_norm": 2.610842542083833,
      "language_loss": 0.8631078,
      "learning_rate": 1.740303028124246e-07,
      "loss": 0.88499987,
      "num_input_tokens_seen": 156597720,
      "step": 7237,
      "time_per_iteration": 2.6850359439849854
    },
    {
      "auxiliary_loss_clip": 0.01157672,
      "auxiliary_loss_mlp": 0.01027333,
      "balance_loss_clip": 1.04970813,
      "balance_loss_mlp": 1.01932788,
      "epoch": 0.8703180424457404,
      "flos": 30555707362560.0,
      "grad_norm": 1.9050170355513993,
      "language_loss": 0.75687879,
      "learning_rate": 1.7371262597920212e-07,
      "loss": 0.77872884,
      "num_input_tokens_seen": 156619780,
      "step": 7238,
      "time_per_iteration": 3.6333343982696533
    },
    {
      "auxiliary_loss_clip": 0.01129233,
      "auxiliary_loss_mlp": 0.01034856,
      "balance_loss_clip": 1.04726791,
      "balance_loss_mlp": 1.02722955,
      "epoch": 0.8704382853363795,
      "flos": 19608986223360.0,
      "grad_norm": 1.5350451497328466,
      "language_loss": 0.76572514,
      "learning_rate": 1.7339522619232195e-07,
      "loss": 0.78736603,
      "num_input_tokens_seen": 156638160,
      "step": 7239,
      "time_per_iteration": 3.656190872192383
    },
    {
      "auxiliary_loss_clip": 0.01123861,
      "auxiliary_loss_mlp": 0.01032541,
      "balance_loss_clip": 1.04688048,
      "balance_loss_mlp": 1.02435756,
      "epoch": 0.8705585282270186,
      "flos": 26613900846720.0,
      "grad_norm": 1.9593384555740248,
      "language_loss": 0.75517768,
      "learning_rate": 1.730781034999338e-07,
      "loss": 0.77674168,
      "num_input_tokens_seen": 156659740,
      "step": 7240,
      "time_per_iteration": 2.7705562114715576
    },
    {
      "auxiliary_loss_clip": 0.01141124,
      "auxiliary_loss_mlp": 0.01024508,
      "balance_loss_clip": 1.04935169,
      "balance_loss_mlp": 1.01728082,
      "epoch": 0.8706787711176577,
      "flos": 34090465979520.0,
      "grad_norm": 1.9711573983220585,
      "language_loss": 0.73434037,
      "learning_rate": 1.7276125795014497e-07,
      "loss": 0.7559967,
      "num_input_tokens_seen": 156678190,
      "step": 7241,
      "time_per_iteration": 2.805898666381836
    },
    {
      "auxiliary_loss_clip": 0.01134991,
      "auxiliary_loss_mlp": 0.00762571,
      "balance_loss_clip": 1.04857099,
      "balance_loss_mlp": 1.00046253,
      "epoch": 0.8707990140082967,
      "flos": 14611513968000.0,
      "grad_norm": 3.7631127918428957,
      "language_loss": 0.67565906,
      "learning_rate": 1.7244468959102054e-07,
      "loss": 0.69463468,
      "num_input_tokens_seen": 156695245,
      "step": 7242,
      "time_per_iteration": 2.728208541870117
    },
    {
      "auxiliary_loss_clip": 0.01146014,
      "auxiliary_loss_mlp": 0.01030925,
      "balance_loss_clip": 1.04996789,
      "balance_loss_mlp": 1.02291393,
      "epoch": 0.8709192568989359,
      "flos": 20084156265600.0,
      "grad_norm": 2.5352175676012534,
      "language_loss": 0.85406768,
      "learning_rate": 1.7212839847058348e-07,
      "loss": 0.87583715,
      "num_input_tokens_seen": 156710375,
      "step": 7243,
      "time_per_iteration": 3.6276168823242188
    },
    {
      "auxiliary_loss_clip": 0.01137126,
      "auxiliary_loss_mlp": 0.01024305,
      "balance_loss_clip": 1.04671359,
      "balance_loss_mlp": 1.01669967,
      "epoch": 0.871039499789575,
      "flos": 16727083251840.0,
      "grad_norm": 2.145789227920685,
      "language_loss": 0.73822522,
      "learning_rate": 1.718123846368147e-07,
      "loss": 0.75983953,
      "num_input_tokens_seen": 156729420,
      "step": 7244,
      "time_per_iteration": 2.7543232440948486
    },
    {
      "auxiliary_loss_clip": 0.01144617,
      "auxiliary_loss_mlp": 0.01022743,
      "balance_loss_clip": 1.0461967,
      "balance_loss_mlp": 1.01559043,
      "epoch": 0.871159742680214,
      "flos": 21068790860160.0,
      "grad_norm": 1.6834107068569355,
      "language_loss": 0.71642131,
      "learning_rate": 1.714966481376543e-07,
      "loss": 0.73809493,
      "num_input_tokens_seen": 156746100,
      "step": 7245,
      "time_per_iteration": 2.721273899078369
    },
    {
      "auxiliary_loss_clip": 0.01105896,
      "auxiliary_loss_mlp": 0.01026537,
      "balance_loss_clip": 1.04300535,
      "balance_loss_mlp": 1.01873529,
      "epoch": 0.8712799855708532,
      "flos": 28256526731520.0,
      "grad_norm": 2.1096922356010643,
      "language_loss": 0.83224595,
      "learning_rate": 1.7118118902099797e-07,
      "loss": 0.85357028,
      "num_input_tokens_seen": 156764185,
      "step": 7246,
      "time_per_iteration": 2.825852394104004
    },
    {
      "auxiliary_loss_clip": 0.01126833,
      "auxiliary_loss_mlp": 0.01030792,
      "balance_loss_clip": 1.04487538,
      "balance_loss_mlp": 1.02343404,
      "epoch": 0.8714002284614922,
      "flos": 22236677665920.0,
      "grad_norm": 1.6597206149079549,
      "language_loss": 0.80890501,
      "learning_rate": 1.7086600733470146e-07,
      "loss": 0.83048123,
      "num_input_tokens_seen": 156784855,
      "step": 7247,
      "time_per_iteration": 2.8688907623291016
    },
    {
      "auxiliary_loss_clip": 0.01109617,
      "auxiliary_loss_mlp": 0.01026016,
      "balance_loss_clip": 1.04376364,
      "balance_loss_mlp": 1.0185473,
      "epoch": 0.8715204713521313,
      "flos": 21431919404160.0,
      "grad_norm": 3.373525811180215,
      "language_loss": 0.77209461,
      "learning_rate": 1.7055110312657738e-07,
      "loss": 0.79345095,
      "num_input_tokens_seen": 156804350,
      "step": 7248,
      "time_per_iteration": 2.7331721782684326
    },
    {
      "auxiliary_loss_clip": 0.01074028,
      "auxiliary_loss_mlp": 0.01031053,
      "balance_loss_clip": 1.03425932,
      "balance_loss_mlp": 1.02277398,
      "epoch": 0.8716407142427703,
      "flos": 23440439180160.0,
      "grad_norm": 2.5413938777962013,
      "language_loss": 0.74560595,
      "learning_rate": 1.702364764443962e-07,
      "loss": 0.76665682,
      "num_input_tokens_seen": 156823425,
      "step": 7249,
      "time_per_iteration": 2.911924123764038
    },
    {
      "auxiliary_loss_clip": 0.01113801,
      "auxiliary_loss_mlp": 0.01031099,
      "balance_loss_clip": 1.04572487,
      "balance_loss_mlp": 1.02261782,
      "epoch": 0.8717609571334095,
      "flos": 27958683156480.0,
      "grad_norm": 2.1242966258817506,
      "language_loss": 0.72533643,
      "learning_rate": 1.6992212733588685e-07,
      "loss": 0.7467854,
      "num_input_tokens_seen": 156843090,
      "step": 7250,
      "time_per_iteration": 2.988816738128662
    },
    {
      "auxiliary_loss_clip": 0.01124756,
      "auxiliary_loss_mlp": 0.01028072,
      "balance_loss_clip": 1.04848695,
      "balance_loss_mlp": 1.02061212,
      "epoch": 0.8718812000240486,
      "flos": 25479482538240.0,
      "grad_norm": 1.7456615161633828,
      "language_loss": 0.74747729,
      "learning_rate": 1.6960805584873538e-07,
      "loss": 0.7690056,
      "num_input_tokens_seen": 156861090,
      "step": 7251,
      "time_per_iteration": 3.7219769954681396
    },
    {
      "auxiliary_loss_clip": 0.01134496,
      "auxiliary_loss_mlp": 0.01028483,
      "balance_loss_clip": 1.04305053,
      "balance_loss_mlp": 1.02087736,
      "epoch": 0.8720014429146876,
      "flos": 23403056100480.0,
      "grad_norm": 2.149784169277667,
      "language_loss": 0.78453732,
      "learning_rate": 1.6929426203058684e-07,
      "loss": 0.80616713,
      "num_input_tokens_seen": 156881515,
      "step": 7252,
      "time_per_iteration": 2.7183313369750977
    },
    {
      "auxiliary_loss_clip": 0.01164943,
      "auxiliary_loss_mlp": 0.00762491,
      "balance_loss_clip": 1.04791105,
      "balance_loss_mlp": 1.00043142,
      "epoch": 0.8721216858053268,
      "flos": 24352821567360.0,
      "grad_norm": 2.4286857976848713,
      "language_loss": 0.80326295,
      "learning_rate": 1.689807459290431e-07,
      "loss": 0.82253724,
      "num_input_tokens_seen": 156900170,
      "step": 7253,
      "time_per_iteration": 2.656439781188965
    },
    {
      "auxiliary_loss_clip": 0.01096699,
      "auxiliary_loss_mlp": 0.01028698,
      "balance_loss_clip": 1.04271722,
      "balance_loss_mlp": 1.02104449,
      "epoch": 0.8722419286959658,
      "flos": 33869687034240.0,
      "grad_norm": 2.0368034852954704,
      "language_loss": 0.71320677,
      "learning_rate": 1.6866750759166437e-07,
      "loss": 0.73446077,
      "num_input_tokens_seen": 156920150,
      "step": 7254,
      "time_per_iteration": 2.9208734035491943
    },
    {
      "auxiliary_loss_clip": 0.01173997,
      "auxiliary_loss_mlp": 0.01024106,
      "balance_loss_clip": 1.04923701,
      "balance_loss_mlp": 1.01607752,
      "epoch": 0.8723621715866049,
      "flos": 18369385914240.0,
      "grad_norm": 2.4455237793436444,
      "language_loss": 0.77676821,
      "learning_rate": 1.6835454706596865e-07,
      "loss": 0.79874921,
      "num_input_tokens_seen": 156937980,
      "step": 7255,
      "time_per_iteration": 2.5722856521606445
    },
    {
      "auxiliary_loss_clip": 0.01113544,
      "auxiliary_loss_mlp": 0.01034536,
      "balance_loss_clip": 1.04558635,
      "balance_loss_mlp": 1.02678776,
      "epoch": 0.8724824144772441,
      "flos": 22013348855040.0,
      "grad_norm": 1.5804025072491497,
      "language_loss": 0.73689586,
      "learning_rate": 1.680418643994317e-07,
      "loss": 0.7583766,
      "num_input_tokens_seen": 156956550,
      "step": 7256,
      "time_per_iteration": 2.889806032180786
    },
    {
      "auxiliary_loss_clip": 0.0103806,
      "auxiliary_loss_mlp": 0.01004683,
      "balance_loss_clip": 1.00834417,
      "balance_loss_mlp": 1.00373495,
      "epoch": 0.8726026573678831,
      "flos": 66698720213760.0,
      "grad_norm": 0.8983263073843686,
      "language_loss": 0.64444792,
      "learning_rate": 1.6772945963948738e-07,
      "loss": 0.66487539,
      "num_input_tokens_seen": 157014715,
      "step": 7257,
      "time_per_iteration": 3.3230373859405518
    },
    {
      "auxiliary_loss_clip": 0.01152251,
      "auxiliary_loss_mlp": 0.01028062,
      "balance_loss_clip": 1.04717958,
      "balance_loss_mlp": 1.01960707,
      "epoch": 0.8727229002585222,
      "flos": 13370908078080.0,
      "grad_norm": 2.170265252036347,
      "language_loss": 0.77547777,
      "learning_rate": 1.6741733283352733e-07,
      "loss": 0.79728091,
      "num_input_tokens_seen": 157032320,
      "step": 7258,
      "time_per_iteration": 2.7095253467559814
    },
    {
      "auxiliary_loss_clip": 0.01137382,
      "auxiliary_loss_mlp": 0.01025166,
      "balance_loss_clip": 1.04636252,
      "balance_loss_mlp": 1.01743507,
      "epoch": 0.8728431431491613,
      "flos": 21796987282560.0,
      "grad_norm": 2.545601324606841,
      "language_loss": 0.84213269,
      "learning_rate": 1.6710548402890102e-07,
      "loss": 0.86375809,
      "num_input_tokens_seen": 157052845,
      "step": 7259,
      "time_per_iteration": 2.7651093006134033
    },
    {
      "auxiliary_loss_clip": 0.01131604,
      "auxiliary_loss_mlp": 0.01023574,
      "balance_loss_clip": 1.0432117,
      "balance_loss_mlp": 1.01499701,
      "epoch": 0.8729633860398004,
      "flos": 36173823742080.0,
      "grad_norm": 2.215920060861705,
      "language_loss": 0.67086107,
      "learning_rate": 1.6679391327291527e-07,
      "loss": 0.69241285,
      "num_input_tokens_seen": 157074050,
      "step": 7260,
      "time_per_iteration": 2.798624277114868
    },
    {
      "auxiliary_loss_clip": 0.01133689,
      "auxiliary_loss_mlp": 0.01027979,
      "balance_loss_clip": 1.04678738,
      "balance_loss_mlp": 1.02082062,
      "epoch": 0.8730836289304394,
      "flos": 16359680989440.0,
      "grad_norm": 2.8395307438849238,
      "language_loss": 0.68404818,
      "learning_rate": 1.6648262061283492e-07,
      "loss": 0.70566487,
      "num_input_tokens_seen": 157089350,
      "step": 7261,
      "time_per_iteration": 2.6684131622314453
    },
    {
      "auxiliary_loss_clip": 0.01156865,
      "auxiliary_loss_mlp": 0.01021954,
      "balance_loss_clip": 1.04858232,
      "balance_loss_mlp": 1.01459575,
      "epoch": 0.8732038718210786,
      "flos": 21215126868480.0,
      "grad_norm": 2.8190012332617553,
      "language_loss": 0.73508036,
      "learning_rate": 1.6617160609588353e-07,
      "loss": 0.7568686,
      "num_input_tokens_seen": 157108525,
      "step": 7262,
      "time_per_iteration": 2.6268584728240967
    },
    {
      "auxiliary_loss_clip": 0.01174208,
      "auxiliary_loss_mlp": 0.01031522,
      "balance_loss_clip": 1.05197048,
      "balance_loss_mlp": 1.02343321,
      "epoch": 0.8733241147117177,
      "flos": 16610696208000.0,
      "grad_norm": 2.966790729005703,
      "language_loss": 0.72669208,
      "learning_rate": 1.6586086976924163e-07,
      "loss": 0.74874938,
      "num_input_tokens_seen": 157124025,
      "step": 7263,
      "time_per_iteration": 3.547071933746338
    },
    {
      "auxiliary_loss_clip": 0.01134831,
      "auxiliary_loss_mlp": 0.01026845,
      "balance_loss_clip": 1.04903495,
      "balance_loss_mlp": 1.01942372,
      "epoch": 0.8734443576023567,
      "flos": 20193935207040.0,
      "grad_norm": 2.096345203861438,
      "language_loss": 0.78486973,
      "learning_rate": 1.6555041168004747e-07,
      "loss": 0.80648649,
      "num_input_tokens_seen": 157143345,
      "step": 7264,
      "time_per_iteration": 3.687664031982422
    },
    {
      "auxiliary_loss_clip": 0.01117054,
      "auxiliary_loss_mlp": 0.01030206,
      "balance_loss_clip": 1.04585218,
      "balance_loss_mlp": 1.02262449,
      "epoch": 0.8735646004929959,
      "flos": 18041162411520.0,
      "grad_norm": 1.9636601268215788,
      "language_loss": 0.69234097,
      "learning_rate": 1.6524023187539715e-07,
      "loss": 0.71381354,
      "num_input_tokens_seen": 157161630,
      "step": 7265,
      "time_per_iteration": 2.701925277709961
    },
    {
      "auxiliary_loss_clip": 0.01137882,
      "auxiliary_loss_mlp": 0.00762196,
      "balance_loss_clip": 1.04769754,
      "balance_loss_mlp": 1.00053382,
      "epoch": 0.873684843383635,
      "flos": 20262344659200.0,
      "grad_norm": 1.761307933847148,
      "language_loss": 0.74675941,
      "learning_rate": 1.649303304023446e-07,
      "loss": 0.76576018,
      "num_input_tokens_seen": 157181385,
      "step": 7266,
      "time_per_iteration": 2.763763666152954
    },
    {
      "auxiliary_loss_clip": 0.01140566,
      "auxiliary_loss_mlp": 0.01029114,
      "balance_loss_clip": 1.04629481,
      "balance_loss_mlp": 1.0216248,
      "epoch": 0.873805086274274,
      "flos": 16947287579520.0,
      "grad_norm": 1.7428370391000128,
      "language_loss": 0.78866023,
      "learning_rate": 1.6462070730790246e-07,
      "loss": 0.81035703,
      "num_input_tokens_seen": 157200545,
      "step": 7267,
      "time_per_iteration": 2.6747491359710693
    },
    {
      "auxiliary_loss_clip": 0.01105463,
      "auxiliary_loss_mlp": 0.01030975,
      "balance_loss_clip": 1.03911316,
      "balance_loss_mlp": 1.02316689,
      "epoch": 0.8739253291649132,
      "flos": 18041270152320.0,
      "grad_norm": 4.44570657617556,
      "language_loss": 0.79142416,
      "learning_rate": 1.6431136263903912e-07,
      "loss": 0.81278849,
      "num_input_tokens_seen": 157219545,
      "step": 7268,
      "time_per_iteration": 2.7907092571258545
    },
    {
      "auxiliary_loss_clip": 0.01145077,
      "auxiliary_loss_mlp": 0.0102889,
      "balance_loss_clip": 1.04595125,
      "balance_loss_mlp": 1.02034307,
      "epoch": 0.8740455720555522,
      "flos": 21325085377920.0,
      "grad_norm": 3.0616433461429757,
      "language_loss": 0.73412317,
      "learning_rate": 1.6400229644268282e-07,
      "loss": 0.75586283,
      "num_input_tokens_seen": 157237900,
      "step": 7269,
      "time_per_iteration": 3.625784397125244
    },
    {
      "auxiliary_loss_clip": 0.01154863,
      "auxiliary_loss_mlp": 0.01023972,
      "balance_loss_clip": 1.04972088,
      "balance_loss_mlp": 1.01597941,
      "epoch": 0.8741658149461913,
      "flos": 15158684822400.0,
      "grad_norm": 2.764666468723766,
      "language_loss": 0.81327718,
      "learning_rate": 1.6369350876571852e-07,
      "loss": 0.8350656,
      "num_input_tokens_seen": 157256055,
      "step": 7270,
      "time_per_iteration": 2.6385538578033447
    },
    {
      "auxiliary_loss_clip": 0.01090204,
      "auxiliary_loss_mlp": 0.01026146,
      "balance_loss_clip": 1.04336762,
      "balance_loss_mlp": 1.01807284,
      "epoch": 0.8742860578368304,
      "flos": 23039855729280.0,
      "grad_norm": 2.1219364644298153,
      "language_loss": 0.8162728,
      "learning_rate": 1.6338499965498874e-07,
      "loss": 0.83743632,
      "num_input_tokens_seen": 157274785,
      "step": 7271,
      "time_per_iteration": 2.843634843826294
    },
    {
      "auxiliary_loss_clip": 0.01125903,
      "auxiliary_loss_mlp": 0.01025194,
      "balance_loss_clip": 1.041044,
      "balance_loss_mlp": 1.01727235,
      "epoch": 0.8744063007274695,
      "flos": 28145347159680.0,
      "grad_norm": 1.5760002129439092,
      "language_loss": 0.77633047,
      "learning_rate": 1.630767691572943e-07,
      "loss": 0.79784143,
      "num_input_tokens_seen": 157294805,
      "step": 7272,
      "time_per_iteration": 2.7387428283691406
    },
    {
      "auxiliary_loss_clip": 0.01041299,
      "auxiliary_loss_mlp": 0.01003056,
      "balance_loss_clip": 1.00780702,
      "balance_loss_mlp": 1.00209033,
      "epoch": 0.8745265436181086,
      "flos": 64034076654720.0,
      "grad_norm": 0.7465541939145158,
      "language_loss": 0.53384364,
      "learning_rate": 1.6276881731939306e-07,
      "loss": 0.5542872,
      "num_input_tokens_seen": 157356695,
      "step": 7273,
      "time_per_iteration": 3.3094301223754883
    },
    {
      "auxiliary_loss_clip": 0.01114314,
      "auxiliary_loss_mlp": 0.0102424,
      "balance_loss_clip": 1.043782,
      "balance_loss_mlp": 1.01680708,
      "epoch": 0.8746467865087477,
      "flos": 28658618553600.0,
      "grad_norm": 2.055679411611374,
      "language_loss": 0.75594425,
      "learning_rate": 1.6246114418800193e-07,
      "loss": 0.7773298,
      "num_input_tokens_seen": 157376975,
      "step": 7274,
      "time_per_iteration": 2.8297762870788574
    },
    {
      "auxiliary_loss_clip": 0.0114373,
      "auxiliary_loss_mlp": 0.0102411,
      "balance_loss_clip": 1.04606009,
      "balance_loss_mlp": 1.01670694,
      "epoch": 0.8747670293993868,
      "flos": 23985850268160.0,
      "grad_norm": 1.787574666100598,
      "language_loss": 0.77013576,
      "learning_rate": 1.6215374980979423e-07,
      "loss": 0.79181415,
      "num_input_tokens_seen": 157397385,
      "step": 7275,
      "time_per_iteration": 2.716691493988037
    },
    {
      "auxiliary_loss_clip": 0.01093668,
      "auxiliary_loss_mlp": 0.01029384,
      "balance_loss_clip": 1.04443526,
      "balance_loss_mlp": 1.02210331,
      "epoch": 0.8748872722900258,
      "flos": 45221624478720.0,
      "grad_norm": 2.329612384527014,
      "language_loss": 0.68903387,
      "learning_rate": 1.6184663423140133e-07,
      "loss": 0.71026438,
      "num_input_tokens_seen": 157417685,
      "step": 7276,
      "time_per_iteration": 3.0339107513427734
    },
    {
      "auxiliary_loss_clip": 0.01136693,
      "auxiliary_loss_mlp": 0.01040226,
      "balance_loss_clip": 1.04820633,
      "balance_loss_mlp": 1.03245997,
      "epoch": 0.875007515180665,
      "flos": 19754280737280.0,
      "grad_norm": 1.9668550332139298,
      "language_loss": 0.64336765,
      "learning_rate": 1.615397974994126e-07,
      "loss": 0.66513681,
      "num_input_tokens_seen": 157435490,
      "step": 7277,
      "time_per_iteration": 3.5833401679992676
    },
    {
      "auxiliary_loss_clip": 0.01146771,
      "auxiliary_loss_mlp": 0.01021253,
      "balance_loss_clip": 1.04869425,
      "balance_loss_mlp": 1.01431489,
      "epoch": 0.875127758071304,
      "flos": 22710734386560.0,
      "grad_norm": 1.7467101820010278,
      "language_loss": 0.80696714,
      "learning_rate": 1.6123323966037438e-07,
      "loss": 0.82864738,
      "num_input_tokens_seen": 157454010,
      "step": 7278,
      "time_per_iteration": 2.7201106548309326
    },
    {
      "auxiliary_loss_clip": 0.01121291,
      "auxiliary_loss_mlp": 0.01027433,
      "balance_loss_clip": 1.05003786,
      "balance_loss_mlp": 1.0200572,
      "epoch": 0.8752480009619431,
      "flos": 23403846199680.0,
      "grad_norm": 2.155980901248034,
      "language_loss": 0.78918278,
      "learning_rate": 1.6092696076079216e-07,
      "loss": 0.81067002,
      "num_input_tokens_seen": 157472385,
      "step": 7279,
      "time_per_iteration": 2.8344979286193848
    },
    {
      "auxiliary_loss_clip": 0.01101525,
      "auxiliary_loss_mlp": 0.01025848,
      "balance_loss_clip": 1.04133511,
      "balance_loss_mlp": 1.01766968,
      "epoch": 0.8753682438525822,
      "flos": 26213101914240.0,
      "grad_norm": 1.9462167832236255,
      "language_loss": 0.73922604,
      "learning_rate": 1.6062096084712785e-07,
      "loss": 0.76049972,
      "num_input_tokens_seen": 157493735,
      "step": 7280,
      "time_per_iteration": 2.803884267807007
    },
    {
      "auxiliary_loss_clip": 0.01111132,
      "auxiliary_loss_mlp": 0.0102269,
      "balance_loss_clip": 1.04170227,
      "balance_loss_mlp": 1.0142796,
      "epoch": 0.8754884867432213,
      "flos": 23326745656320.0,
      "grad_norm": 3.341664844371012,
      "language_loss": 0.70794415,
      "learning_rate": 1.6031523996580098e-07,
      "loss": 0.72928238,
      "num_input_tokens_seen": 157511295,
      "step": 7281,
      "time_per_iteration": 2.803762912750244
    },
    {
      "auxiliary_loss_clip": 0.0115204,
      "auxiliary_loss_mlp": 0.01028228,
      "balance_loss_clip": 1.04954529,
      "balance_loss_mlp": 1.01985955,
      "epoch": 0.8756087296338604,
      "flos": 12495226412160.0,
      "grad_norm": 2.149638644449772,
      "language_loss": 0.66293722,
      "learning_rate": 1.6000979816318981e-07,
      "loss": 0.68473989,
      "num_input_tokens_seen": 157529760,
      "step": 7282,
      "time_per_iteration": 2.639570951461792
    },
    {
      "auxiliary_loss_clip": 0.01158524,
      "auxiliary_loss_mlp": 0.01024902,
      "balance_loss_clip": 1.05024314,
      "balance_loss_mlp": 1.01769841,
      "epoch": 0.8757289725244994,
      "flos": 18952898353920.0,
      "grad_norm": 2.3080348139190496,
      "language_loss": 0.75448179,
      "learning_rate": 1.5970463548562886e-07,
      "loss": 0.77631605,
      "num_input_tokens_seen": 157548915,
      "step": 7283,
      "time_per_iteration": 2.691882848739624
    },
    {
      "auxiliary_loss_clip": 0.01155006,
      "auxiliary_loss_mlp": 0.01029616,
      "balance_loss_clip": 1.05112112,
      "balance_loss_mlp": 1.02254081,
      "epoch": 0.8758492154151386,
      "flos": 25265958140160.0,
      "grad_norm": 2.137074777765674,
      "language_loss": 0.71275902,
      "learning_rate": 1.5939975197941192e-07,
      "loss": 0.73460525,
      "num_input_tokens_seen": 157570570,
      "step": 7284,
      "time_per_iteration": 2.664425849914551
    },
    {
      "auxiliary_loss_clip": 0.01030498,
      "auxiliary_loss_mlp": 0.01001555,
      "balance_loss_clip": 1.00859189,
      "balance_loss_mlp": 1.00064337,
      "epoch": 0.8759694583057777,
      "flos": 65571664193280.0,
      "grad_norm": 0.8080189432369186,
      "language_loss": 0.53321415,
      "learning_rate": 1.5909514769078892e-07,
      "loss": 0.55353469,
      "num_input_tokens_seen": 157635675,
      "step": 7285,
      "time_per_iteration": 3.360034227371216
    },
    {
      "auxiliary_loss_clip": 0.01162469,
      "auxiliary_loss_mlp": 0.01028732,
      "balance_loss_clip": 1.05464673,
      "balance_loss_mlp": 1.02135313,
      "epoch": 0.8760897011964167,
      "flos": 25446193608960.0,
      "grad_norm": 1.5486354583136597,
      "language_loss": 0.77925736,
      "learning_rate": 1.5879082266596867e-07,
      "loss": 0.8011694,
      "num_input_tokens_seen": 157657015,
      "step": 7286,
      "time_per_iteration": 2.70729398727417
    },
    {
      "auxiliary_loss_clip": 0.01156723,
      "auxiliary_loss_mlp": 0.01029148,
      "balance_loss_clip": 1.04851305,
      "balance_loss_mlp": 1.02084494,
      "epoch": 0.8762099440870559,
      "flos": 28984830894720.0,
      "grad_norm": 1.7022685157525212,
      "language_loss": 0.7203685,
      "learning_rate": 1.5848677695111645e-07,
      "loss": 0.7422272,
      "num_input_tokens_seen": 157678615,
      "step": 7287,
      "time_per_iteration": 2.727764129638672
    },
    {
      "auxiliary_loss_clip": 0.01134495,
      "auxiliary_loss_mlp": 0.01033339,
      "balance_loss_clip": 1.04987311,
      "balance_loss_mlp": 1.02485168,
      "epoch": 0.8763301869776949,
      "flos": 21609461352960.0,
      "grad_norm": 2.638512066828975,
      "language_loss": 0.69479847,
      "learning_rate": 1.5818301059235562e-07,
      "loss": 0.7164768,
      "num_input_tokens_seen": 157693790,
      "step": 7288,
      "time_per_iteration": 2.73382830619812
    },
    {
      "auxiliary_loss_clip": 0.0114997,
      "auxiliary_loss_mlp": 0.0102889,
      "balance_loss_clip": 1.04946113,
      "balance_loss_mlp": 1.02089453,
      "epoch": 0.876450429868334,
      "flos": 24644416176000.0,
      "grad_norm": 1.5586642107029287,
      "language_loss": 0.81341016,
      "learning_rate": 1.578795236357684e-07,
      "loss": 0.83519876,
      "num_input_tokens_seen": 157715255,
      "step": 7289,
      "time_per_iteration": 3.657431125640869
    },
    {
      "auxiliary_loss_clip": 0.01134771,
      "auxiliary_loss_mlp": 0.01021974,
      "balance_loss_clip": 1.0492456,
      "balance_loss_mlp": 1.0146184,
      "epoch": 0.8765706727589732,
      "flos": 20260046188800.0,
      "grad_norm": 1.9782695783249489,
      "language_loss": 0.855165,
      "learning_rate": 1.5757631612739218e-07,
      "loss": 0.87673247,
      "num_input_tokens_seen": 157728800,
      "step": 7290,
      "time_per_iteration": 3.607171058654785
    },
    {
      "auxiliary_loss_clip": 0.01060901,
      "auxiliary_loss_mlp": 0.01000686,
      "balance_loss_clip": 1.00729156,
      "balance_loss_mlp": 0.99980336,
      "epoch": 0.8766909156496122,
      "flos": 71371165276800.0,
      "grad_norm": 0.7771238274671045,
      "language_loss": 0.61400425,
      "learning_rate": 1.572733881132242e-07,
      "loss": 0.63462019,
      "num_input_tokens_seen": 157789445,
      "step": 7291,
      "time_per_iteration": 3.232154369354248
    },
    {
      "auxiliary_loss_clip": 0.01039229,
      "auxiliary_loss_mlp": 0.01002236,
      "balance_loss_clip": 1.00784802,
      "balance_loss_mlp": 1.00126398,
      "epoch": 0.8768111585402513,
      "flos": 69523490603520.0,
      "grad_norm": 0.7813087746391901,
      "language_loss": 0.5847832,
      "learning_rate": 1.5697073963921814e-07,
      "loss": 0.60519791,
      "num_input_tokens_seen": 157848685,
      "step": 7292,
      "time_per_iteration": 3.1565091609954834
    },
    {
      "auxiliary_loss_clip": 0.01115162,
      "auxiliary_loss_mlp": 0.01031055,
      "balance_loss_clip": 1.04269385,
      "balance_loss_mlp": 1.02278459,
      "epoch": 0.8769314014308904,
      "flos": 18838558385280.0,
      "grad_norm": 2.685331088048781,
      "language_loss": 0.85037279,
      "learning_rate": 1.566683707512857e-07,
      "loss": 0.87183487,
      "num_input_tokens_seen": 157866360,
      "step": 7293,
      "time_per_iteration": 2.833693027496338
    },
    {
      "auxiliary_loss_clip": 0.01138241,
      "auxiliary_loss_mlp": 0.01026244,
      "balance_loss_clip": 1.05078292,
      "balance_loss_mlp": 1.01752412,
      "epoch": 0.8770516443215295,
      "flos": 14976402278400.0,
      "grad_norm": 2.205855092768923,
      "language_loss": 0.79559994,
      "learning_rate": 1.5636628149529553e-07,
      "loss": 0.81724483,
      "num_input_tokens_seen": 157884150,
      "step": 7294,
      "time_per_iteration": 3.681020736694336
    },
    {
      "auxiliary_loss_clip": 0.01135928,
      "auxiliary_loss_mlp": 0.01028191,
      "balance_loss_clip": 1.04659486,
      "balance_loss_mlp": 1.02094865,
      "epoch": 0.8771718872121685,
      "flos": 31649654021760.0,
      "grad_norm": 2.2742681528088378,
      "language_loss": 0.80085337,
      "learning_rate": 1.560644719170743e-07,
      "loss": 0.82249451,
      "num_input_tokens_seen": 157905020,
      "step": 7295,
      "time_per_iteration": 2.790996789932251
    },
    {
      "auxiliary_loss_clip": 0.01166355,
      "auxiliary_loss_mlp": 0.01036815,
      "balance_loss_clip": 1.04936886,
      "balance_loss_mlp": 1.02805889,
      "epoch": 0.8772921301028077,
      "flos": 36095466222720.0,
      "grad_norm": 1.8113673411584839,
      "language_loss": 0.7235058,
      "learning_rate": 1.5576294206240692e-07,
      "loss": 0.74553752,
      "num_input_tokens_seen": 157924545,
      "step": 7296,
      "time_per_iteration": 2.790539503097534
    },
    {
      "auxiliary_loss_clip": 0.01120369,
      "auxiliary_loss_mlp": 0.01027005,
      "balance_loss_clip": 1.04423189,
      "balance_loss_mlp": 1.01879704,
      "epoch": 0.8774123729934468,
      "flos": 57116961849600.0,
      "grad_norm": 1.6826638910411502,
      "language_loss": 0.67998397,
      "learning_rate": 1.5546169197703507e-07,
      "loss": 0.70145762,
      "num_input_tokens_seen": 157950820,
      "step": 7297,
      "time_per_iteration": 3.080724000930786
    },
    {
      "auxiliary_loss_clip": 0.01143883,
      "auxiliary_loss_mlp": 0.01025934,
      "balance_loss_clip": 1.04946327,
      "balance_loss_mlp": 1.01824808,
      "epoch": 0.8775326158840858,
      "flos": 23914495900800.0,
      "grad_norm": 3.0456205701203056,
      "language_loss": 0.77294731,
      "learning_rate": 1.5516072170665774e-07,
      "loss": 0.79464543,
      "num_input_tokens_seen": 157968790,
      "step": 7298,
      "time_per_iteration": 2.778409957885742
    },
    {
      "auxiliary_loss_clip": 0.01146665,
      "auxiliary_loss_mlp": 0.01032037,
      "balance_loss_clip": 1.05089378,
      "balance_loss_mlp": 1.02462482,
      "epoch": 0.877652858774725,
      "flos": 17123285243520.0,
      "grad_norm": 2.145465116524248,
      "language_loss": 0.86749285,
      "learning_rate": 1.5486003129693214e-07,
      "loss": 0.88927996,
      "num_input_tokens_seen": 157986155,
      "step": 7299,
      "time_per_iteration": 2.6514530181884766
    },
    {
      "auxiliary_loss_clip": 0.01162249,
      "auxiliary_loss_mlp": 0.01028655,
      "balance_loss_clip": 1.05203485,
      "balance_loss_mlp": 1.02088213,
      "epoch": 0.877773101665364,
      "flos": 16508961912960.0,
      "grad_norm": 1.877447723569222,
      "language_loss": 0.78388023,
      "learning_rate": 1.545596207934725e-07,
      "loss": 0.80578935,
      "num_input_tokens_seen": 158004640,
      "step": 7300,
      "time_per_iteration": 2.7214722633361816
    },
    {
      "auxiliary_loss_clip": 0.01157652,
      "auxiliary_loss_mlp": 0.00761423,
      "balance_loss_clip": 1.05011249,
      "balance_loss_mlp": 1.00049567,
      "epoch": 0.8778933445560031,
      "flos": 22053209973120.0,
      "grad_norm": 2.177298018763775,
      "language_loss": 0.78000486,
      "learning_rate": 1.5425949024185147e-07,
      "loss": 0.79919565,
      "num_input_tokens_seen": 158024665,
      "step": 7301,
      "time_per_iteration": 2.696871280670166
    },
    {
      "auxiliary_loss_clip": 0.01147749,
      "auxiliary_loss_mlp": 0.01028774,
      "balance_loss_clip": 1.05042446,
      "balance_loss_mlp": 1.02111495,
      "epoch": 0.8780135874466423,
      "flos": 22564757514240.0,
      "grad_norm": 2.275914914422081,
      "language_loss": 0.67716014,
      "learning_rate": 1.5395963968759818e-07,
      "loss": 0.69892538,
      "num_input_tokens_seen": 158044940,
      "step": 7302,
      "time_per_iteration": 2.7206037044525146
    },
    {
      "auxiliary_loss_clip": 0.01119622,
      "auxiliary_loss_mlp": 0.01032878,
      "balance_loss_clip": 1.04465652,
      "balance_loss_mlp": 1.02491808,
      "epoch": 0.8781338303372813,
      "flos": 61531999073280.0,
      "grad_norm": 1.9781099168071965,
      "language_loss": 0.64569676,
      "learning_rate": 1.536600691761998e-07,
      "loss": 0.66722178,
      "num_input_tokens_seen": 158070770,
      "step": 7303,
      "time_per_iteration": 4.06214165687561
    },
    {
      "auxiliary_loss_clip": 0.01174295,
      "auxiliary_loss_mlp": 0.0103265,
      "balance_loss_clip": 1.05199575,
      "balance_loss_mlp": 1.02484131,
      "epoch": 0.8782540732279204,
      "flos": 22674751937280.0,
      "grad_norm": 2.654220591742918,
      "language_loss": 0.71557307,
      "learning_rate": 1.5336077875310084e-07,
      "loss": 0.73764253,
      "num_input_tokens_seen": 158089995,
      "step": 7304,
      "time_per_iteration": 2.579514741897583
    },
    {
      "auxiliary_loss_clip": 0.01151748,
      "auxiliary_loss_mlp": 0.01030635,
      "balance_loss_clip": 1.04842556,
      "balance_loss_mlp": 1.02211118,
      "epoch": 0.8783743161185595,
      "flos": 16070348937600.0,
      "grad_norm": 2.30925857149256,
      "language_loss": 0.73941255,
      "learning_rate": 1.5306176846370321e-07,
      "loss": 0.76123643,
      "num_input_tokens_seen": 158108140,
      "step": 7305,
      "time_per_iteration": 2.666335344314575
    },
    {
      "auxiliary_loss_clip": 0.01137792,
      "auxiliary_loss_mlp": 0.01025767,
      "balance_loss_clip": 1.04484296,
      "balance_loss_mlp": 1.01779175,
      "epoch": 0.8784945590091986,
      "flos": 26067879227520.0,
      "grad_norm": 2.08513193307838,
      "language_loss": 0.74455833,
      "learning_rate": 1.5276303835336712e-07,
      "loss": 0.76619393,
      "num_input_tokens_seen": 158128680,
      "step": 7306,
      "time_per_iteration": 2.774913787841797
    },
    {
      "auxiliary_loss_clip": 0.01044186,
      "auxiliary_loss_mlp": 0.01000357,
      "balance_loss_clip": 1.01544905,
      "balance_loss_mlp": 0.99936122,
      "epoch": 0.8786148018998376,
      "flos": 62720643939840.0,
      "grad_norm": 0.7614179563042456,
      "language_loss": 0.53409493,
      "learning_rate": 1.524645884674094e-07,
      "loss": 0.55454034,
      "num_input_tokens_seen": 158185610,
      "step": 7307,
      "time_per_iteration": 3.2639107704162598
    },
    {
      "auxiliary_loss_clip": 0.01099755,
      "auxiliary_loss_mlp": 0.01025153,
      "balance_loss_clip": 1.03992951,
      "balance_loss_mlp": 1.01725519,
      "epoch": 0.8787350447904768,
      "flos": 21652734263040.0,
      "grad_norm": 2.0593943892124056,
      "language_loss": 0.79071105,
      "learning_rate": 1.521664188511047e-07,
      "loss": 0.8119601,
      "num_input_tokens_seen": 158205635,
      "step": 7308,
      "time_per_iteration": 2.812842607498169
    },
    {
      "auxiliary_loss_clip": 0.01098028,
      "auxiliary_loss_mlp": 0.01025619,
      "balance_loss_clip": 1.04391682,
      "balance_loss_mlp": 1.0185318,
      "epoch": 0.8788552876811159,
      "flos": 25478476957440.0,
      "grad_norm": 2.0685988984880086,
      "language_loss": 0.80086958,
      "learning_rate": 1.518685295496851e-07,
      "loss": 0.82210612,
      "num_input_tokens_seen": 158223495,
      "step": 7309,
      "time_per_iteration": 2.775596857070923
    },
    {
      "auxiliary_loss_clip": 0.01116867,
      "auxiliary_loss_mlp": 0.01025275,
      "balance_loss_clip": 1.0429368,
      "balance_loss_mlp": 1.01811695,
      "epoch": 0.8789755305717549,
      "flos": 22310222762880.0,
      "grad_norm": 1.863430936391657,
      "language_loss": 0.85684288,
      "learning_rate": 1.5157092060833975e-07,
      "loss": 0.87826431,
      "num_input_tokens_seen": 158243145,
      "step": 7310,
      "time_per_iteration": 2.809077501296997
    },
    {
      "auxiliary_loss_clip": 0.01173043,
      "auxiliary_loss_mlp": 0.01022573,
      "balance_loss_clip": 1.05246341,
      "balance_loss_mlp": 1.01523566,
      "epoch": 0.879095773462394,
      "flos": 29310971408640.0,
      "grad_norm": 1.791360123427297,
      "language_loss": 0.66334915,
      "learning_rate": 1.5127359207221658e-07,
      "loss": 0.68530536,
      "num_input_tokens_seen": 158262625,
      "step": 7311,
      "time_per_iteration": 2.7215652465820312
    },
    {
      "auxiliary_loss_clip": 0.01124368,
      "auxiliary_loss_mlp": 0.01028277,
      "balance_loss_clip": 1.04378641,
      "balance_loss_mlp": 1.02000403,
      "epoch": 0.8792160163530331,
      "flos": 16690023394560.0,
      "grad_norm": 2.061023945214478,
      "language_loss": 0.73420596,
      "learning_rate": 1.5097654398641923e-07,
      "loss": 0.75573242,
      "num_input_tokens_seen": 158280530,
      "step": 7312,
      "time_per_iteration": 2.739384889602661
    },
    {
      "auxiliary_loss_clip": 0.0113124,
      "auxiliary_loss_mlp": 0.01030717,
      "balance_loss_clip": 1.04721904,
      "balance_loss_mlp": 1.02369547,
      "epoch": 0.8793362592436722,
      "flos": 24499301230080.0,
      "grad_norm": 1.714409844345111,
      "language_loss": 0.7325688,
      "learning_rate": 1.5067977639601014e-07,
      "loss": 0.75418836,
      "num_input_tokens_seen": 158303290,
      "step": 7313,
      "time_per_iteration": 2.8041110038757324
    },
    {
      "auxiliary_loss_clip": 0.01155079,
      "auxiliary_loss_mlp": 0.01025103,
      "balance_loss_clip": 1.04643738,
      "balance_loss_mlp": 1.0180161,
      "epoch": 0.8794565021343113,
      "flos": 14538399834240.0,
      "grad_norm": 3.518340074034393,
      "language_loss": 0.71001619,
      "learning_rate": 1.5038328934600864e-07,
      "loss": 0.73181802,
      "num_input_tokens_seen": 158319925,
      "step": 7314,
      "time_per_iteration": 2.6373183727264404
    },
    {
      "auxiliary_loss_clip": 0.01100387,
      "auxiliary_loss_mlp": 0.01024059,
      "balance_loss_clip": 1.04058313,
      "balance_loss_mlp": 1.0164181,
      "epoch": 0.8795767450249504,
      "flos": 39530286224640.0,
      "grad_norm": 1.9408962044967952,
      "language_loss": 0.7005223,
      "learning_rate": 1.5008708288139161e-07,
      "loss": 0.72176671,
      "num_input_tokens_seen": 158342285,
      "step": 7315,
      "time_per_iteration": 3.8287458419799805
    },
    {
      "auxiliary_loss_clip": 0.01150193,
      "auxiliary_loss_mlp": 0.01026459,
      "balance_loss_clip": 1.04953313,
      "balance_loss_mlp": 1.01841795,
      "epoch": 0.8796969879155895,
      "flos": 22960672197120.0,
      "grad_norm": 2.19526294429556,
      "language_loss": 0.73490137,
      "learning_rate": 1.497911570470931e-07,
      "loss": 0.75666785,
      "num_input_tokens_seen": 158362290,
      "step": 7316,
      "time_per_iteration": 3.5696463584899902
    },
    {
      "auxiliary_loss_clip": 0.011711,
      "auxiliary_loss_mlp": 0.01030636,
      "balance_loss_clip": 1.05058086,
      "balance_loss_mlp": 1.02266061,
      "epoch": 0.8798172308062285,
      "flos": 28362427004160.0,
      "grad_norm": 1.6928384710005555,
      "language_loss": 0.85542226,
      "learning_rate": 1.494955118880048e-07,
      "loss": 0.87743962,
      "num_input_tokens_seen": 158383275,
      "step": 7317,
      "time_per_iteration": 2.754767894744873
    },
    {
      "auxiliary_loss_clip": 0.01094179,
      "auxiliary_loss_mlp": 0.01022522,
      "balance_loss_clip": 1.04142654,
      "balance_loss_mlp": 1.0151875,
      "epoch": 0.8799374736968677,
      "flos": 23988974751360.0,
      "grad_norm": 1.6822132865595967,
      "language_loss": 0.72841108,
      "learning_rate": 1.4920014744897634e-07,
      "loss": 0.74957806,
      "num_input_tokens_seen": 158402690,
      "step": 7318,
      "time_per_iteration": 2.797797679901123
    },
    {
      "auxiliary_loss_clip": 0.01132527,
      "auxiliary_loss_mlp": 0.01023959,
      "balance_loss_clip": 1.04997468,
      "balance_loss_mlp": 1.01604366,
      "epoch": 0.8800577165875068,
      "flos": 25630271832960.0,
      "grad_norm": 1.8313088289341426,
      "language_loss": 0.8651973,
      "learning_rate": 1.4890506377481392e-07,
      "loss": 0.88676214,
      "num_input_tokens_seen": 158421780,
      "step": 7319,
      "time_per_iteration": 2.7391738891601562
    },
    {
      "auxiliary_loss_clip": 0.01107692,
      "auxiliary_loss_mlp": 0.0103131,
      "balance_loss_clip": 1.04289126,
      "balance_loss_mlp": 1.02396119,
      "epoch": 0.8801779594781458,
      "flos": 23440331439360.0,
      "grad_norm": 1.4485687946945425,
      "language_loss": 0.63963699,
      "learning_rate": 1.486102609102815e-07,
      "loss": 0.66102701,
      "num_input_tokens_seen": 158442330,
      "step": 7320,
      "time_per_iteration": 2.826251268386841
    },
    {
      "auxiliary_loss_clip": 0.01121473,
      "auxiliary_loss_mlp": 0.00761775,
      "balance_loss_clip": 1.04203022,
      "balance_loss_mlp": 1.0005064,
      "epoch": 0.880298202368785,
      "flos": 11508580656000.0,
      "grad_norm": 2.3066555372799447,
      "language_loss": 0.85887319,
      "learning_rate": 1.483157389001004e-07,
      "loss": 0.87770563,
      "num_input_tokens_seen": 158459890,
      "step": 7321,
      "time_per_iteration": 3.6849188804626465
    },
    {
      "auxiliary_loss_clip": 0.01147465,
      "auxiliary_loss_mlp": 0.01028464,
      "balance_loss_clip": 1.05095601,
      "balance_loss_mlp": 1.01992857,
      "epoch": 0.880418445259424,
      "flos": 22671447886080.0,
      "grad_norm": 4.812080345678931,
      "language_loss": 0.79099381,
      "learning_rate": 1.4802149778894933e-07,
      "loss": 0.81275308,
      "num_input_tokens_seen": 158478680,
      "step": 7322,
      "time_per_iteration": 2.7528066635131836
    },
    {
      "auxiliary_loss_clip": 0.01150998,
      "auxiliary_loss_mlp": 0.01023941,
      "balance_loss_clip": 1.04587114,
      "balance_loss_mlp": 1.0162698,
      "epoch": 0.8805386881500631,
      "flos": 20522158709760.0,
      "grad_norm": 2.1713846764859825,
      "language_loss": 0.87748021,
      "learning_rate": 1.4772753762146484e-07,
      "loss": 0.89922959,
      "num_input_tokens_seen": 158497935,
      "step": 7323,
      "time_per_iteration": 2.660870313644409
    },
    {
      "auxiliary_loss_clip": 0.01109275,
      "auxiliary_loss_mlp": 0.01025894,
      "balance_loss_clip": 1.04067802,
      "balance_loss_mlp": 1.01808,
      "epoch": 0.8806589310407023,
      "flos": 36538891620480.0,
      "grad_norm": 1.6456817482181867,
      "language_loss": 0.70680153,
      "learning_rate": 1.474338584422401e-07,
      "loss": 0.72815329,
      "num_input_tokens_seen": 158523145,
      "step": 7324,
      "time_per_iteration": 2.8819477558135986
    },
    {
      "auxiliary_loss_clip": 0.01162142,
      "auxiliary_loss_mlp": 0.01026791,
      "balance_loss_clip": 1.04980183,
      "balance_loss_mlp": 1.0198766,
      "epoch": 0.8807791739313413,
      "flos": 23440187784960.0,
      "grad_norm": 1.7300847676486837,
      "language_loss": 0.7583673,
      "learning_rate": 1.4714046029582595e-07,
      "loss": 0.78025663,
      "num_input_tokens_seen": 158542210,
      "step": 7325,
      "time_per_iteration": 2.693119764328003
    },
    {
      "auxiliary_loss_clip": 0.01144517,
      "auxiliary_loss_mlp": 0.00761771,
      "balance_loss_clip": 1.04534721,
      "balance_loss_mlp": 1.00046301,
      "epoch": 0.8808994168219804,
      "flos": 25956843310080.0,
      "grad_norm": 2.8702484734065243,
      "language_loss": 0.75961399,
      "learning_rate": 1.46847343226731e-07,
      "loss": 0.77867687,
      "num_input_tokens_seen": 158563250,
      "step": 7326,
      "time_per_iteration": 2.7357521057128906
    },
    {
      "auxiliary_loss_clip": 0.0117543,
      "auxiliary_loss_mlp": 0.01025921,
      "balance_loss_clip": 1.05355906,
      "balance_loss_mlp": 1.01847661,
      "epoch": 0.8810196597126195,
      "flos": 17092079303040.0,
      "grad_norm": 2.900939031663252,
      "language_loss": 0.69330132,
      "learning_rate": 1.465545072794203e-07,
      "loss": 0.71531487,
      "num_input_tokens_seen": 158581125,
      "step": 7327,
      "time_per_iteration": 2.609461545944214
    },
    {
      "auxiliary_loss_clip": 0.01126507,
      "auxiliary_loss_mlp": 0.01028359,
      "balance_loss_clip": 1.04561305,
      "balance_loss_mlp": 1.02020204,
      "epoch": 0.8811399026032586,
      "flos": 23002831785600.0,
      "grad_norm": 1.6985104543537923,
      "language_loss": 0.75881857,
      "learning_rate": 1.4626195249831774e-07,
      "loss": 0.78036726,
      "num_input_tokens_seen": 158602025,
      "step": 7328,
      "time_per_iteration": 2.7704479694366455
    },
    {
      "auxiliary_loss_clip": 0.01158684,
      "auxiliary_loss_mlp": 0.0102783,
      "balance_loss_clip": 1.04743838,
      "balance_loss_mlp": 1.02059698,
      "epoch": 0.8812601454938976,
      "flos": 14463813242880.0,
      "grad_norm": 2.057205488426424,
      "language_loss": 0.72321737,
      "learning_rate": 1.4596967892780244e-07,
      "loss": 0.7450825,
      "num_input_tokens_seen": 158618355,
      "step": 7329,
      "time_per_iteration": 3.520064115524292
    },
    {
      "auxiliary_loss_clip": 0.01155492,
      "auxiliary_loss_mlp": 0.01027719,
      "balance_loss_clip": 1.05008674,
      "balance_loss_mlp": 1.02032495,
      "epoch": 0.8813803883845368,
      "flos": 22493223578880.0,
      "grad_norm": 1.8150727123856725,
      "language_loss": 0.7453413,
      "learning_rate": 1.4567768661221314e-07,
      "loss": 0.76717341,
      "num_input_tokens_seen": 158638925,
      "step": 7330,
      "time_per_iteration": 2.6416423320770264
    },
    {
      "auxiliary_loss_clip": 0.01147624,
      "auxiliary_loss_mlp": 0.01031436,
      "balance_loss_clip": 1.04906249,
      "balance_loss_mlp": 1.02403867,
      "epoch": 0.8815006312751759,
      "flos": 21506901045120.0,
      "grad_norm": 2.3581829204697415,
      "language_loss": 0.74722368,
      "learning_rate": 1.4538597559584442e-07,
      "loss": 0.76901424,
      "num_input_tokens_seen": 158656715,
      "step": 7331,
      "time_per_iteration": 2.672309160232544
    },
    {
      "auxiliary_loss_clip": 0.01159799,
      "auxiliary_loss_mlp": 0.0102351,
      "balance_loss_clip": 1.05035818,
      "balance_loss_mlp": 1.01564837,
      "epoch": 0.8816208741658149,
      "flos": 22784566792320.0,
      "grad_norm": 2.455430240157703,
      "language_loss": 0.78805923,
      "learning_rate": 1.4509454592294823e-07,
      "loss": 0.80989236,
      "num_input_tokens_seen": 158677200,
      "step": 7332,
      "time_per_iteration": 2.667738676071167
    },
    {
      "auxiliary_loss_clip": 0.0114393,
      "auxiliary_loss_mlp": 0.01022145,
      "balance_loss_clip": 1.04655039,
      "balance_loss_mlp": 1.01390481,
      "epoch": 0.8817411170564541,
      "flos": 17779409026560.0,
      "grad_norm": 2.4093070244382826,
      "language_loss": 0.79078186,
      "learning_rate": 1.448033976377354e-07,
      "loss": 0.81244266,
      "num_input_tokens_seen": 158692185,
      "step": 7333,
      "time_per_iteration": 2.7606797218322754
    },
    {
      "auxiliary_loss_clip": 0.01137799,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.04468632,
      "balance_loss_mlp": 1.02261186,
      "epoch": 0.8818613599470931,
      "flos": 18551812112640.0,
      "grad_norm": 3.3796418737489637,
      "language_loss": 0.74110335,
      "learning_rate": 1.445125307843713e-07,
      "loss": 0.76278114,
      "num_input_tokens_seen": 158710410,
      "step": 7334,
      "time_per_iteration": 2.7113444805145264
    },
    {
      "auxiliary_loss_clip": 0.01140411,
      "auxiliary_loss_mlp": 0.01024512,
      "balance_loss_clip": 1.0478369,
      "balance_loss_mlp": 1.01725173,
      "epoch": 0.8819816028377322,
      "flos": 27599792417280.0,
      "grad_norm": 1.743336136869392,
      "language_loss": 0.75866699,
      "learning_rate": 1.442219454069813e-07,
      "loss": 0.78031623,
      "num_input_tokens_seen": 158731435,
      "step": 7335,
      "time_per_iteration": 2.6965231895446777
    },
    {
      "auxiliary_loss_clip": 0.01121954,
      "auxiliary_loss_mlp": 0.01031068,
      "balance_loss_clip": 1.04481196,
      "balance_loss_mlp": 1.02345908,
      "epoch": 0.8821018457283714,
      "flos": 23404600385280.0,
      "grad_norm": 1.9380687051006433,
      "language_loss": 0.66631466,
      "learning_rate": 1.4393164154964676e-07,
      "loss": 0.68784487,
      "num_input_tokens_seen": 158750965,
      "step": 7336,
      "time_per_iteration": 2.744532346725464
    },
    {
      "auxiliary_loss_clip": 0.01135466,
      "auxiliary_loss_mlp": 0.0076213,
      "balance_loss_clip": 1.04644656,
      "balance_loss_mlp": 1.00050533,
      "epoch": 0.8822220886190104,
      "flos": 29132459792640.0,
      "grad_norm": 1.8345044656133904,
      "language_loss": 0.9385519,
      "learning_rate": 1.4364161925640649e-07,
      "loss": 0.95752788,
      "num_input_tokens_seen": 158772365,
      "step": 7337,
      "time_per_iteration": 2.7326996326446533
    },
    {
      "auxiliary_loss_clip": 0.01158116,
      "auxiliary_loss_mlp": 0.01024844,
      "balance_loss_clip": 1.05000854,
      "balance_loss_mlp": 1.01764643,
      "epoch": 0.8823423315096495,
      "flos": 20485422074880.0,
      "grad_norm": 4.538700856243449,
      "language_loss": 0.85372472,
      "learning_rate": 1.4335187857125663e-07,
      "loss": 0.87555432,
      "num_input_tokens_seen": 158791065,
      "step": 7338,
      "time_per_iteration": 2.7066879272460938
    },
    {
      "auxiliary_loss_clip": 0.01146047,
      "auxiliary_loss_mlp": 0.01024872,
      "balance_loss_clip": 1.04801822,
      "balance_loss_mlp": 1.01784718,
      "epoch": 0.8824625744002886,
      "flos": 24206377818240.0,
      "grad_norm": 1.742444146053828,
      "language_loss": 0.75532436,
      "learning_rate": 1.4306241953815023e-07,
      "loss": 0.77703357,
      "num_input_tokens_seen": 158812125,
      "step": 7339,
      "time_per_iteration": 2.6862783432006836
    },
    {
      "auxiliary_loss_clip": 0.01132258,
      "auxiliary_loss_mlp": 0.0102999,
      "balance_loss_clip": 1.048895,
      "balance_loss_mlp": 1.02201533,
      "epoch": 0.8825828172909277,
      "flos": 24679500785280.0,
      "grad_norm": 17.790899393947676,
      "language_loss": 0.71120799,
      "learning_rate": 1.4277324220099862e-07,
      "loss": 0.73283041,
      "num_input_tokens_seen": 158834035,
      "step": 7340,
      "time_per_iteration": 2.825230836868286
    },
    {
      "auxiliary_loss_clip": 0.01159942,
      "auxiliary_loss_mlp": 0.01025114,
      "balance_loss_clip": 1.04896486,
      "balance_loss_mlp": 1.01773477,
      "epoch": 0.8827030601815667,
      "flos": 22456163721600.0,
      "grad_norm": 2.1496079334800284,
      "language_loss": 0.7453844,
      "learning_rate": 1.4248434660366938e-07,
      "loss": 0.76723498,
      "num_input_tokens_seen": 158853510,
      "step": 7341,
      "time_per_iteration": 3.780674695968628
    },
    {
      "auxiliary_loss_clip": 0.01124883,
      "auxiliary_loss_mlp": 0.01027051,
      "balance_loss_clip": 1.04606831,
      "balance_loss_mlp": 1.01887918,
      "epoch": 0.8828233030722058,
      "flos": 19865639877120.0,
      "grad_norm": 1.755500033692445,
      "language_loss": 0.70468581,
      "learning_rate": 1.4219573278998808e-07,
      "loss": 0.72620517,
      "num_input_tokens_seen": 158871970,
      "step": 7342,
      "time_per_iteration": 3.7029712200164795
    },
    {
      "auxiliary_loss_clip": 0.01175184,
      "auxiliary_loss_mlp": 0.01034333,
      "balance_loss_clip": 1.05155873,
      "balance_loss_mlp": 1.02611327,
      "epoch": 0.882943545962845,
      "flos": 39347213581440.0,
      "grad_norm": 2.543311192541817,
      "language_loss": 0.6503799,
      "learning_rate": 1.4190740080373685e-07,
      "loss": 0.6724751,
      "num_input_tokens_seen": 158892250,
      "step": 7343,
      "time_per_iteration": 2.8139917850494385
    },
    {
      "auxiliary_loss_clip": 0.01161394,
      "auxiliary_loss_mlp": 0.0102543,
      "balance_loss_clip": 1.05401421,
      "balance_loss_mlp": 1.01745498,
      "epoch": 0.883063788853484,
      "flos": 19054524908160.0,
      "grad_norm": 2.893443745989961,
      "language_loss": 0.84308702,
      "learning_rate": 1.4161935068865538e-07,
      "loss": 0.86495531,
      "num_input_tokens_seen": 158907395,
      "step": 7344,
      "time_per_iteration": 2.7550079822540283
    },
    {
      "auxiliary_loss_clip": 0.01099805,
      "auxiliary_loss_mlp": 0.01027342,
      "balance_loss_clip": 1.03852725,
      "balance_loss_mlp": 1.019683,
      "epoch": 0.8831840317441231,
      "flos": 18733196816640.0,
      "grad_norm": 3.0582067962553303,
      "language_loss": 0.76226568,
      "learning_rate": 1.4133158248844113e-07,
      "loss": 0.78353715,
      "num_input_tokens_seen": 158926300,
      "step": 7345,
      "time_per_iteration": 2.755136251449585
    },
    {
      "auxiliary_loss_clip": 0.0112986,
      "auxiliary_loss_mlp": 0.01027009,
      "balance_loss_clip": 1.04680204,
      "balance_loss_mlp": 1.01931405,
      "epoch": 0.8833042746347622,
      "flos": 26827712553600.0,
      "grad_norm": 2.1400949110737493,
      "language_loss": 0.73591471,
      "learning_rate": 1.4104409624674785e-07,
      "loss": 0.75748342,
      "num_input_tokens_seen": 158946085,
      "step": 7346,
      "time_per_iteration": 3.743356227874756
    },
    {
      "auxiliary_loss_clip": 0.01160283,
      "auxiliary_loss_mlp": 0.01028491,
      "balance_loss_clip": 1.05279207,
      "balance_loss_mlp": 1.02106416,
      "epoch": 0.8834245175254013,
      "flos": 26104077158400.0,
      "grad_norm": 1.7793340305025585,
      "language_loss": 0.78494954,
      "learning_rate": 1.407568920071873e-07,
      "loss": 0.80683726,
      "num_input_tokens_seen": 158964950,
      "step": 7347,
      "time_per_iteration": 2.640138864517212
    },
    {
      "auxiliary_loss_clip": 0.01116813,
      "auxiliary_loss_mlp": 0.01034914,
      "balance_loss_clip": 1.0434649,
      "balance_loss_mlp": 1.02677166,
      "epoch": 0.8835447604160404,
      "flos": 30629036977920.0,
      "grad_norm": 5.56170770141013,
      "language_loss": 0.68596172,
      "learning_rate": 1.4046996981332782e-07,
      "loss": 0.707479,
      "num_input_tokens_seen": 158984835,
      "step": 7348,
      "time_per_iteration": 2.8685922622680664
    },
    {
      "auxiliary_loss_clip": 0.01174853,
      "auxiliary_loss_mlp": 0.01025788,
      "balance_loss_clip": 1.04877567,
      "balance_loss_mlp": 1.01733017,
      "epoch": 0.8836650033066795,
      "flos": 24718356322560.0,
      "grad_norm": 2.6777109083720894,
      "language_loss": 0.78658718,
      "learning_rate": 1.4018332970869516e-07,
      "loss": 0.80859363,
      "num_input_tokens_seen": 159002775,
      "step": 7349,
      "time_per_iteration": 2.638267755508423
    },
    {
      "auxiliary_loss_clip": 0.01133907,
      "auxiliary_loss_mlp": 0.01038025,
      "balance_loss_clip": 1.0430094,
      "balance_loss_mlp": 1.02981782,
      "epoch": 0.8837852461973186,
      "flos": 25413371556480.0,
      "grad_norm": 1.8405152009494752,
      "language_loss": 0.85202765,
      "learning_rate": 1.398969717367733e-07,
      "loss": 0.87374699,
      "num_input_tokens_seen": 159024100,
      "step": 7350,
      "time_per_iteration": 2.7451560497283936
    },
    {
      "auxiliary_loss_clip": 0.01156347,
      "auxiliary_loss_mlp": 0.01028024,
      "balance_loss_clip": 1.0497812,
      "balance_loss_mlp": 1.0206871,
      "epoch": 0.8839054890879576,
      "flos": 17822574195840.0,
      "grad_norm": 1.9250129132485738,
      "language_loss": 0.76181948,
      "learning_rate": 1.396108959410014e-07,
      "loss": 0.78366321,
      "num_input_tokens_seen": 159043315,
      "step": 7351,
      "time_per_iteration": 2.7979321479797363
    },
    {
      "auxiliary_loss_clip": 0.01174697,
      "auxiliary_loss_mlp": 0.00762152,
      "balance_loss_clip": 1.05237317,
      "balance_loss_mlp": 1.00045562,
      "epoch": 0.8840257319785968,
      "flos": 23769021818880.0,
      "grad_norm": 1.5928210474170406,
      "language_loss": 0.81377041,
      "learning_rate": 1.3932510236477745e-07,
      "loss": 0.83313888,
      "num_input_tokens_seen": 159063985,
      "step": 7352,
      "time_per_iteration": 2.6027307510375977
    },
    {
      "auxiliary_loss_clip": 0.01132176,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.04695499,
      "balance_loss_mlp": 1.02157307,
      "epoch": 0.8841459748692359,
      "flos": 29059776622080.0,
      "grad_norm": 1.7639812865659252,
      "language_loss": 0.56127548,
      "learning_rate": 1.3903959105145636e-07,
      "loss": 0.58289289,
      "num_input_tokens_seen": 159084475,
      "step": 7353,
      "time_per_iteration": 2.794560432434082
    },
    {
      "auxiliary_loss_clip": 0.0114306,
      "auxiliary_loss_mlp": 0.01027765,
      "balance_loss_clip": 1.0494504,
      "balance_loss_mlp": 1.0201056,
      "epoch": 0.8842662177598749,
      "flos": 24311523905280.0,
      "grad_norm": 1.8692701851770477,
      "language_loss": 0.82963336,
      "learning_rate": 1.387543620443492e-07,
      "loss": 0.85134161,
      "num_input_tokens_seen": 159101320,
      "step": 7354,
      "time_per_iteration": 2.7047603130340576
    },
    {
      "auxiliary_loss_clip": 0.01135526,
      "auxiliary_loss_mlp": 0.01029671,
      "balance_loss_clip": 1.0444808,
      "balance_loss_mlp": 1.02250624,
      "epoch": 0.8843864606505141,
      "flos": 25007867942400.0,
      "grad_norm": 1.909810597818011,
      "language_loss": 0.8429898,
      "learning_rate": 1.3846941538672606e-07,
      "loss": 0.86464185,
      "num_input_tokens_seen": 159120025,
      "step": 7355,
      "time_per_iteration": 3.7058677673339844
    },
    {
      "auxiliary_loss_clip": 0.01085264,
      "auxiliary_loss_mlp": 0.01026546,
      "balance_loss_clip": 1.04099798,
      "balance_loss_mlp": 1.01873755,
      "epoch": 0.8845067035411531,
      "flos": 28183915388160.0,
      "grad_norm": 2.0416487219026194,
      "language_loss": 0.80866683,
      "learning_rate": 1.3818475112181193e-07,
      "loss": 0.82978499,
      "num_input_tokens_seen": 159138820,
      "step": 7356,
      "time_per_iteration": 2.8929364681243896
    },
    {
      "auxiliary_loss_clip": 0.01156307,
      "auxiliary_loss_mlp": 0.01023564,
      "balance_loss_clip": 1.05113006,
      "balance_loss_mlp": 1.0163275,
      "epoch": 0.8846269464317922,
      "flos": 12853219311360.0,
      "grad_norm": 3.3049448467540086,
      "language_loss": 0.79746693,
      "learning_rate": 1.3790036929279091e-07,
      "loss": 0.8192656,
      "num_input_tokens_seen": 159155975,
      "step": 7357,
      "time_per_iteration": 2.6125190258026123
    },
    {
      "auxiliary_loss_clip": 0.01096041,
      "auxiliary_loss_mlp": 0.0102432,
      "balance_loss_clip": 1.04402757,
      "balance_loss_mlp": 1.01629174,
      "epoch": 0.8847471893224313,
      "flos": 18624351628800.0,
      "grad_norm": 2.5835287645011493,
      "language_loss": 0.58710432,
      "learning_rate": 1.3761626994280363e-07,
      "loss": 0.6083079,
      "num_input_tokens_seen": 159173445,
      "step": 7358,
      "time_per_iteration": 2.7619435787200928
    },
    {
      "auxiliary_loss_clip": 0.0117262,
      "auxiliary_loss_mlp": 0.01026863,
      "balance_loss_clip": 1.04924035,
      "balance_loss_mlp": 1.01913869,
      "epoch": 0.8848674322130704,
      "flos": 35769433449600.0,
      "grad_norm": 1.6397658507637374,
      "language_loss": 0.73845851,
      "learning_rate": 1.3733245311494735e-07,
      "loss": 0.76045334,
      "num_input_tokens_seen": 159196100,
      "step": 7359,
      "time_per_iteration": 2.7134714126586914
    },
    {
      "auxiliary_loss_clip": 0.01105066,
      "auxiliary_loss_mlp": 0.0102813,
      "balance_loss_clip": 1.04476666,
      "balance_loss_mlp": 1.0206852,
      "epoch": 0.8849876751037095,
      "flos": 24243760897920.0,
      "grad_norm": 2.132137176576723,
      "language_loss": 0.70912087,
      "learning_rate": 1.3704891885227676e-07,
      "loss": 0.73045278,
      "num_input_tokens_seen": 159216145,
      "step": 7360,
      "time_per_iteration": 2.815351963043213
    },
    {
      "auxiliary_loss_clip": 0.01160675,
      "auxiliary_loss_mlp": 0.01033221,
      "balance_loss_clip": 1.04883432,
      "balance_loss_mlp": 1.02473903,
      "epoch": 0.8851079179943486,
      "flos": 21500580251520.0,
      "grad_norm": 2.614357409493415,
      "language_loss": 0.78049469,
      "learning_rate": 1.367656671978037e-07,
      "loss": 0.80243361,
      "num_input_tokens_seen": 159233610,
      "step": 7361,
      "time_per_iteration": 2.608555555343628
    },
    {
      "auxiliary_loss_clip": 0.01162961,
      "auxiliary_loss_mlp": 0.01023623,
      "balance_loss_clip": 1.04941225,
      "balance_loss_mlp": 1.01558185,
      "epoch": 0.8852281608849877,
      "flos": 15300711198720.0,
      "grad_norm": 2.5479552512331294,
      "language_loss": 0.73856157,
      "learning_rate": 1.36482698194498e-07,
      "loss": 0.76042742,
      "num_input_tokens_seen": 159250155,
      "step": 7362,
      "time_per_iteration": 2.6431386470794678
    },
    {
      "auxiliary_loss_clip": 0.01176191,
      "auxiliary_loss_mlp": 0.01023235,
      "balance_loss_clip": 1.05194414,
      "balance_loss_mlp": 1.01554549,
      "epoch": 0.8853484037756267,
      "flos": 23295719283840.0,
      "grad_norm": 2.5473653717512175,
      "language_loss": 0.72389776,
      "learning_rate": 1.3620001188528506e-07,
      "loss": 0.74589205,
      "num_input_tokens_seen": 159270875,
      "step": 7363,
      "time_per_iteration": 2.6114041805267334
    },
    {
      "auxiliary_loss_clip": 0.01139355,
      "auxiliary_loss_mlp": 0.01029165,
      "balance_loss_clip": 1.04559326,
      "balance_loss_mlp": 1.02101696,
      "epoch": 0.8854686466662659,
      "flos": 25114773795840.0,
      "grad_norm": 2.8521466879893405,
      "language_loss": 0.73582506,
      "learning_rate": 1.3591760831304865e-07,
      "loss": 0.75751024,
      "num_input_tokens_seen": 159288565,
      "step": 7364,
      "time_per_iteration": 2.750915288925171
    },
    {
      "auxiliary_loss_clip": 0.01173797,
      "auxiliary_loss_mlp": 0.01025934,
      "balance_loss_clip": 1.05200386,
      "balance_loss_mlp": 1.01807177,
      "epoch": 0.885588889556905,
      "flos": 21390873137280.0,
      "grad_norm": 2.2559429610392825,
      "language_loss": 0.79578239,
      "learning_rate": 1.356354875206287e-07,
      "loss": 0.81777966,
      "num_input_tokens_seen": 159306400,
      "step": 7365,
      "time_per_iteration": 2.5768189430236816
    },
    {
      "auxiliary_loss_clip": 0.01079285,
      "auxiliary_loss_mlp": 0.00762034,
      "balance_loss_clip": 1.03970492,
      "balance_loss_mlp": 1.00049591,
      "epoch": 0.885709132447544,
      "flos": 26906752431360.0,
      "grad_norm": 2.882651246384728,
      "language_loss": 0.70221883,
      "learning_rate": 1.3535364955082296e-07,
      "loss": 0.72063208,
      "num_input_tokens_seen": 159326250,
      "step": 7366,
      "time_per_iteration": 2.926259756088257
    },
    {
      "auxiliary_loss_clip": 0.01173205,
      "auxiliary_loss_mlp": 0.01024299,
      "balance_loss_clip": 1.05269885,
      "balance_loss_mlp": 1.01701498,
      "epoch": 0.8858293753381832,
      "flos": 26103394800000.0,
      "grad_norm": 3.573941712411158,
      "language_loss": 0.64406419,
      "learning_rate": 1.3507209444638613e-07,
      "loss": 0.66603923,
      "num_input_tokens_seen": 159348250,
      "step": 7367,
      "time_per_iteration": 4.535213947296143
    },
    {
      "auxiliary_loss_clip": 0.01135701,
      "auxiliary_loss_mlp": 0.01024315,
      "balance_loss_clip": 1.04657912,
      "balance_loss_mlp": 1.01634014,
      "epoch": 0.8859496182288222,
      "flos": 23292810282240.0,
      "grad_norm": 2.104570343656309,
      "language_loss": 0.73977679,
      "learning_rate": 1.347908222500298e-07,
      "loss": 0.76137686,
      "num_input_tokens_seen": 159368325,
      "step": 7368,
      "time_per_iteration": 2.6567399501800537
    },
    {
      "auxiliary_loss_clip": 0.01144226,
      "auxiliary_loss_mlp": 0.01028075,
      "balance_loss_clip": 1.04918122,
      "balance_loss_mlp": 1.0208987,
      "epoch": 0.8860698611194613,
      "flos": 16872916469760.0,
      "grad_norm": 2.0043207746645675,
      "language_loss": 0.69871235,
      "learning_rate": 1.3450983300442276e-07,
      "loss": 0.72043538,
      "num_input_tokens_seen": 159387555,
      "step": 7369,
      "time_per_iteration": 2.7424397468566895
    },
    {
      "auxiliary_loss_clip": 0.01152199,
      "auxiliary_loss_mlp": 0.01031533,
      "balance_loss_clip": 1.04844356,
      "balance_loss_mlp": 1.02407956,
      "epoch": 0.8861901040101005,
      "flos": 24681404206080.0,
      "grad_norm": 1.9910755051238587,
      "language_loss": 0.73261607,
      "learning_rate": 1.3422912675219068e-07,
      "loss": 0.75445342,
      "num_input_tokens_seen": 159407310,
      "step": 7370,
      "time_per_iteration": 2.7054920196533203
    },
    {
      "auxiliary_loss_clip": 0.01132858,
      "auxiliary_loss_mlp": 0.01029889,
      "balance_loss_clip": 1.04688215,
      "balance_loss_mlp": 1.02220881,
      "epoch": 0.8863103469007395,
      "flos": 24423026699520.0,
      "grad_norm": 2.407114725444136,
      "language_loss": 0.79054391,
      "learning_rate": 1.339487035359166e-07,
      "loss": 0.81217134,
      "num_input_tokens_seen": 159427680,
      "step": 7371,
      "time_per_iteration": 2.792947769165039
    },
    {
      "auxiliary_loss_clip": 0.01157951,
      "auxiliary_loss_mlp": 0.01024254,
      "balance_loss_clip": 1.04912353,
      "balance_loss_mlp": 1.01716387,
      "epoch": 0.8864305897913786,
      "flos": 22053964158720.0,
      "grad_norm": 1.5867571766557005,
      "language_loss": 0.85112786,
      "learning_rate": 1.336685633981409e-07,
      "loss": 0.87294996,
      "num_input_tokens_seen": 159448765,
      "step": 7372,
      "time_per_iteration": 3.6243062019348145
    },
    {
      "auxiliary_loss_clip": 0.01154017,
      "auxiliary_loss_mlp": 0.01028716,
      "balance_loss_clip": 1.05021346,
      "balance_loss_mlp": 1.02053225,
      "epoch": 0.8865508326820177,
      "flos": 19099449843840.0,
      "grad_norm": 2.0764762110047137,
      "language_loss": 0.75100303,
      "learning_rate": 1.333887063813597e-07,
      "loss": 0.77283037,
      "num_input_tokens_seen": 159466870,
      "step": 7373,
      "time_per_iteration": 2.6150195598602295
    },
    {
      "auxiliary_loss_clip": 0.01127569,
      "auxiliary_loss_mlp": 0.01028673,
      "balance_loss_clip": 1.04632711,
      "balance_loss_mlp": 1.0213002,
      "epoch": 0.8866710755726568,
      "flos": 15414189240960.0,
      "grad_norm": 1.7853017884645503,
      "language_loss": 0.66335976,
      "learning_rate": 1.331091325280278e-07,
      "loss": 0.68492216,
      "num_input_tokens_seen": 159485840,
      "step": 7374,
      "time_per_iteration": 2.7245090007781982
    },
    {
      "auxiliary_loss_clip": 0.01143538,
      "auxiliary_loss_mlp": 0.01025724,
      "balance_loss_clip": 1.04882121,
      "balance_loss_mlp": 1.01801395,
      "epoch": 0.8867913184632958,
      "flos": 20083689388800.0,
      "grad_norm": 1.8036148263598604,
      "language_loss": 0.7859382,
      "learning_rate": 1.3282984188055625e-07,
      "loss": 0.8076309,
      "num_input_tokens_seen": 159505630,
      "step": 7375,
      "time_per_iteration": 2.659451723098755
    },
    {
      "auxiliary_loss_clip": 0.01125319,
      "auxiliary_loss_mlp": 0.01028508,
      "balance_loss_clip": 1.04399109,
      "balance_loss_mlp": 1.02032101,
      "epoch": 0.8869115613539349,
      "flos": 23365852588800.0,
      "grad_norm": 1.9306424775716677,
      "language_loss": 0.79820305,
      "learning_rate": 1.3255083448131288e-07,
      "loss": 0.81974131,
      "num_input_tokens_seen": 159524675,
      "step": 7376,
      "time_per_iteration": 2.7089028358459473
    },
    {
      "auxiliary_loss_clip": 0.01124103,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 1.04395509,
      "balance_loss_mlp": 1.01859415,
      "epoch": 0.8870318042445741,
      "flos": 21286840371840.0,
      "grad_norm": 3.370967137258449,
      "language_loss": 0.79209375,
      "learning_rate": 1.3227211037262365e-07,
      "loss": 0.81360435,
      "num_input_tokens_seen": 159541915,
      "step": 7377,
      "time_per_iteration": 2.7062718868255615
    },
    {
      "auxiliary_loss_clip": 0.01159178,
      "auxiliary_loss_mlp": 0.0103022,
      "balance_loss_clip": 1.0500772,
      "balance_loss_mlp": 1.02206028,
      "epoch": 0.8871520471352131,
      "flos": 20010862563840.0,
      "grad_norm": 2.7676796896632965,
      "language_loss": 0.85723239,
      "learning_rate": 1.319936695967696e-07,
      "loss": 0.87912631,
      "num_input_tokens_seen": 159559740,
      "step": 7378,
      "time_per_iteration": 2.6893177032470703
    },
    {
      "auxiliary_loss_clip": 0.01123925,
      "auxiliary_loss_mlp": 0.01026154,
      "balance_loss_clip": 1.04483569,
      "balance_loss_mlp": 1.01769042,
      "epoch": 0.8872722900258522,
      "flos": 22601422321920.0,
      "grad_norm": 2.2892113347351244,
      "language_loss": 0.81977236,
      "learning_rate": 1.3171551219599097e-07,
      "loss": 0.84127313,
      "num_input_tokens_seen": 159578265,
      "step": 7379,
      "time_per_iteration": 2.7732250690460205
    },
    {
      "auxiliary_loss_clip": 0.01114782,
      "auxiliary_loss_mlp": 0.01026811,
      "balance_loss_clip": 1.0428617,
      "balance_loss_mlp": 1.01876426,
      "epoch": 0.8873925329164913,
      "flos": 22163276223360.0,
      "grad_norm": 3.120530269769321,
      "language_loss": 0.78143322,
      "learning_rate": 1.3143763821248377e-07,
      "loss": 0.80284917,
      "num_input_tokens_seen": 159595350,
      "step": 7380,
      "time_per_iteration": 3.662724494934082
    },
    {
      "auxiliary_loss_clip": 0.01140538,
      "auxiliary_loss_mlp": 0.01027368,
      "balance_loss_clip": 1.05180085,
      "balance_loss_mlp": 1.02047467,
      "epoch": 0.8875127758071304,
      "flos": 19208223204480.0,
      "grad_norm": 2.0708766506508103,
      "language_loss": 0.72498453,
      "learning_rate": 1.3116004768840118e-07,
      "loss": 0.74666363,
      "num_input_tokens_seen": 159613725,
      "step": 7381,
      "time_per_iteration": 2.7138054370880127
    },
    {
      "auxiliary_loss_clip": 0.01158854,
      "auxiliary_loss_mlp": 0.01026376,
      "balance_loss_clip": 1.04831874,
      "balance_loss_mlp": 1.01818633,
      "epoch": 0.8876330186977694,
      "flos": 18110900666880.0,
      "grad_norm": 2.4606194476503433,
      "language_loss": 0.74253201,
      "learning_rate": 1.3088274066585348e-07,
      "loss": 0.76438433,
      "num_input_tokens_seen": 159631335,
      "step": 7382,
      "time_per_iteration": 2.61901593208313
    },
    {
      "auxiliary_loss_clip": 0.01150097,
      "auxiliary_loss_mlp": 0.00761915,
      "balance_loss_clip": 1.04710734,
      "balance_loss_mlp": 1.00045466,
      "epoch": 0.8877532615884086,
      "flos": 22009434272640.0,
      "grad_norm": 2.2387069177694117,
      "language_loss": 0.91139412,
      "learning_rate": 1.3060571718690749e-07,
      "loss": 0.93051428,
      "num_input_tokens_seen": 159648830,
      "step": 7383,
      "time_per_iteration": 2.6562230587005615
    },
    {
      "auxiliary_loss_clip": 0.01035745,
      "auxiliary_loss_mlp": 0.00999705,
      "balance_loss_clip": 1.00662041,
      "balance_loss_mlp": 0.99875689,
      "epoch": 0.8878735044790477,
      "flos": 72136924346880.0,
      "grad_norm": 0.7454911404546687,
      "language_loss": 0.568156,
      "learning_rate": 1.3032897729358805e-07,
      "loss": 0.58851051,
      "num_input_tokens_seen": 159709785,
      "step": 7384,
      "time_per_iteration": 3.2339630126953125
    },
    {
      "auxiliary_loss_clip": 0.01089169,
      "auxiliary_loss_mlp": 0.01035714,
      "balance_loss_clip": 1.04003203,
      "balance_loss_mlp": 1.02674329,
      "epoch": 0.8879937473696867,
      "flos": 27526355061120.0,
      "grad_norm": 1.9399254571720057,
      "language_loss": 0.80275428,
      "learning_rate": 1.3005252102787645e-07,
      "loss": 0.8240031,
      "num_input_tokens_seen": 159728725,
      "step": 7385,
      "time_per_iteration": 2.801798105239868
    },
    {
      "auxiliary_loss_clip": 0.01129993,
      "auxiliary_loss_mlp": 0.01028224,
      "balance_loss_clip": 1.04627931,
      "balance_loss_mlp": 1.0206182,
      "epoch": 0.8881139902603259,
      "flos": 22234091886720.0,
      "grad_norm": 1.5715783992244625,
      "language_loss": 0.7374841,
      "learning_rate": 1.297763484317105e-07,
      "loss": 0.75906622,
      "num_input_tokens_seen": 159747020,
      "step": 7386,
      "time_per_iteration": 2.7289810180664062
    },
    {
      "auxiliary_loss_clip": 0.01148531,
      "auxiliary_loss_mlp": 0.01031396,
      "balance_loss_clip": 1.04883277,
      "balance_loss_mlp": 1.02316499,
      "epoch": 0.888234233150965,
      "flos": 20299548170880.0,
      "grad_norm": 2.1004947730888204,
      "language_loss": 0.70800138,
      "learning_rate": 1.2950045954698551e-07,
      "loss": 0.7298007,
      "num_input_tokens_seen": 159764855,
      "step": 7387,
      "time_per_iteration": 2.6445164680480957
    },
    {
      "auxiliary_loss_clip": 0.01115915,
      "auxiliary_loss_mlp": 0.01030976,
      "balance_loss_clip": 1.04585469,
      "balance_loss_mlp": 1.02406442,
      "epoch": 0.888354476041604,
      "flos": 18147996437760.0,
      "grad_norm": 3.3584020210347365,
      "language_loss": 0.75530815,
      "learning_rate": 1.2922485441555343e-07,
      "loss": 0.77677703,
      "num_input_tokens_seen": 159783935,
      "step": 7388,
      "time_per_iteration": 2.8362743854522705
    },
    {
      "auxiliary_loss_clip": 0.01171302,
      "auxiliary_loss_mlp": 0.01022123,
      "balance_loss_clip": 1.04898059,
      "balance_loss_mlp": 1.01486337,
      "epoch": 0.8884747189322432,
      "flos": 22014282608640.0,
      "grad_norm": 1.7506308109244606,
      "language_loss": 0.82026935,
      "learning_rate": 1.2894953307922363e-07,
      "loss": 0.84220362,
      "num_input_tokens_seen": 159802895,
      "step": 7389,
      "time_per_iteration": 2.615461826324463
    },
    {
      "auxiliary_loss_clip": 0.01161169,
      "auxiliary_loss_mlp": 0.01023664,
      "balance_loss_clip": 1.0517112,
      "balance_loss_mlp": 1.01604033,
      "epoch": 0.8885949618228822,
      "flos": 19786779567360.0,
      "grad_norm": 2.019032189444583,
      "language_loss": 0.83914065,
      "learning_rate": 1.2867449557976208e-07,
      "loss": 0.86098897,
      "num_input_tokens_seen": 159820995,
      "step": 7390,
      "time_per_iteration": 2.6303176879882812
    },
    {
      "auxiliary_loss_clip": 0.01174213,
      "auxiliary_loss_mlp": 0.01028412,
      "balance_loss_clip": 1.05117297,
      "balance_loss_mlp": 1.02105403,
      "epoch": 0.8887152047135213,
      "flos": 20047599198720.0,
      "grad_norm": 2.245171256967049,
      "language_loss": 0.75677305,
      "learning_rate": 1.283997419588916e-07,
      "loss": 0.7787993,
      "num_input_tokens_seen": 159840465,
      "step": 7391,
      "time_per_iteration": 2.5531721115112305
    },
    {
      "auxiliary_loss_clip": 0.01148089,
      "auxiliary_loss_mlp": 0.0102755,
      "balance_loss_clip": 1.04791439,
      "balance_loss_mlp": 1.01950884,
      "epoch": 0.8888354476041604,
      "flos": 18588117784320.0,
      "grad_norm": 2.055706615548871,
      "language_loss": 0.62229896,
      "learning_rate": 1.2812527225829216e-07,
      "loss": 0.64405537,
      "num_input_tokens_seen": 159858690,
      "step": 7392,
      "time_per_iteration": 2.702301263809204
    },
    {
      "auxiliary_loss_clip": 0.01153421,
      "auxiliary_loss_mlp": 0.01029506,
      "balance_loss_clip": 1.05098534,
      "balance_loss_mlp": 1.02111959,
      "epoch": 0.8889556904947995,
      "flos": 21689794120320.0,
      "grad_norm": 1.9542335750961057,
      "language_loss": 0.76725811,
      "learning_rate": 1.2785108651960052e-07,
      "loss": 0.78908741,
      "num_input_tokens_seen": 159880325,
      "step": 7393,
      "time_per_iteration": 3.670783281326294
    },
    {
      "auxiliary_loss_clip": 0.01161126,
      "auxiliary_loss_mlp": 0.01026349,
      "balance_loss_clip": 1.05125964,
      "balance_loss_mlp": 1.0190444,
      "epoch": 0.8890759333854386,
      "flos": 27381204201600.0,
      "grad_norm": 2.2486080826227903,
      "language_loss": 0.8104322,
      "learning_rate": 1.2757718478441094e-07,
      "loss": 0.83230692,
      "num_input_tokens_seen": 159901070,
      "step": 7394,
      "time_per_iteration": 2.7570455074310303
    },
    {
      "auxiliary_loss_clip": 0.01126625,
      "auxiliary_loss_mlp": 0.01026694,
      "balance_loss_clip": 1.04182804,
      "balance_loss_mlp": 1.01816154,
      "epoch": 0.8891961762760777,
      "flos": 24498834353280.0,
      "grad_norm": 1.728767681438295,
      "language_loss": 0.77378416,
      "learning_rate": 1.2730356709427302e-07,
      "loss": 0.79531729,
      "num_input_tokens_seen": 159919750,
      "step": 7395,
      "time_per_iteration": 2.76289439201355
    },
    {
      "auxiliary_loss_clip": 0.01129388,
      "auxiliary_loss_mlp": 0.01027603,
      "balance_loss_clip": 1.04525757,
      "balance_loss_mlp": 1.02004266,
      "epoch": 0.8893164191667168,
      "flos": 41499770895360.0,
      "grad_norm": 1.4725820536003773,
      "language_loss": 0.5980252,
      "learning_rate": 1.2703023349069542e-07,
      "loss": 0.61959505,
      "num_input_tokens_seen": 159944600,
      "step": 7396,
      "time_per_iteration": 2.8738293647766113
    },
    {
      "auxiliary_loss_clip": 0.01142429,
      "auxiliary_loss_mlp": 0.01021793,
      "balance_loss_clip": 1.04964924,
      "balance_loss_mlp": 1.01446724,
      "epoch": 0.8894366620573558,
      "flos": 33583623120000.0,
      "grad_norm": 2.2215174280494248,
      "language_loss": 0.61826754,
      "learning_rate": 1.2675718401514223e-07,
      "loss": 0.63990974,
      "num_input_tokens_seen": 159968780,
      "step": 7397,
      "time_per_iteration": 2.8256797790527344
    },
    {
      "auxiliary_loss_clip": 0.01173965,
      "auxiliary_loss_mlp": 0.01028928,
      "balance_loss_clip": 1.05367517,
      "balance_loss_mlp": 1.02103031,
      "epoch": 0.889556904947995,
      "flos": 16909832672640.0,
      "grad_norm": 2.110053618367281,
      "language_loss": 0.74976468,
      "learning_rate": 1.264844187090346e-07,
      "loss": 0.7717936,
      "num_input_tokens_seen": 159985905,
      "step": 7398,
      "time_per_iteration": 3.5565571784973145
    },
    {
      "auxiliary_loss_clip": 0.01151229,
      "auxiliary_loss_mlp": 0.01029313,
      "balance_loss_clip": 1.04706597,
      "balance_loss_mlp": 1.02186823,
      "epoch": 0.889677147838634,
      "flos": 26030855283840.0,
      "grad_norm": 1.9972963936172667,
      "language_loss": 0.752707,
      "learning_rate": 1.262119376137516e-07,
      "loss": 0.77451241,
      "num_input_tokens_seen": 160006965,
      "step": 7399,
      "time_per_iteration": 2.7338287830352783
    },
    {
      "auxiliary_loss_clip": 0.01129464,
      "auxiliary_loss_mlp": 0.01034348,
      "balance_loss_clip": 1.04709899,
      "balance_loss_mlp": 1.02713037,
      "epoch": 0.8897973907292731,
      "flos": 26468283110400.0,
      "grad_norm": 1.6185433986023183,
      "language_loss": 0.8495388,
      "learning_rate": 1.2593974077062707e-07,
      "loss": 0.87117696,
      "num_input_tokens_seen": 160028585,
      "step": 7400,
      "time_per_iteration": 2.782371759414673
    },
    {
      "auxiliary_loss_clip": 0.01131659,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.04662967,
      "balance_loss_mlp": 1.02126038,
      "epoch": 0.8899176336199123,
      "flos": 26249694894720.0,
      "grad_norm": 2.0939597459112265,
      "language_loss": 0.63883471,
      "learning_rate": 1.2566782822095423e-07,
      "loss": 0.66043919,
      "num_input_tokens_seen": 160048840,
      "step": 7401,
      "time_per_iteration": 2.717695474624634
    },
    {
      "auxiliary_loss_clip": 0.01115115,
      "auxiliary_loss_mlp": 0.00761722,
      "balance_loss_clip": 1.04394054,
      "balance_loss_mlp": 1.00050831,
      "epoch": 0.8900378765105513,
      "flos": 20811742156800.0,
      "grad_norm": 1.6487576662403123,
      "language_loss": 0.71944427,
      "learning_rate": 1.2539620000598162e-07,
      "loss": 0.73821259,
      "num_input_tokens_seen": 160068175,
      "step": 7402,
      "time_per_iteration": 2.7639377117156982
    },
    {
      "auxiliary_loss_clip": 0.01139222,
      "auxiliary_loss_mlp": 0.01033609,
      "balance_loss_clip": 1.04511142,
      "balance_loss_mlp": 1.02524638,
      "epoch": 0.8901581194011904,
      "flos": 16472333018880.0,
      "grad_norm": 3.364527899060452,
      "language_loss": 0.79593372,
      "learning_rate": 1.2512485616691492e-07,
      "loss": 0.81766206,
      "num_input_tokens_seen": 160085230,
      "step": 7403,
      "time_per_iteration": 2.618126392364502
    },
    {
      "auxiliary_loss_clip": 0.01106721,
      "auxiliary_loss_mlp": 0.00761822,
      "balance_loss_clip": 1.04604697,
      "balance_loss_mlp": 1.00048828,
      "epoch": 0.8902783622918296,
      "flos": 35155253773440.0,
      "grad_norm": 1.4905233818645827,
      "language_loss": 0.8102231,
      "learning_rate": 1.2485379674491681e-07,
      "loss": 0.82890856,
      "num_input_tokens_seen": 160111425,
      "step": 7404,
      "time_per_iteration": 2.945892810821533
    },
    {
      "auxiliary_loss_clip": 0.01145299,
      "auxiliary_loss_mlp": 0.01027705,
      "balance_loss_clip": 1.04757619,
      "balance_loss_mlp": 1.0202961,
      "epoch": 0.8903986051824686,
      "flos": 17201068145280.0,
      "grad_norm": 2.108827749512052,
      "language_loss": 0.79233658,
      "learning_rate": 1.2458302178110657e-07,
      "loss": 0.81406665,
      "num_input_tokens_seen": 160129790,
      "step": 7405,
      "time_per_iteration": 2.6231651306152344
    },
    {
      "auxiliary_loss_clip": 0.01138202,
      "auxiliary_loss_mlp": 0.01025088,
      "balance_loss_clip": 1.0474031,
      "balance_loss_mlp": 1.01800382,
      "epoch": 0.8905188480731077,
      "flos": 25483863997440.0,
      "grad_norm": 2.454425540966896,
      "language_loss": 0.82521242,
      "learning_rate": 1.2431253131656118e-07,
      "loss": 0.84684527,
      "num_input_tokens_seen": 160149265,
      "step": 7406,
      "time_per_iteration": 3.6963050365448
    },
    {
      "auxiliary_loss_clip": 0.01127428,
      "auxiliary_loss_mlp": 0.01028338,
      "balance_loss_clip": 1.04389167,
      "balance_loss_mlp": 1.02060676,
      "epoch": 0.8906390909637467,
      "flos": 23365888502400.0,
      "grad_norm": 2.6178658329006095,
      "language_loss": 0.76817572,
      "learning_rate": 1.240423253923133e-07,
      "loss": 0.78973341,
      "num_input_tokens_seen": 160168870,
      "step": 7407,
      "time_per_iteration": 2.7300734519958496
    },
    {
      "auxiliary_loss_clip": 0.0111172,
      "auxiliary_loss_mlp": 0.01027221,
      "balance_loss_clip": 1.0460813,
      "balance_loss_mlp": 1.01928794,
      "epoch": 0.8907593338543859,
      "flos": 21068790860160.0,
      "grad_norm": 3.059723251599233,
      "language_loss": 0.69291729,
      "learning_rate": 1.237724040493533e-07,
      "loss": 0.71430671,
      "num_input_tokens_seen": 160187495,
      "step": 7408,
      "time_per_iteration": 2.6835269927978516
    },
    {
      "auxiliary_loss_clip": 0.01155441,
      "auxiliary_loss_mlp": 0.01035959,
      "balance_loss_clip": 1.05170584,
      "balance_loss_mlp": 1.02682173,
      "epoch": 0.8908795767450249,
      "flos": 21869562712320.0,
      "grad_norm": 2.4230130694062875,
      "language_loss": 0.72975522,
      "learning_rate": 1.2350276732862773e-07,
      "loss": 0.75166917,
      "num_input_tokens_seen": 160208520,
      "step": 7409,
      "time_per_iteration": 2.7005105018615723
    },
    {
      "auxiliary_loss_clip": 0.01040103,
      "auxiliary_loss_mlp": 0.01001919,
      "balance_loss_clip": 1.00905061,
      "balance_loss_mlp": 1.0009712,
      "epoch": 0.890999819635664,
      "flos": 66307869348480.0,
      "grad_norm": 0.8495802499733985,
      "language_loss": 0.56644833,
      "learning_rate": 1.2323341527103993e-07,
      "loss": 0.58686852,
      "num_input_tokens_seen": 160263720,
      "step": 7410,
      "time_per_iteration": 3.190088987350464
    },
    {
      "auxiliary_loss_clip": 0.01140229,
      "auxiliary_loss_mlp": 0.01025737,
      "balance_loss_clip": 1.04601836,
      "balance_loss_mlp": 1.01786351,
      "epoch": 0.8911200625263032,
      "flos": 26869908055680.0,
      "grad_norm": 2.2575339697730095,
      "language_loss": 0.8527267,
      "learning_rate": 1.2296434791745135e-07,
      "loss": 0.87438637,
      "num_input_tokens_seen": 160282170,
      "step": 7411,
      "time_per_iteration": 2.748673677444458
    },
    {
      "auxiliary_loss_clip": 0.01159263,
      "auxiliary_loss_mlp": 0.00762233,
      "balance_loss_clip": 1.05234528,
      "balance_loss_mlp": 1.00045669,
      "epoch": 0.8912403054169422,
      "flos": 20885825957760.0,
      "grad_norm": 1.659354605319327,
      "language_loss": 0.7710349,
      "learning_rate": 1.2269556530867875e-07,
      "loss": 0.79024988,
      "num_input_tokens_seen": 160300725,
      "step": 7412,
      "time_per_iteration": 2.6808969974517822
    },
    {
      "auxiliary_loss_clip": 0.01154306,
      "auxiliary_loss_mlp": 0.01031118,
      "balance_loss_clip": 1.05275905,
      "balance_loss_mlp": 1.02287471,
      "epoch": 0.8913605483075813,
      "flos": 27016567286400.0,
      "grad_norm": 1.961553252413355,
      "language_loss": 0.82220244,
      "learning_rate": 1.2242706748549614e-07,
      "loss": 0.84405673,
      "num_input_tokens_seen": 160318720,
      "step": 7413,
      "time_per_iteration": 2.795976161956787
    },
    {
      "auxiliary_loss_clip": 0.01160708,
      "auxiliary_loss_mlp": 0.01031737,
      "balance_loss_clip": 1.0491035,
      "balance_loss_mlp": 1.0230763,
      "epoch": 0.8914807911982204,
      "flos": 23621500661760.0,
      "grad_norm": 2.460776351303339,
      "language_loss": 0.8262105,
      "learning_rate": 1.2215885448863473e-07,
      "loss": 0.84813488,
      "num_input_tokens_seen": 160339595,
      "step": 7414,
      "time_per_iteration": 2.7082290649414062
    },
    {
      "auxiliary_loss_clip": 0.01124355,
      "auxiliary_loss_mlp": 0.01028898,
      "balance_loss_clip": 1.04539466,
      "balance_loss_mlp": 1.02151036,
      "epoch": 0.8916010340888595,
      "flos": 24462277286400.0,
      "grad_norm": 1.88678741072764,
      "language_loss": 0.80417997,
      "learning_rate": 1.2189092635878152e-07,
      "loss": 0.82571244,
      "num_input_tokens_seen": 160361045,
      "step": 7415,
      "time_per_iteration": 2.7481300830841064
    },
    {
      "auxiliary_loss_clip": 0.0108972,
      "auxiliary_loss_mlp": 0.01025797,
      "balance_loss_clip": 1.03949797,
      "balance_loss_mlp": 1.01812553,
      "epoch": 0.8917212769794985,
      "flos": 21215773313280.0,
      "grad_norm": 1.7307929572042724,
      "language_loss": 0.77528012,
      "learning_rate": 1.216232831365822e-07,
      "loss": 0.7964353,
      "num_input_tokens_seen": 160379990,
      "step": 7416,
      "time_per_iteration": 2.807093620300293
    },
    {
      "auxiliary_loss_clip": 0.01144101,
      "auxiliary_loss_mlp": 0.01034403,
      "balance_loss_clip": 1.04958355,
      "balance_loss_mlp": 1.02633274,
      "epoch": 0.8918415198701377,
      "flos": 25513992529920.0,
      "grad_norm": 2.455707361000989,
      "language_loss": 0.80979908,
      "learning_rate": 1.2135592486263678e-07,
      "loss": 0.83158416,
      "num_input_tokens_seen": 160399240,
      "step": 7417,
      "time_per_iteration": 2.6829240322113037
    },
    {
      "auxiliary_loss_clip": 0.01100991,
      "auxiliary_loss_mlp": 0.01029202,
      "balance_loss_clip": 1.04222047,
      "balance_loss_mlp": 1.02176356,
      "epoch": 0.8919617627607768,
      "flos": 37853006693760.0,
      "grad_norm": 1.7844606927456392,
      "language_loss": 0.61343789,
      "learning_rate": 1.2108885157750415e-07,
      "loss": 0.63473982,
      "num_input_tokens_seen": 160421600,
      "step": 7418,
      "time_per_iteration": 2.9113852977752686
    },
    {
      "auxiliary_loss_clip": 0.01143159,
      "auxiliary_loss_mlp": 0.01029237,
      "balance_loss_clip": 1.05171275,
      "balance_loss_mlp": 1.02162862,
      "epoch": 0.8920820056514158,
      "flos": 26213676531840.0,
      "grad_norm": 1.9921454940144612,
      "language_loss": 0.80435753,
      "learning_rate": 1.2082206332169897e-07,
      "loss": 0.82608151,
      "num_input_tokens_seen": 160441695,
      "step": 7419,
      "time_per_iteration": 4.572452068328857
    },
    {
      "auxiliary_loss_clip": 0.01113535,
      "auxiliary_loss_mlp": 0.0102657,
      "balance_loss_clip": 1.04611659,
      "balance_loss_mlp": 1.01948309,
      "epoch": 0.892202248542055,
      "flos": 17383135207680.0,
      "grad_norm": 2.720304701831911,
      "language_loss": 0.7303102,
      "learning_rate": 1.2055556013569225e-07,
      "loss": 0.75171125,
      "num_input_tokens_seen": 160457205,
      "step": 7420,
      "time_per_iteration": 2.7056102752685547
    },
    {
      "auxiliary_loss_clip": 0.0113508,
      "auxiliary_loss_mlp": 0.01028535,
      "balance_loss_clip": 1.04639816,
      "balance_loss_mlp": 1.02078962,
      "epoch": 0.892322491432694,
      "flos": 21324223451520.0,
      "grad_norm": 2.0593734991204387,
      "language_loss": 0.82209432,
      "learning_rate": 1.2028934205991315e-07,
      "loss": 0.84373045,
      "num_input_tokens_seen": 160476525,
      "step": 7421,
      "time_per_iteration": 2.7378649711608887
    },
    {
      "auxiliary_loss_clip": 0.01124393,
      "auxiliary_loss_mlp": 0.01027362,
      "balance_loss_clip": 1.04545629,
      "balance_loss_mlp": 1.0195148,
      "epoch": 0.8924427343233331,
      "flos": 24029374573440.0,
      "grad_norm": 1.4650340618422617,
      "language_loss": 0.76745391,
      "learning_rate": 1.2002340913474607e-07,
      "loss": 0.78897154,
      "num_input_tokens_seen": 160500160,
      "step": 7422,
      "time_per_iteration": 2.7435672283172607
    },
    {
      "auxiliary_loss_clip": 0.01146909,
      "auxiliary_loss_mlp": 0.01027037,
      "balance_loss_clip": 1.04646587,
      "balance_loss_mlp": 1.01872826,
      "epoch": 0.8925629772139723,
      "flos": 30008069631360.0,
      "grad_norm": 4.970309685853115,
      "language_loss": 0.7403177,
      "learning_rate": 1.1975776140053317e-07,
      "loss": 0.76205713,
      "num_input_tokens_seen": 160520130,
      "step": 7423,
      "time_per_iteration": 2.7674126625061035
    },
    {
      "auxiliary_loss_clip": 0.01137304,
      "auxiliary_loss_mlp": 0.01030166,
      "balance_loss_clip": 1.04792261,
      "balance_loss_mlp": 1.02147579,
      "epoch": 0.8926832201046113,
      "flos": 22601709630720.0,
      "grad_norm": 2.91370443490376,
      "language_loss": 0.73713988,
      "learning_rate": 1.194923988975729e-07,
      "loss": 0.75881457,
      "num_input_tokens_seen": 160539730,
      "step": 7424,
      "time_per_iteration": 3.78094482421875
    },
    {
      "auxiliary_loss_clip": 0.01128227,
      "auxiliary_loss_mlp": 0.01025384,
      "balance_loss_clip": 1.04229474,
      "balance_loss_mlp": 1.01764703,
      "epoch": 0.8928034629952504,
      "flos": 13297722117120.0,
      "grad_norm": 2.2220433176144496,
      "language_loss": 0.7330004,
      "learning_rate": 1.192273216661206e-07,
      "loss": 0.75453651,
      "num_input_tokens_seen": 160557820,
      "step": 7425,
      "time_per_iteration": 2.734940528869629
    },
    {
      "auxiliary_loss_clip": 0.01026099,
      "auxiliary_loss_mlp": 0.01000867,
      "balance_loss_clip": 1.00792813,
      "balance_loss_mlp": 0.9999429,
      "epoch": 0.8929237058858895,
      "flos": 54854556744960.0,
      "grad_norm": 0.7693152731224997,
      "language_loss": 0.5747143,
      "learning_rate": 1.189625297463881e-07,
      "loss": 0.59498399,
      "num_input_tokens_seen": 160619510,
      "step": 7426,
      "time_per_iteration": 3.297910690307617
    },
    {
      "auxiliary_loss_clip": 0.01141936,
      "auxiliary_loss_mlp": 0.01027635,
      "balance_loss_clip": 1.0466671,
      "balance_loss_mlp": 1.02020812,
      "epoch": 0.8930439487765286,
      "flos": 28883850785280.0,
      "grad_norm": 1.562308635736932,
      "language_loss": 0.79615015,
      "learning_rate": 1.1869802317854394e-07,
      "loss": 0.81784582,
      "num_input_tokens_seen": 160643295,
      "step": 7427,
      "time_per_iteration": 2.751685619354248
    },
    {
      "auxiliary_loss_clip": 0.01125207,
      "auxiliary_loss_mlp": 0.01027878,
      "balance_loss_clip": 1.04417157,
      "balance_loss_mlp": 1.01997399,
      "epoch": 0.8931641916671677,
      "flos": 22419283432320.0,
      "grad_norm": 2.4069002367090357,
      "language_loss": 0.7184903,
      "learning_rate": 1.1843380200271425e-07,
      "loss": 0.74002111,
      "num_input_tokens_seen": 160662495,
      "step": 7428,
      "time_per_iteration": 2.7426087856292725
    },
    {
      "auxiliary_loss_clip": 0.0113858,
      "auxiliary_loss_mlp": 0.01033073,
      "balance_loss_clip": 1.04567742,
      "balance_loss_mlp": 1.02479947,
      "epoch": 0.8932844345578068,
      "flos": 25843149786240.0,
      "grad_norm": 2.3040037345506676,
      "language_loss": 0.80475098,
      "learning_rate": 1.181698662589805e-07,
      "loss": 0.82646751,
      "num_input_tokens_seen": 160682080,
      "step": 7429,
      "time_per_iteration": 2.733351945877075
    },
    {
      "auxiliary_loss_clip": 0.01150079,
      "auxiliary_loss_mlp": 0.01023977,
      "balance_loss_clip": 1.0457058,
      "balance_loss_mlp": 1.01581085,
      "epoch": 0.8934046774484459,
      "flos": 22925803069440.0,
      "grad_norm": 1.9401085840890524,
      "language_loss": 0.76238418,
      "learning_rate": 1.1790621598738249e-07,
      "loss": 0.78412473,
      "num_input_tokens_seen": 160700395,
      "step": 7430,
      "time_per_iteration": 2.6443824768066406
    },
    {
      "auxiliary_loss_clip": 0.01119706,
      "auxiliary_loss_mlp": 0.01022452,
      "balance_loss_clip": 1.04619908,
      "balance_loss_mlp": 1.01612449,
      "epoch": 0.8935249203390849,
      "flos": 24462097718400.0,
      "grad_norm": 1.907581039067834,
      "language_loss": 0.75127137,
      "learning_rate": 1.1764285122791461e-07,
      "loss": 0.77269292,
      "num_input_tokens_seen": 160721115,
      "step": 7431,
      "time_per_iteration": 2.846845865249634
    },
    {
      "auxiliary_loss_clip": 0.01119038,
      "auxiliary_loss_mlp": 0.01023927,
      "balance_loss_clip": 1.04562736,
      "balance_loss_mlp": 1.01621687,
      "epoch": 0.8936451632297241,
      "flos": 15742735966080.0,
      "grad_norm": 1.7437774030376003,
      "language_loss": 0.7736448,
      "learning_rate": 1.173797720205294e-07,
      "loss": 0.79507446,
      "num_input_tokens_seen": 160739150,
      "step": 7432,
      "time_per_iteration": 2.690141201019287
    },
    {
      "auxiliary_loss_clip": 0.01141675,
      "auxiliary_loss_mlp": 0.01032336,
      "balance_loss_clip": 1.04701173,
      "balance_loss_mlp": 1.02339482,
      "epoch": 0.8937654061203631,
      "flos": 35115500396160.0,
      "grad_norm": 2.415081831056623,
      "language_loss": 0.71666515,
      "learning_rate": 1.1711697840513602e-07,
      "loss": 0.73840523,
      "num_input_tokens_seen": 160758585,
      "step": 7433,
      "time_per_iteration": 3.7904934883117676
    },
    {
      "auxiliary_loss_clip": 0.01156716,
      "auxiliary_loss_mlp": 0.01030702,
      "balance_loss_clip": 1.05082273,
      "balance_loss_mlp": 1.02275121,
      "epoch": 0.8938856490110022,
      "flos": 16107444708480.0,
      "grad_norm": 2.1364443250842675,
      "language_loss": 0.71118027,
      "learning_rate": 1.1685447042160012e-07,
      "loss": 0.7330544,
      "num_input_tokens_seen": 160776620,
      "step": 7434,
      "time_per_iteration": 2.636467933654785
    },
    {
      "auxiliary_loss_clip": 0.01147503,
      "auxiliary_loss_mlp": 0.01028967,
      "balance_loss_clip": 1.05054867,
      "balance_loss_mlp": 1.02075338,
      "epoch": 0.8940058919016414,
      "flos": 20704189858560.0,
      "grad_norm": 1.8646162873479128,
      "language_loss": 0.71725607,
      "learning_rate": 1.1659224810974367e-07,
      "loss": 0.73902076,
      "num_input_tokens_seen": 160796580,
      "step": 7435,
      "time_per_iteration": 2.780308246612549
    },
    {
      "auxiliary_loss_clip": 0.01142514,
      "auxiliary_loss_mlp": 0.01029388,
      "balance_loss_clip": 1.04896092,
      "balance_loss_mlp": 1.02085853,
      "epoch": 0.8941261347922804,
      "flos": 25229041937280.0,
      "grad_norm": 1.4058205795861762,
      "language_loss": 0.68474787,
      "learning_rate": 1.1633031150934591e-07,
      "loss": 0.70646691,
      "num_input_tokens_seen": 160819610,
      "step": 7436,
      "time_per_iteration": 2.7844643592834473
    },
    {
      "auxiliary_loss_clip": 0.01178017,
      "auxiliary_loss_mlp": 0.01026618,
      "balance_loss_clip": 1.05321097,
      "balance_loss_mlp": 1.01884294,
      "epoch": 0.8942463776829195,
      "flos": 19537236806400.0,
      "grad_norm": 1.8667330751184126,
      "language_loss": 0.80039549,
      "learning_rate": 1.1606866066014176e-07,
      "loss": 0.82244188,
      "num_input_tokens_seen": 160838660,
      "step": 7437,
      "time_per_iteration": 2.693211078643799
    },
    {
      "auxiliary_loss_clip": 0.01153964,
      "auxiliary_loss_mlp": 0.01029072,
      "balance_loss_clip": 1.04600143,
      "balance_loss_mlp": 1.02101922,
      "epoch": 0.8943666205735585,
      "flos": 22301567585280.0,
      "grad_norm": 2.4112285866446315,
      "language_loss": 0.75841063,
      "learning_rate": 1.1580729560182434e-07,
      "loss": 0.78024101,
      "num_input_tokens_seen": 160854515,
      "step": 7438,
      "time_per_iteration": 2.672736883163452
    },
    {
      "auxiliary_loss_clip": 0.01120774,
      "auxiliary_loss_mlp": 0.01025986,
      "balance_loss_clip": 1.04308271,
      "balance_loss_mlp": 1.01776063,
      "epoch": 0.8944868634641977,
      "flos": 18912893581440.0,
      "grad_norm": 3.6402205183714464,
      "language_loss": 0.70996249,
      "learning_rate": 1.1554621637404171e-07,
      "loss": 0.73143005,
      "num_input_tokens_seen": 160872605,
      "step": 7439,
      "time_per_iteration": 2.8019511699676514
    },
    {
      "auxiliary_loss_clip": 0.01158557,
      "auxiliary_loss_mlp": 0.01023199,
      "balance_loss_clip": 1.04868841,
      "balance_loss_mlp": 1.01610923,
      "epoch": 0.8946071063548368,
      "flos": 14460904241280.0,
      "grad_norm": 2.341740832171538,
      "language_loss": 0.61382115,
      "learning_rate": 1.1528542301639999e-07,
      "loss": 0.63563865,
      "num_input_tokens_seen": 160889395,
      "step": 7440,
      "time_per_iteration": 2.629870891571045
    },
    {
      "auxiliary_loss_clip": 0.01117967,
      "auxiliary_loss_mlp": 0.01029317,
      "balance_loss_clip": 1.04467237,
      "balance_loss_mlp": 1.02099037,
      "epoch": 0.8947273492454758,
      "flos": 20084084438400.0,
      "grad_norm": 2.933684415425619,
      "language_loss": 0.82557011,
      "learning_rate": 1.1502491556846105e-07,
      "loss": 0.84704292,
      "num_input_tokens_seen": 160907890,
      "step": 7441,
      "time_per_iteration": 2.7800252437591553
    },
    {
      "auxiliary_loss_clip": 0.01172423,
      "auxiliary_loss_mlp": 0.01026811,
      "balance_loss_clip": 1.05057383,
      "balance_loss_mlp": 1.01929498,
      "epoch": 0.894847592136115,
      "flos": 18550555136640.0,
      "grad_norm": 2.6475562120632152,
      "language_loss": 0.81368816,
      "learning_rate": 1.1476469406974331e-07,
      "loss": 0.83568054,
      "num_input_tokens_seen": 160923490,
      "step": 7442,
      "time_per_iteration": 2.679720640182495
    },
    {
      "auxiliary_loss_clip": 0.0111069,
      "auxiliary_loss_mlp": 0.01026526,
      "balance_loss_clip": 1.04453731,
      "balance_loss_mlp": 1.01884341,
      "epoch": 0.894967835026754,
      "flos": 23478468704640.0,
      "grad_norm": 1.739628771263465,
      "language_loss": 0.77209365,
      "learning_rate": 1.1450475855972341e-07,
      "loss": 0.79346585,
      "num_input_tokens_seen": 160944280,
      "step": 7443,
      "time_per_iteration": 2.764779567718506
    },
    {
      "auxiliary_loss_clip": 0.01156662,
      "auxiliary_loss_mlp": 0.01028972,
      "balance_loss_clip": 1.04787374,
      "balance_loss_mlp": 1.02124774,
      "epoch": 0.8950880779173931,
      "flos": 15188310564480.0,
      "grad_norm": 13.34435538571431,
      "language_loss": 0.70662951,
      "learning_rate": 1.1424510907783158e-07,
      "loss": 0.72848582,
      "num_input_tokens_seen": 160961560,
      "step": 7444,
      "time_per_iteration": 2.6385371685028076
    },
    {
      "auxiliary_loss_clip": 0.01144591,
      "auxiliary_loss_mlp": 0.01025767,
      "balance_loss_clip": 1.04905975,
      "balance_loss_mlp": 1.01905203,
      "epoch": 0.8952083208080323,
      "flos": 22091957769600.0,
      "grad_norm": 1.9401125147432285,
      "language_loss": 0.82781911,
      "learning_rate": 1.1398574566345787e-07,
      "loss": 0.84952265,
      "num_input_tokens_seen": 160982195,
      "step": 7445,
      "time_per_iteration": 3.6846814155578613
    },
    {
      "auxiliary_loss_clip": 0.01142521,
      "auxiliary_loss_mlp": 0.00762265,
      "balance_loss_clip": 1.04727983,
      "balance_loss_mlp": 1.00039816,
      "epoch": 0.8953285636986713,
      "flos": 23254026572160.0,
      "grad_norm": 4.543878019347808,
      "language_loss": 0.83042395,
      "learning_rate": 1.1372666835594702e-07,
      "loss": 0.84947187,
      "num_input_tokens_seen": 161000520,
      "step": 7446,
      "time_per_iteration": 3.643979072570801
    },
    {
      "auxiliary_loss_clip": 0.01120412,
      "auxiliary_loss_mlp": 0.01029477,
      "balance_loss_clip": 1.045012,
      "balance_loss_mlp": 1.02177644,
      "epoch": 0.8954488065893104,
      "flos": 16362661818240.0,
      "grad_norm": 2.225353549718969,
      "language_loss": 0.71587586,
      "learning_rate": 1.1346787719460071e-07,
      "loss": 0.73737478,
      "num_input_tokens_seen": 161019405,
      "step": 7447,
      "time_per_iteration": 2.7615835666656494
    },
    {
      "auxiliary_loss_clip": 0.01143508,
      "auxiliary_loss_mlp": 0.0076188,
      "balance_loss_clip": 1.04614592,
      "balance_loss_mlp": 1.00046146,
      "epoch": 0.8955690494799495,
      "flos": 18257883120000.0,
      "grad_norm": 1.7063975967767802,
      "language_loss": 0.72198373,
      "learning_rate": 1.1320937221867732e-07,
      "loss": 0.74103761,
      "num_input_tokens_seen": 161036985,
      "step": 7448,
      "time_per_iteration": 2.663926601409912
    },
    {
      "auxiliary_loss_clip": 0.01146399,
      "auxiliary_loss_mlp": 0.01028827,
      "balance_loss_clip": 1.04727054,
      "balance_loss_mlp": 1.02132308,
      "epoch": 0.8956892923705886,
      "flos": 25447486498560.0,
      "grad_norm": 1.8710845394681204,
      "language_loss": 0.79546189,
      "learning_rate": 1.1295115346739192e-07,
      "loss": 0.81721419,
      "num_input_tokens_seen": 161056985,
      "step": 7449,
      "time_per_iteration": 2.763733386993408
    },
    {
      "auxiliary_loss_clip": 0.01159593,
      "auxiliary_loss_mlp": 0.01030495,
      "balance_loss_clip": 1.04902482,
      "balance_loss_mlp": 1.0221951,
      "epoch": 0.8958095352612276,
      "flos": 52661883939840.0,
      "grad_norm": 3.144610440335558,
      "language_loss": 0.73233306,
      "learning_rate": 1.1269322097991629e-07,
      "loss": 0.7542339,
      "num_input_tokens_seen": 161080270,
      "step": 7450,
      "time_per_iteration": 3.8256821632385254
    },
    {
      "auxiliary_loss_clip": 0.01149494,
      "auxiliary_loss_mlp": 0.01037905,
      "balance_loss_clip": 1.05079889,
      "balance_loss_mlp": 1.03004634,
      "epoch": 0.8959297781518668,
      "flos": 23186335392000.0,
      "grad_norm": 6.0974715574396985,
      "language_loss": 0.68113375,
      "learning_rate": 1.1243557479537846e-07,
      "loss": 0.70300782,
      "num_input_tokens_seen": 161100160,
      "step": 7451,
      "time_per_iteration": 2.69295597076416
    },
    {
      "auxiliary_loss_clip": 0.01105512,
      "auxiliary_loss_mlp": 0.01029538,
      "balance_loss_clip": 1.04217291,
      "balance_loss_mlp": 1.02147901,
      "epoch": 0.8960500210425059,
      "flos": 20334309557760.0,
      "grad_norm": 3.360029164163069,
      "language_loss": 0.68785667,
      "learning_rate": 1.121782149528634e-07,
      "loss": 0.70920724,
      "num_input_tokens_seen": 161117260,
      "step": 7452,
      "time_per_iteration": 2.7920517921447754
    },
    {
      "auxiliary_loss_clip": 0.01120863,
      "auxiliary_loss_mlp": 0.01024313,
      "balance_loss_clip": 1.0432725,
      "balance_loss_mlp": 1.01682675,
      "epoch": 0.8961702639331449,
      "flos": 19901694153600.0,
      "grad_norm": 2.056057038351073,
      "language_loss": 0.78649342,
      "learning_rate": 1.1192114149141208e-07,
      "loss": 0.80794519,
      "num_input_tokens_seen": 161136895,
      "step": 7453,
      "time_per_iteration": 2.8008363246917725
    },
    {
      "auxiliary_loss_clip": 0.01149423,
      "auxiliary_loss_mlp": 0.01025556,
      "balance_loss_clip": 1.04859114,
      "balance_loss_mlp": 1.017241,
      "epoch": 0.8962905068237841,
      "flos": 12896348567040.0,
      "grad_norm": 2.7067849359478697,
      "language_loss": 0.6576755,
      "learning_rate": 1.1166435445002197e-07,
      "loss": 0.6794253,
      "num_input_tokens_seen": 161154565,
      "step": 7454,
      "time_per_iteration": 2.7565672397613525
    },
    {
      "auxiliary_loss_clip": 0.01108733,
      "auxiliary_loss_mlp": 0.01028051,
      "balance_loss_clip": 1.04618561,
      "balance_loss_mlp": 1.02002263,
      "epoch": 0.8964107497144231,
      "flos": 23440331439360.0,
      "grad_norm": 2.1207641916074627,
      "language_loss": 0.68783784,
      "learning_rate": 1.1140785386764818e-07,
      "loss": 0.70920569,
      "num_input_tokens_seen": 161173265,
      "step": 7455,
      "time_per_iteration": 2.857571601867676
    },
    {
      "auxiliary_loss_clip": 0.01152537,
      "auxiliary_loss_mlp": 0.01031058,
      "balance_loss_clip": 1.05013263,
      "balance_loss_mlp": 1.02317262,
      "epoch": 0.8965309926050622,
      "flos": 19500176949120.0,
      "grad_norm": 2.003473237406585,
      "language_loss": 0.70178306,
      "learning_rate": 1.1115163978320153e-07,
      "loss": 0.72361898,
      "num_input_tokens_seen": 161191995,
      "step": 7456,
      "time_per_iteration": 2.670517921447754
    },
    {
      "auxiliary_loss_clip": 0.01150442,
      "auxiliary_loss_mlp": 0.01027642,
      "balance_loss_clip": 1.0499258,
      "balance_loss_mlp": 1.01908875,
      "epoch": 0.8966512354957014,
      "flos": 28658008022400.0,
      "grad_norm": 2.4421613964865965,
      "language_loss": 0.82639635,
      "learning_rate": 1.1089571223554917e-07,
      "loss": 0.84817731,
      "num_input_tokens_seen": 161212880,
      "step": 7457,
      "time_per_iteration": 2.884352922439575
    },
    {
      "auxiliary_loss_clip": 0.0117421,
      "auxiliary_loss_mlp": 0.01035451,
      "balance_loss_clip": 1.05185997,
      "balance_loss_mlp": 1.02743983,
      "epoch": 0.8967714783863404,
      "flos": 23370916406400.0,
      "grad_norm": 1.7616007090666383,
      "language_loss": 0.85471606,
      "learning_rate": 1.1064007126351537e-07,
      "loss": 0.8768127,
      "num_input_tokens_seen": 161233595,
      "step": 7458,
      "time_per_iteration": 2.6132142543792725
    },
    {
      "auxiliary_loss_clip": 0.01124835,
      "auxiliary_loss_mlp": 0.00761934,
      "balance_loss_clip": 1.0441972,
      "balance_loss_mlp": 1.0004425,
      "epoch": 0.8968917212769795,
      "flos": 24535175938560.0,
      "grad_norm": 4.766359037228922,
      "language_loss": 0.7653538,
      "learning_rate": 1.1038471690588003e-07,
      "loss": 0.78422153,
      "num_input_tokens_seen": 161252740,
      "step": 7459,
      "time_per_iteration": 3.691600799560547
    },
    {
      "auxiliary_loss_clip": 0.01170874,
      "auxiliary_loss_mlp": 0.01025967,
      "balance_loss_clip": 1.05049419,
      "balance_loss_mlp": 1.01856458,
      "epoch": 0.8970119641676186,
      "flos": 23475416048640.0,
      "grad_norm": 3.7423627823001784,
      "language_loss": 0.80043113,
      "learning_rate": 1.1012964920138145e-07,
      "loss": 0.8223995,
      "num_input_tokens_seen": 161272325,
      "step": 7460,
      "time_per_iteration": 2.6353704929351807
    },
    {
      "auxiliary_loss_clip": 0.0113859,
      "auxiliary_loss_mlp": 0.01027908,
      "balance_loss_clip": 1.04676819,
      "balance_loss_mlp": 1.02030873,
      "epoch": 0.8971322070582577,
      "flos": 24538192680960.0,
      "grad_norm": 2.1660561880916163,
      "language_loss": 0.75862682,
      "learning_rate": 1.0987486818871205e-07,
      "loss": 0.7802918,
      "num_input_tokens_seen": 161295915,
      "step": 7461,
      "time_per_iteration": 2.7691726684570312
    },
    {
      "auxiliary_loss_clip": 0.01175473,
      "auxiliary_loss_mlp": 0.01027249,
      "balance_loss_clip": 1.05171525,
      "balance_loss_mlp": 1.01907754,
      "epoch": 0.8972524499488967,
      "flos": 21797454159360.0,
      "grad_norm": 5.879333130275937,
      "language_loss": 0.73487306,
      "learning_rate": 1.0962037390652245e-07,
      "loss": 0.75690031,
      "num_input_tokens_seen": 161314935,
      "step": 7462,
      "time_per_iteration": 2.627454996109009
    },
    {
      "auxiliary_loss_clip": 0.01101176,
      "auxiliary_loss_mlp": 0.01035625,
      "balance_loss_clip": 1.04030633,
      "balance_loss_mlp": 1.02702379,
      "epoch": 0.8973726928395359,
      "flos": 21726243446400.0,
      "grad_norm": 1.982529945088344,
      "language_loss": 0.7227115,
      "learning_rate": 1.0936616639341911e-07,
      "loss": 0.74407959,
      "num_input_tokens_seen": 161335225,
      "step": 7463,
      "time_per_iteration": 2.800414800643921
    },
    {
      "auxiliary_loss_clip": 0.01002096,
      "auxiliary_loss_mlp": 0.00999373,
      "balance_loss_clip": 1.0080111,
      "balance_loss_mlp": 0.99849111,
      "epoch": 0.897492935730175,
      "flos": 53837100097920.0,
      "grad_norm": 0.7510087712389274,
      "language_loss": 0.54677045,
      "learning_rate": 1.0911224568796473e-07,
      "loss": 0.5667851,
      "num_input_tokens_seen": 161393420,
      "step": 7464,
      "time_per_iteration": 3.5078768730163574
    },
    {
      "auxiliary_loss_clip": 0.01131087,
      "auxiliary_loss_mlp": 0.01026965,
      "balance_loss_clip": 1.05041766,
      "balance_loss_mlp": 1.01955581,
      "epoch": 0.897613178620814,
      "flos": 18290346036480.0,
      "grad_norm": 2.2729584940873355,
      "language_loss": 0.71109247,
      "learning_rate": 1.0885861182867984e-07,
      "loss": 0.73267293,
      "num_input_tokens_seen": 161411525,
      "step": 7465,
      "time_per_iteration": 3.0398476123809814
    },
    {
      "auxiliary_loss_clip": 0.01149233,
      "auxiliary_loss_mlp": 0.01027827,
      "balance_loss_clip": 1.04624867,
      "balance_loss_mlp": 1.02008998,
      "epoch": 0.8977334215114532,
      "flos": 32993718059520.0,
      "grad_norm": 5.238946763076238,
      "language_loss": 0.70925963,
      "learning_rate": 1.0860526485403942e-07,
      "loss": 0.73103029,
      "num_input_tokens_seen": 161432800,
      "step": 7466,
      "time_per_iteration": 2.8271732330322266
    },
    {
      "auxiliary_loss_clip": 0.01098924,
      "auxiliary_loss_mlp": 0.01023662,
      "balance_loss_clip": 1.03916037,
      "balance_loss_mlp": 1.01612747,
      "epoch": 0.8978536644020922,
      "flos": 15195636938880.0,
      "grad_norm": 1.553549493635097,
      "language_loss": 0.77456725,
      "learning_rate": 1.0835220480247675e-07,
      "loss": 0.79579312,
      "num_input_tokens_seen": 161451295,
      "step": 7467,
      "time_per_iteration": 2.7862460613250732
    },
    {
      "auxiliary_loss_clip": 0.01133863,
      "auxiliary_loss_mlp": 0.00762094,
      "balance_loss_clip": 1.04447055,
      "balance_loss_mlp": 1.00049019,
      "epoch": 0.8979739072927313,
      "flos": 18004389863040.0,
      "grad_norm": 4.513614291154688,
      "language_loss": 0.83865207,
      "learning_rate": 1.0809943171238067e-07,
      "loss": 0.8576116,
      "num_input_tokens_seen": 161469220,
      "step": 7468,
      "time_per_iteration": 2.7117910385131836
    },
    {
      "auxiliary_loss_clip": 0.01138312,
      "auxiliary_loss_mlp": 0.01031098,
      "balance_loss_clip": 1.04437804,
      "balance_loss_mlp": 1.02281284,
      "epoch": 0.8980941501833704,
      "flos": 22271546793600.0,
      "grad_norm": 4.159755369379204,
      "language_loss": 0.62929964,
      "learning_rate": 1.078469456220965e-07,
      "loss": 0.6509937,
      "num_input_tokens_seen": 161489375,
      "step": 7469,
      "time_per_iteration": 2.8116977214813232
    },
    {
      "auxiliary_loss_clip": 0.01157232,
      "auxiliary_loss_mlp": 0.01026431,
      "balance_loss_clip": 1.0493952,
      "balance_loss_mlp": 1.01898026,
      "epoch": 0.8982143930740095,
      "flos": 37560729726720.0,
      "grad_norm": 2.572708853078807,
      "language_loss": 0.69792032,
      "learning_rate": 1.0759474656992606e-07,
      "loss": 0.71975702,
      "num_input_tokens_seen": 161512145,
      "step": 7470,
      "time_per_iteration": 2.7780160903930664
    },
    {
      "auxiliary_loss_clip": 0.01116312,
      "auxiliary_loss_mlp": 0.01029619,
      "balance_loss_clip": 1.04337406,
      "balance_loss_mlp": 1.0220437,
      "epoch": 0.8983346359646486,
      "flos": 18076893465600.0,
      "grad_norm": 2.582372058040045,
      "language_loss": 0.78234857,
      "learning_rate": 1.0734283459412785e-07,
      "loss": 0.80380785,
      "num_input_tokens_seen": 161528995,
      "step": 7471,
      "time_per_iteration": 4.833739757537842
    },
    {
      "auxiliary_loss_clip": 0.01176469,
      "auxiliary_loss_mlp": 0.01026239,
      "balance_loss_clip": 1.0521723,
      "balance_loss_mlp": 1.01750708,
      "epoch": 0.8984548788552876,
      "flos": 20558895344640.0,
      "grad_norm": 1.9787226144489252,
      "language_loss": 0.81130385,
      "learning_rate": 1.0709120973291707e-07,
      "loss": 0.83333087,
      "num_input_tokens_seen": 161548775,
      "step": 7472,
      "time_per_iteration": 2.689734935760498
    },
    {
      "auxiliary_loss_clip": 0.01122269,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.045928,
      "balance_loss_mlp": 1.01800978,
      "epoch": 0.8985751217459268,
      "flos": 17785442511360.0,
      "grad_norm": 5.319937375559341,
      "language_loss": 0.77668273,
      "learning_rate": 1.0683987202446475e-07,
      "loss": 0.79816914,
      "num_input_tokens_seen": 161566960,
      "step": 7473,
      "time_per_iteration": 2.7358667850494385
    },
    {
      "auxiliary_loss_clip": 0.01123129,
      "auxiliary_loss_mlp": 0.0102696,
      "balance_loss_clip": 1.04594445,
      "balance_loss_mlp": 1.01834667,
      "epoch": 0.8986953646365659,
      "flos": 21617003208960.0,
      "grad_norm": 1.8196501135857719,
      "language_loss": 0.70018214,
      "learning_rate": 1.0658882150689862e-07,
      "loss": 0.72168303,
      "num_input_tokens_seen": 161585820,
      "step": 7474,
      "time_per_iteration": 2.7803854942321777
    },
    {
      "auxiliary_loss_clip": 0.01158303,
      "auxiliary_loss_mlp": 0.01024301,
      "balance_loss_clip": 1.04910922,
      "balance_loss_mlp": 1.0158844,
      "epoch": 0.8988156075272049,
      "flos": 14027355083520.0,
      "grad_norm": 2.465506273020457,
      "language_loss": 0.78559041,
      "learning_rate": 1.0633805821830288e-07,
      "loss": 0.80741644,
      "num_input_tokens_seen": 161602505,
      "step": 7475,
      "time_per_iteration": 2.6255626678466797
    },
    {
      "auxiliary_loss_clip": 0.01163326,
      "auxiliary_loss_mlp": 0.01024738,
      "balance_loss_clip": 1.04903483,
      "balance_loss_mlp": 1.01640558,
      "epoch": 0.8989358504178441,
      "flos": 29059202004480.0,
      "grad_norm": 7.9528063104551885,
      "language_loss": 0.8306936,
      "learning_rate": 1.0608758219671753e-07,
      "loss": 0.85257423,
      "num_input_tokens_seen": 161621545,
      "step": 7476,
      "time_per_iteration": 3.6828668117523193
    },
    {
      "auxiliary_loss_clip": 0.0114854,
      "auxiliary_loss_mlp": 0.01027027,
      "balance_loss_clip": 1.04617977,
      "balance_loss_mlp": 1.01974368,
      "epoch": 0.8990560933084831,
      "flos": 20230420446720.0,
      "grad_norm": 1.6372642249063045,
      "language_loss": 0.70970416,
      "learning_rate": 1.0583739348014065e-07,
      "loss": 0.73145986,
      "num_input_tokens_seen": 161642630,
      "step": 7477,
      "time_per_iteration": 2.7477853298187256
    },
    {
      "auxiliary_loss_clip": 0.0113872,
      "auxiliary_loss_mlp": 0.01027707,
      "balance_loss_clip": 1.04923725,
      "balance_loss_mlp": 1.0197916,
      "epoch": 0.8991763361991222,
      "flos": 25520672459520.0,
      "grad_norm": 1.8454117108792008,
      "language_loss": 0.85036767,
      "learning_rate": 1.0558749210652518e-07,
      "loss": 0.87203193,
      "num_input_tokens_seen": 161662560,
      "step": 7478,
      "time_per_iteration": 2.6959176063537598
    },
    {
      "auxiliary_loss_clip": 0.0114609,
      "auxiliary_loss_mlp": 0.01029246,
      "balance_loss_clip": 1.04851508,
      "balance_loss_mlp": 1.02105081,
      "epoch": 0.8992965790897613,
      "flos": 25119191168640.0,
      "grad_norm": 2.919045281100521,
      "language_loss": 0.85809314,
      "learning_rate": 1.053378781137808e-07,
      "loss": 0.87984645,
      "num_input_tokens_seen": 161683480,
      "step": 7479,
      "time_per_iteration": 2.723107099533081
    },
    {
      "auxiliary_loss_clip": 0.01102892,
      "auxiliary_loss_mlp": 0.01025931,
      "balance_loss_clip": 1.04112482,
      "balance_loss_mlp": 1.01797998,
      "epoch": 0.8994168219804004,
      "flos": 16070815814400.0,
      "grad_norm": 1.9883545349227143,
      "language_loss": 0.77963096,
      "learning_rate": 1.0508855153977392e-07,
      "loss": 0.80091918,
      "num_input_tokens_seen": 161699945,
      "step": 7480,
      "time_per_iteration": 2.7966134548187256
    },
    {
      "auxiliary_loss_clip": 0.01131193,
      "auxiliary_loss_mlp": 0.00761739,
      "balance_loss_clip": 1.04444504,
      "balance_loss_mlp": 1.00046277,
      "epoch": 0.8995370648710395,
      "flos": 24825764966400.0,
      "grad_norm": 4.08832185951716,
      "language_loss": 0.67294586,
      "learning_rate": 1.0483951242232669e-07,
      "loss": 0.69187516,
      "num_input_tokens_seen": 161720420,
      "step": 7481,
      "time_per_iteration": 2.7902684211730957
    },
    {
      "auxiliary_loss_clip": 0.01022684,
      "auxiliary_loss_mlp": 0.0100026,
      "balance_loss_clip": 1.00896585,
      "balance_loss_mlp": 0.99940819,
      "epoch": 0.8996573077616786,
      "flos": 63116238378240.0,
      "grad_norm": 0.9735753373765793,
      "language_loss": 0.57625127,
      "learning_rate": 1.0459076079921936e-07,
      "loss": 0.59648073,
      "num_input_tokens_seen": 161773080,
      "step": 7482,
      "time_per_iteration": 3.3204336166381836
    },
    {
      "auxiliary_loss_clip": 0.01066086,
      "auxiliary_loss_mlp": 0.01029495,
      "balance_loss_clip": 1.04015815,
      "balance_loss_mlp": 1.02174604,
      "epoch": 0.8997775506523177,
      "flos": 18219674027520.0,
      "grad_norm": 6.5578092243008586,
      "language_loss": 0.85113734,
      "learning_rate": 1.0434229670818618e-07,
      "loss": 0.8720932,
      "num_input_tokens_seen": 161789755,
      "step": 7483,
      "time_per_iteration": 3.007187604904175
    },
    {
      "auxiliary_loss_clip": 0.01133557,
      "auxiliary_loss_mlp": 0.01030604,
      "balance_loss_clip": 1.04713809,
      "balance_loss_mlp": 1.02292717,
      "epoch": 0.8998977935429567,
      "flos": 24166768095360.0,
      "grad_norm": 1.9019306102862532,
      "language_loss": 0.8005656,
      "learning_rate": 1.0409412018691944e-07,
      "loss": 0.82220721,
      "num_input_tokens_seen": 161810220,
      "step": 7484,
      "time_per_iteration": 3.133582353591919
    },
    {
      "auxiliary_loss_clip": 0.01099316,
      "auxiliary_loss_mlp": 0.01030286,
      "balance_loss_clip": 1.0423578,
      "balance_loss_mlp": 1.02240634,
      "epoch": 0.9000180364335959,
      "flos": 20773030273920.0,
      "grad_norm": 1.9531377524990272,
      "language_loss": 0.75487363,
      "learning_rate": 1.0384623127306724e-07,
      "loss": 0.7761696,
      "num_input_tokens_seen": 161827565,
      "step": 7485,
      "time_per_iteration": 3.7065820693969727
    },
    {
      "auxiliary_loss_clip": 0.01146372,
      "auxiliary_loss_mlp": 0.01027785,
      "balance_loss_clip": 1.0471673,
      "balance_loss_mlp": 1.02010787,
      "epoch": 0.900138279324235,
      "flos": 19205745166080.0,
      "grad_norm": 1.893073735161405,
      "language_loss": 0.79777735,
      "learning_rate": 1.0359863000423397e-07,
      "loss": 0.81951892,
      "num_input_tokens_seen": 161845700,
      "step": 7486,
      "time_per_iteration": 2.688422679901123
    },
    {
      "auxiliary_loss_clip": 0.01111091,
      "auxiliary_loss_mlp": 0.01024624,
      "balance_loss_clip": 1.04370773,
      "balance_loss_mlp": 1.01739717,
      "epoch": 0.900258522214874,
      "flos": 28731158069760.0,
      "grad_norm": 1.8612456090356428,
      "language_loss": 0.72156382,
      "learning_rate": 1.0335131641798112e-07,
      "loss": 0.74292099,
      "num_input_tokens_seen": 161867660,
      "step": 7487,
      "time_per_iteration": 2.785817861557007
    },
    {
      "auxiliary_loss_clip": 0.01020902,
      "auxiliary_loss_mlp": 0.01004014,
      "balance_loss_clip": 1.0086894,
      "balance_loss_mlp": 1.00300717,
      "epoch": 0.9003787651055132,
      "flos": 58280685655680.0,
      "grad_norm": 0.8044406282615952,
      "language_loss": 0.55509627,
      "learning_rate": 1.0310429055182512e-07,
      "loss": 0.5753454,
      "num_input_tokens_seen": 161921980,
      "step": 7488,
      "time_per_iteration": 3.1480491161346436
    },
    {
      "auxiliary_loss_clip": 0.01135227,
      "auxiliary_loss_mlp": 0.01028312,
      "balance_loss_clip": 1.04657483,
      "balance_loss_mlp": 1.02034271,
      "epoch": 0.9004990079961522,
      "flos": 25556475340800.0,
      "grad_norm": 1.7222275259968,
      "language_loss": 0.74415743,
      "learning_rate": 1.0285755244324024e-07,
      "loss": 0.76579285,
      "num_input_tokens_seen": 161942725,
      "step": 7489,
      "time_per_iteration": 3.147169589996338
    },
    {
      "auxiliary_loss_clip": 0.01138646,
      "auxiliary_loss_mlp": 0.0102627,
      "balance_loss_clip": 1.04649651,
      "balance_loss_mlp": 1.0187,
      "epoch": 0.9006192508867913,
      "flos": 23335185352320.0,
      "grad_norm": 1.8856342645716617,
      "language_loss": 0.68804914,
      "learning_rate": 1.0261110212965629e-07,
      "loss": 0.70969826,
      "num_input_tokens_seen": 161964520,
      "step": 7490,
      "time_per_iteration": 2.7433664798736572
    },
    {
      "auxiliary_loss_clip": 0.01154288,
      "auxiliary_loss_mlp": 0.01032178,
      "balance_loss_clip": 1.05114555,
      "balance_loss_mlp": 1.02437615,
      "epoch": 0.9007394937774305,
      "flos": 18040300485120.0,
      "grad_norm": 2.1758041689428826,
      "language_loss": 0.79379559,
      "learning_rate": 1.023649396484596e-07,
      "loss": 0.81566024,
      "num_input_tokens_seen": 161983575,
      "step": 7491,
      "time_per_iteration": 2.7258694171905518
    },
    {
      "auxiliary_loss_clip": 0.01103937,
      "auxiliary_loss_mlp": 0.01025256,
      "balance_loss_clip": 1.04340363,
      "balance_loss_mlp": 1.01779044,
      "epoch": 0.9008597366680695,
      "flos": 43068456633600.0,
      "grad_norm": 2.3174718311068334,
      "language_loss": 0.6735844,
      "learning_rate": 1.0211906503699275e-07,
      "loss": 0.69487631,
      "num_input_tokens_seen": 162006550,
      "step": 7492,
      "time_per_iteration": 2.9346392154693604
    },
    {
      "auxiliary_loss_clip": 0.01143074,
      "auxiliary_loss_mlp": 0.01031815,
      "balance_loss_clip": 1.04627395,
      "balance_loss_mlp": 1.0236994,
      "epoch": 0.9009799795587086,
      "flos": 14939055112320.0,
      "grad_norm": 8.151039351663508,
      "language_loss": 0.82244289,
      "learning_rate": 1.0187347833255455e-07,
      "loss": 0.84419179,
      "num_input_tokens_seen": 162022455,
      "step": 7493,
      "time_per_iteration": 2.708005666732788
    },
    {
      "auxiliary_loss_clip": 0.01125067,
      "auxiliary_loss_mlp": 0.01027569,
      "balance_loss_clip": 1.04500175,
      "balance_loss_mlp": 1.01973724,
      "epoch": 0.9011002224493477,
      "flos": 21579584215680.0,
      "grad_norm": 1.8338825188916188,
      "language_loss": 0.7929967,
      "learning_rate": 1.0162817957240056e-07,
      "loss": 0.81452304,
      "num_input_tokens_seen": 162042350,
      "step": 7494,
      "time_per_iteration": 2.7324538230895996
    },
    {
      "auxiliary_loss_clip": 0.01052676,
      "auxiliary_loss_mlp": 0.01002171,
      "balance_loss_clip": 1.00739145,
      "balance_loss_mlp": 1.00119996,
      "epoch": 0.9012204653399868,
      "flos": 71166367883520.0,
      "grad_norm": 0.9492951337052733,
      "language_loss": 0.62956244,
      "learning_rate": 1.0138316879374253e-07,
      "loss": 0.65011084,
      "num_input_tokens_seen": 162111640,
      "step": 7495,
      "time_per_iteration": 3.4377734661102295
    },
    {
      "auxiliary_loss_clip": 0.0116321,
      "auxiliary_loss_mlp": 0.01029106,
      "balance_loss_clip": 1.04996765,
      "balance_loss_mlp": 1.02161944,
      "epoch": 0.9013407082306258,
      "flos": 15594963413760.0,
      "grad_norm": 2.561381343806623,
      "language_loss": 0.74120843,
      "learning_rate": 1.0113844603374833e-07,
      "loss": 0.76313162,
      "num_input_tokens_seen": 162128165,
      "step": 7496,
      "time_per_iteration": 2.7082467079162598
    },
    {
      "auxiliary_loss_clip": 0.01128919,
      "auxiliary_loss_mlp": 0.007626,
      "balance_loss_clip": 1.04578161,
      "balance_loss_mlp": 1.00049472,
      "epoch": 0.901460951121265,
      "flos": 15049157276160.0,
      "grad_norm": 4.094089878123296,
      "language_loss": 0.71890926,
      "learning_rate": 1.0089401132954178e-07,
      "loss": 0.73782444,
      "num_input_tokens_seen": 162146145,
      "step": 7497,
      "time_per_iteration": 3.625793695449829
    },
    {
      "auxiliary_loss_clip": 0.01119959,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.04535687,
      "balance_loss_mlp": 1.02112222,
      "epoch": 0.9015811940119041,
      "flos": 22236857233920.0,
      "grad_norm": 1.9280610033550156,
      "language_loss": 0.7232362,
      "learning_rate": 1.006498647182037e-07,
      "loss": 0.74472308,
      "num_input_tokens_seen": 162164800,
      "step": 7498,
      "time_per_iteration": 2.814945697784424
    },
    {
      "auxiliary_loss_clip": 0.01149388,
      "auxiliary_loss_mlp": 0.01027058,
      "balance_loss_clip": 1.04952121,
      "balance_loss_mlp": 1.01947033,
      "epoch": 0.9017014369025431,
      "flos": 24973824827520.0,
      "grad_norm": 2.093927301939855,
      "language_loss": 0.71568418,
      "learning_rate": 1.004060062367713e-07,
      "loss": 0.73744869,
      "num_input_tokens_seen": 162185895,
      "step": 7499,
      "time_per_iteration": 2.7292420864105225
    },
    {
      "auxiliary_loss_clip": 0.01147727,
      "auxiliary_loss_mlp": 0.01028456,
      "balance_loss_clip": 1.04801893,
      "balance_loss_mlp": 1.02071917,
      "epoch": 0.9018216797931822,
      "flos": 18114168804480.0,
      "grad_norm": 3.797881208480783,
      "language_loss": 0.69600219,
      "learning_rate": 1.0016243592223728e-07,
      "loss": 0.71776402,
      "num_input_tokens_seen": 162206295,
      "step": 7500,
      "time_per_iteration": 2.6435022354125977
    },
    {
      "auxiliary_loss_clip": 0.0116052,
      "auxiliary_loss_mlp": 0.01022458,
      "balance_loss_clip": 1.05059755,
      "balance_loss_mlp": 1.01521301,
      "epoch": 0.9019419226838213,
      "flos": 37268452759680.0,
      "grad_norm": 2.2828969264763526,
      "language_loss": 0.65594774,
      "learning_rate": 9.991915381155114e-08,
      "loss": 0.67777753,
      "num_input_tokens_seen": 162229275,
      "step": 7501,
      "time_per_iteration": 2.8186190128326416
    },
    {
      "auxiliary_loss_clip": 0.0116047,
      "auxiliary_loss_mlp": 0.00762465,
      "balance_loss_clip": 1.05304706,
      "balance_loss_mlp": 1.0004741,
      "epoch": 0.9020621655744604,
      "flos": 23441121538560.0,
      "grad_norm": 2.8214206756578872,
      "language_loss": 0.7514717,
      "learning_rate": 9.967615994161871e-08,
      "loss": 0.77070105,
      "num_input_tokens_seen": 162248935,
      "step": 7502,
      "time_per_iteration": 3.6342954635620117
    },
    {
      "auxiliary_loss_clip": 0.01105066,
      "auxiliary_loss_mlp": 0.01026753,
      "balance_loss_clip": 1.04196215,
      "balance_loss_mlp": 1.01905823,
      "epoch": 0.9021824084650995,
      "flos": 22857465444480.0,
      "grad_norm": 3.294826057968253,
      "language_loss": 0.78373861,
      "learning_rate": 9.943345434930161e-08,
      "loss": 0.80505681,
      "num_input_tokens_seen": 162269185,
      "step": 7503,
      "time_per_iteration": 2.8470849990844727
    },
    {
      "auxiliary_loss_clip": 0.01146854,
      "auxiliary_loss_mlp": 0.01023888,
      "balance_loss_clip": 1.04768682,
      "balance_loss_mlp": 1.01630688,
      "epoch": 0.9023026513557386,
      "flos": 22127581082880.0,
      "grad_norm": 1.9313517729679333,
      "language_loss": 0.69108176,
      "learning_rate": 9.919103707141885e-08,
      "loss": 0.71278924,
      "num_input_tokens_seen": 162288065,
      "step": 7504,
      "time_per_iteration": 2.753631830215454
    },
    {
      "auxiliary_loss_clip": 0.0116455,
      "auxiliary_loss_mlp": 0.0102625,
      "balance_loss_clip": 1.04987037,
      "balance_loss_mlp": 1.01811981,
      "epoch": 0.9024228942463777,
      "flos": 24199087357440.0,
      "grad_norm": 2.4370545886254797,
      "language_loss": 0.76647562,
      "learning_rate": 9.89489081447441e-08,
      "loss": 0.7883836,
      "num_input_tokens_seen": 162305265,
      "step": 7505,
      "time_per_iteration": 2.624345302581787
    },
    {
      "auxiliary_loss_clip": 0.01132199,
      "auxiliary_loss_mlp": 0.01026,
      "balance_loss_clip": 1.04794669,
      "balance_loss_mlp": 1.0184449,
      "epoch": 0.9025431371370167,
      "flos": 25008262992000.0,
      "grad_norm": 2.1391363803721197,
      "language_loss": 0.82838815,
      "learning_rate": 9.870706760600844e-08,
      "loss": 0.8499701,
      "num_input_tokens_seen": 162325215,
      "step": 7506,
      "time_per_iteration": 2.799030065536499
    },
    {
      "auxiliary_loss_clip": 0.0113454,
      "auxiliary_loss_mlp": 0.01031368,
      "balance_loss_clip": 1.04886341,
      "balance_loss_mlp": 1.02320814,
      "epoch": 0.9026633800276559,
      "flos": 18952862440320.0,
      "grad_norm": 3.644535241183758,
      "language_loss": 0.72604907,
      "learning_rate": 9.846551549189918e-08,
      "loss": 0.7477082,
      "num_input_tokens_seen": 162344820,
      "step": 7507,
      "time_per_iteration": 2.6692259311676025
    },
    {
      "auxiliary_loss_clip": 0.01126314,
      "auxiliary_loss_mlp": 0.01027702,
      "balance_loss_clip": 1.04468858,
      "balance_loss_mlp": 1.01988149,
      "epoch": 0.902783622918295,
      "flos": 32416059536640.0,
      "grad_norm": 3.656622583643292,
      "language_loss": 0.68817866,
      "learning_rate": 9.822425183905902e-08,
      "loss": 0.70971876,
      "num_input_tokens_seen": 162365345,
      "step": 7508,
      "time_per_iteration": 2.83388614654541
    },
    {
      "auxiliary_loss_clip": 0.01018841,
      "auxiliary_loss_mlp": 0.00999994,
      "balance_loss_clip": 1.01608849,
      "balance_loss_mlp": 0.99904025,
      "epoch": 0.902903865808934,
      "flos": 63717453244800.0,
      "grad_norm": 0.9097600849494125,
      "language_loss": 0.75104511,
      "learning_rate": 9.798327668408823e-08,
      "loss": 0.7712335,
      "num_input_tokens_seen": 162426980,
      "step": 7509,
      "time_per_iteration": 3.394537925720215
    },
    {
      "auxiliary_loss_clip": 0.01103749,
      "auxiliary_loss_mlp": 0.01034052,
      "balance_loss_clip": 1.04076004,
      "balance_loss_mlp": 1.02610648,
      "epoch": 0.9030241086995732,
      "flos": 23804034600960.0,
      "grad_norm": 2.0115459492094083,
      "language_loss": 0.6924088,
      "learning_rate": 9.774259006354158e-08,
      "loss": 0.71378684,
      "num_input_tokens_seen": 162447050,
      "step": 7510,
      "time_per_iteration": 2.945988893508911
    },
    {
      "auxiliary_loss_clip": 0.011377,
      "auxiliary_loss_mlp": 0.01033906,
      "balance_loss_clip": 1.04338861,
      "balance_loss_mlp": 1.02591348,
      "epoch": 0.9031443515902122,
      "flos": 26395887248640.0,
      "grad_norm": 2.277769572733898,
      "language_loss": 0.7629416,
      "learning_rate": 9.750219201393184e-08,
      "loss": 0.7846576,
      "num_input_tokens_seen": 162467015,
      "step": 7511,
      "time_per_iteration": 3.9353349208831787
    },
    {
      "auxiliary_loss_clip": 0.01130935,
      "auxiliary_loss_mlp": 0.01027715,
      "balance_loss_clip": 1.04636765,
      "balance_loss_mlp": 1.02011275,
      "epoch": 0.9032645944808513,
      "flos": 24939350749440.0,
      "grad_norm": 1.8677049033795767,
      "language_loss": 0.77941984,
      "learning_rate": 9.726208257172697e-08,
      "loss": 0.80100632,
      "num_input_tokens_seen": 162488710,
      "step": 7512,
      "time_per_iteration": 2.7413535118103027
    },
    {
      "auxiliary_loss_clip": 0.01098829,
      "auxiliary_loss_mlp": 0.00761905,
      "balance_loss_clip": 1.04648387,
      "balance_loss_mlp": 1.00053251,
      "epoch": 0.9033848373714904,
      "flos": 21178821196800.0,
      "grad_norm": 1.9299643442690377,
      "language_loss": 0.7449888,
      "learning_rate": 9.702226177335115e-08,
      "loss": 0.76359612,
      "num_input_tokens_seen": 162507205,
      "step": 7513,
      "time_per_iteration": 2.8126089572906494
    },
    {
      "auxiliary_loss_clip": 0.01095121,
      "auxiliary_loss_mlp": 0.01033001,
      "balance_loss_clip": 1.04485047,
      "balance_loss_mlp": 1.0247277,
      "epoch": 0.9035050802621295,
      "flos": 26286359702400.0,
      "grad_norm": 3.0475707707605793,
      "language_loss": 0.72694731,
      "learning_rate": 9.67827296551853e-08,
      "loss": 0.74822849,
      "num_input_tokens_seen": 162528490,
      "step": 7514,
      "time_per_iteration": 3.006838798522949
    },
    {
      "auxiliary_loss_clip": 0.0113567,
      "auxiliary_loss_mlp": 0.01022872,
      "balance_loss_clip": 1.0459609,
      "balance_loss_mlp": 1.0153203,
      "epoch": 0.9036253231527686,
      "flos": 24204546224640.0,
      "grad_norm": 2.3254674071977925,
      "language_loss": 0.68266779,
      "learning_rate": 9.65434862535659e-08,
      "loss": 0.7042532,
      "num_input_tokens_seen": 162547860,
      "step": 7515,
      "time_per_iteration": 3.238654136657715
    },
    {
      "auxiliary_loss_clip": 0.01161857,
      "auxiliary_loss_mlp": 0.01035807,
      "balance_loss_clip": 1.04946709,
      "balance_loss_mlp": 1.02793956,
      "epoch": 0.9037455660434077,
      "flos": 18072655660800.0,
      "grad_norm": 3.892722516670127,
      "language_loss": 0.6536485,
      "learning_rate": 9.630453160478635e-08,
      "loss": 0.67562509,
      "num_input_tokens_seen": 162563215,
      "step": 7516,
      "time_per_iteration": 2.6757171154022217
    },
    {
      "auxiliary_loss_clip": 0.01135677,
      "auxiliary_loss_mlp": 0.01034657,
      "balance_loss_clip": 1.04628837,
      "balance_loss_mlp": 1.0271883,
      "epoch": 0.9038658089340468,
      "flos": 24060795995520.0,
      "grad_norm": 1.7185308652143771,
      "language_loss": 0.82710505,
      "learning_rate": 9.60658657450959e-08,
      "loss": 0.84880829,
      "num_input_tokens_seen": 162583515,
      "step": 7517,
      "time_per_iteration": 2.683513879776001
    },
    {
      "auxiliary_loss_clip": 0.0114507,
      "auxiliary_loss_mlp": 0.01027929,
      "balance_loss_clip": 1.04518795,
      "balance_loss_mlp": 1.02050817,
      "epoch": 0.9039860518246858,
      "flos": 21834298535040.0,
      "grad_norm": 1.7550904606558548,
      "language_loss": 0.79629219,
      "learning_rate": 9.582748871069979e-08,
      "loss": 0.81802213,
      "num_input_tokens_seen": 162602955,
      "step": 7518,
      "time_per_iteration": 2.683922529220581
    },
    {
      "auxiliary_loss_clip": 0.01145659,
      "auxiliary_loss_mlp": 0.01030888,
      "balance_loss_clip": 1.05023217,
      "balance_loss_mlp": 1.02366734,
      "epoch": 0.904106294715325,
      "flos": 26614870513920.0,
      "grad_norm": 3.394594271510313,
      "language_loss": 0.8341006,
      "learning_rate": 9.558940053775954e-08,
      "loss": 0.85586601,
      "num_input_tokens_seen": 162621595,
      "step": 7519,
      "time_per_iteration": 2.7167882919311523
    },
    {
      "auxiliary_loss_clip": 0.01130208,
      "auxiliary_loss_mlp": 0.01025089,
      "balance_loss_clip": 1.04722261,
      "balance_loss_mlp": 1.0170722,
      "epoch": 0.904226537605964,
      "flos": 17785693906560.0,
      "grad_norm": 1.8695829287763894,
      "language_loss": 0.68136615,
      "learning_rate": 9.535160126239294e-08,
      "loss": 0.70291913,
      "num_input_tokens_seen": 162638220,
      "step": 7520,
      "time_per_iteration": 2.7698371410369873
    },
    {
      "auxiliary_loss_clip": 0.0110714,
      "auxiliary_loss_mlp": 0.01027184,
      "balance_loss_clip": 1.04181814,
      "balance_loss_mlp": 1.01966155,
      "epoch": 0.9043467804966031,
      "flos": 24790428961920.0,
      "grad_norm": 1.5511840059682878,
      "language_loss": 0.70934379,
      "learning_rate": 9.511409092067424e-08,
      "loss": 0.73068708,
      "num_input_tokens_seen": 162658575,
      "step": 7521,
      "time_per_iteration": 2.817744016647339
    },
    {
      "auxiliary_loss_clip": 0.01137672,
      "auxiliary_loss_mlp": 0.01029168,
      "balance_loss_clip": 1.04381514,
      "balance_loss_mlp": 1.021384,
      "epoch": 0.9044670233872423,
      "flos": 22632125472000.0,
      "grad_norm": 1.9084260735951062,
      "language_loss": 0.67349881,
      "learning_rate": 9.487686954863327e-08,
      "loss": 0.69516718,
      "num_input_tokens_seen": 162678295,
      "step": 7522,
      "time_per_iteration": 2.704944133758545
    },
    {
      "auxiliary_loss_clip": 0.01071886,
      "auxiliary_loss_mlp": 0.01027173,
      "balance_loss_clip": 1.04003239,
      "balance_loss_mlp": 1.01934659,
      "epoch": 0.9045872662778813,
      "flos": 23771320289280.0,
      "grad_norm": 8.503630425358363,
      "language_loss": 0.78146893,
      "learning_rate": 9.46399371822566e-08,
      "loss": 0.80245948,
      "num_input_tokens_seen": 162698070,
      "step": 7523,
      "time_per_iteration": 3.90309476852417
    },
    {
      "auxiliary_loss_clip": 0.01118612,
      "auxiliary_loss_mlp": 0.01038098,
      "balance_loss_clip": 1.04716563,
      "balance_loss_mlp": 1.02993798,
      "epoch": 0.9047075091685204,
      "flos": 15191039998080.0,
      "grad_norm": 2.991516745993287,
      "language_loss": 0.7259208,
      "learning_rate": 9.440329385748657e-08,
      "loss": 0.7474879,
      "num_input_tokens_seen": 162715140,
      "step": 7524,
      "time_per_iteration": 3.173050880432129
    },
    {
      "auxiliary_loss_clip": 0.01115772,
      "auxiliary_loss_mlp": 0.01025184,
      "balance_loss_clip": 1.04513454,
      "balance_loss_mlp": 1.01863682,
      "epoch": 0.9048277520591596,
      "flos": 18003707504640.0,
      "grad_norm": 2.3795509435784976,
      "language_loss": 0.7067734,
      "learning_rate": 9.416693961022137e-08,
      "loss": 0.72818303,
      "num_input_tokens_seen": 162733390,
      "step": 7525,
      "time_per_iteration": 2.82088565826416
    },
    {
      "auxiliary_loss_clip": 0.01110803,
      "auxiliary_loss_mlp": 0.01028392,
      "balance_loss_clip": 1.04018998,
      "balance_loss_mlp": 1.02086401,
      "epoch": 0.9049479949497986,
      "flos": 21872471713920.0,
      "grad_norm": 1.8259172036975089,
      "language_loss": 0.77181607,
      "learning_rate": 9.393087447631654e-08,
      "loss": 0.793208,
      "num_input_tokens_seen": 162751670,
      "step": 7526,
      "time_per_iteration": 2.7401795387268066
    },
    {
      "auxiliary_loss_clip": 0.01168534,
      "auxiliary_loss_mlp": 0.0102444,
      "balance_loss_clip": 1.04931509,
      "balance_loss_mlp": 1.01757383,
      "epoch": 0.9050682378404377,
      "flos": 20773928113920.0,
      "grad_norm": 1.7283419987688897,
      "language_loss": 0.72947747,
      "learning_rate": 9.36950984915823e-08,
      "loss": 0.75140715,
      "num_input_tokens_seen": 162770025,
      "step": 7527,
      "time_per_iteration": 3.8030004501342773
    },
    {
      "auxiliary_loss_clip": 0.01178676,
      "auxiliary_loss_mlp": 0.01031081,
      "balance_loss_clip": 1.05481374,
      "balance_loss_mlp": 1.02256346,
      "epoch": 0.9051884807310768,
      "flos": 21580015178880.0,
      "grad_norm": 1.7389798240403747,
      "language_loss": 0.69144559,
      "learning_rate": 9.345961169178607e-08,
      "loss": 0.71354318,
      "num_input_tokens_seen": 162789710,
      "step": 7528,
      "time_per_iteration": 2.6926112174987793
    },
    {
      "auxiliary_loss_clip": 0.01138041,
      "auxiliary_loss_mlp": 0.01030804,
      "balance_loss_clip": 1.05218315,
      "balance_loss_mlp": 1.02325249,
      "epoch": 0.9053087236217159,
      "flos": 21908059113600.0,
      "grad_norm": 1.457352695100005,
      "language_loss": 0.72890544,
      "learning_rate": 9.322441411265081e-08,
      "loss": 0.7505939,
      "num_input_tokens_seen": 162810695,
      "step": 7529,
      "time_per_iteration": 2.7159950733184814
    },
    {
      "auxiliary_loss_clip": 0.01112877,
      "auxiliary_loss_mlp": 0.01029371,
      "balance_loss_clip": 1.0476855,
      "balance_loss_mlp": 1.02151561,
      "epoch": 0.9054289665123549,
      "flos": 17055809544960.0,
      "grad_norm": 1.8996729166658621,
      "language_loss": 0.72988307,
      "learning_rate": 9.298950578985554e-08,
      "loss": 0.75130552,
      "num_input_tokens_seen": 162827770,
      "step": 7530,
      "time_per_iteration": 2.7457268238067627
    },
    {
      "auxiliary_loss_clip": 0.011091,
      "auxiliary_loss_mlp": 0.01029751,
      "balance_loss_clip": 1.04120958,
      "balance_loss_mlp": 1.02138877,
      "epoch": 0.905549209402994,
      "flos": 20777268078720.0,
      "grad_norm": 1.7487370470130843,
      "language_loss": 0.71204579,
      "learning_rate": 9.275488675903665e-08,
      "loss": 0.73343432,
      "num_input_tokens_seen": 162846715,
      "step": 7531,
      "time_per_iteration": 2.727229595184326
    },
    {
      "auxiliary_loss_clip": 0.0113615,
      "auxiliary_loss_mlp": 0.01034624,
      "balance_loss_clip": 1.04600024,
      "balance_loss_mlp": 1.02642846,
      "epoch": 0.9056694522936332,
      "flos": 21686813291520.0,
      "grad_norm": 7.547943058242803,
      "language_loss": 0.73964697,
      "learning_rate": 9.252055705578454e-08,
      "loss": 0.76135468,
      "num_input_tokens_seen": 162866215,
      "step": 7532,
      "time_per_iteration": 2.724851608276367
    },
    {
      "auxiliary_loss_clip": 0.01073606,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.04010558,
      "balance_loss_mlp": 1.01964951,
      "epoch": 0.9057896951842722,
      "flos": 29569133433600.0,
      "grad_norm": 1.7943384464166274,
      "language_loss": 0.72359538,
      "learning_rate": 9.228651671564747e-08,
      "loss": 0.74460161,
      "num_input_tokens_seen": 162888245,
      "step": 7533,
      "time_per_iteration": 2.916306495666504
    },
    {
      "auxiliary_loss_clip": 0.01162777,
      "auxiliary_loss_mlp": 0.01032612,
      "balance_loss_clip": 1.05167639,
      "balance_loss_mlp": 1.02556133,
      "epoch": 0.9059099380749113,
      "flos": 27892248952320.0,
      "grad_norm": 1.530780432500989,
      "language_loss": 0.77867746,
      "learning_rate": 9.205276577412901e-08,
      "loss": 0.80063134,
      "num_input_tokens_seen": 162911025,
      "step": 7534,
      "time_per_iteration": 2.7089455127716064
    },
    {
      "auxiliary_loss_clip": 0.01173585,
      "auxiliary_loss_mlp": 0.01029577,
      "balance_loss_clip": 1.05025601,
      "balance_loss_mlp": 1.02149487,
      "epoch": 0.9060301809655504,
      "flos": 17748993185280.0,
      "grad_norm": 2.7520246385008513,
      "language_loss": 0.76932991,
      "learning_rate": 9.181930426668905e-08,
      "loss": 0.79136145,
      "num_input_tokens_seen": 162927820,
      "step": 7535,
      "time_per_iteration": 2.641101121902466
    },
    {
      "auxiliary_loss_clip": 0.01158095,
      "auxiliary_loss_mlp": 0.01029396,
      "balance_loss_clip": 1.05161655,
      "balance_loss_mlp": 1.02209163,
      "epoch": 0.9061504238561895,
      "flos": 31759432963200.0,
      "grad_norm": 1.7089015205477078,
      "language_loss": 0.68188393,
      "learning_rate": 9.158613222874346e-08,
      "loss": 0.70375878,
      "num_input_tokens_seen": 162949445,
      "step": 7536,
      "time_per_iteration": 2.7481727600097656
    },
    {
      "auxiliary_loss_clip": 0.01127824,
      "auxiliary_loss_mlp": 0.01024329,
      "balance_loss_clip": 1.04433703,
      "balance_loss_mlp": 1.01672316,
      "epoch": 0.9062706667468285,
      "flos": 20048066075520.0,
      "grad_norm": 1.654924747791527,
      "language_loss": 0.82209432,
      "learning_rate": 9.135324969566394e-08,
      "loss": 0.84361589,
      "num_input_tokens_seen": 162968945,
      "step": 7537,
      "time_per_iteration": 3.654546022415161
    },
    {
      "auxiliary_loss_clip": 0.01162298,
      "auxiliary_loss_mlp": 0.01031429,
      "balance_loss_clip": 1.05257308,
      "balance_loss_mlp": 1.02346587,
      "epoch": 0.9063909096374677,
      "flos": 18437292576000.0,
      "grad_norm": 2.245726710702246,
      "language_loss": 0.75695884,
      "learning_rate": 9.112065670277913e-08,
      "loss": 0.77889609,
      "num_input_tokens_seen": 162985310,
      "step": 7538,
      "time_per_iteration": 2.6849167346954346
    },
    {
      "auxiliary_loss_clip": 0.01154043,
      "auxiliary_loss_mlp": 0.0076153,
      "balance_loss_clip": 1.0492115,
      "balance_loss_mlp": 1.00050163,
      "epoch": 0.9065111525281068,
      "flos": 33547353361920.0,
      "grad_norm": 2.5621278847726563,
      "language_loss": 0.73156846,
      "learning_rate": 9.088835328537303e-08,
      "loss": 0.7507242,
      "num_input_tokens_seen": 163006900,
      "step": 7539,
      "time_per_iteration": 2.775949478149414
    },
    {
      "auxiliary_loss_clip": 0.01175111,
      "auxiliary_loss_mlp": 0.01030279,
      "balance_loss_clip": 1.05287862,
      "balance_loss_mlp": 1.02256012,
      "epoch": 0.9066313954187458,
      "flos": 23367863750400.0,
      "grad_norm": 2.7530973881300813,
      "language_loss": 0.71693397,
      "learning_rate": 9.065633947868568e-08,
      "loss": 0.73898792,
      "num_input_tokens_seen": 163026505,
      "step": 7540,
      "time_per_iteration": 2.663296699523926
    },
    {
      "auxiliary_loss_clip": 0.01115481,
      "auxiliary_loss_mlp": 0.01028356,
      "balance_loss_clip": 1.04217935,
      "balance_loss_mlp": 1.02097988,
      "epoch": 0.906751638309385,
      "flos": 26249623067520.0,
      "grad_norm": 2.447572768566424,
      "language_loss": 0.80145144,
      "learning_rate": 9.042461531791379e-08,
      "loss": 0.8228898,
      "num_input_tokens_seen": 163044925,
      "step": 7541,
      "time_per_iteration": 2.7427024841308594
    },
    {
      "auxiliary_loss_clip": 0.01157883,
      "auxiliary_loss_mlp": 0.01022747,
      "balance_loss_clip": 1.05045688,
      "balance_loss_mlp": 1.01568127,
      "epoch": 0.906871881200024,
      "flos": 16544477485440.0,
      "grad_norm": 2.4131795852802194,
      "language_loss": 0.78388888,
      "learning_rate": 9.019318083820903e-08,
      "loss": 0.80569518,
      "num_input_tokens_seen": 163063505,
      "step": 7542,
      "time_per_iteration": 2.649381160736084
    },
    {
      "auxiliary_loss_clip": 0.01147097,
      "auxiliary_loss_mlp": 0.01031546,
      "balance_loss_clip": 1.04821348,
      "balance_loss_mlp": 1.02261734,
      "epoch": 0.9069921240906631,
      "flos": 24605129675520.0,
      "grad_norm": 1.6789183300369572,
      "language_loss": 0.8559711,
      "learning_rate": 8.996203607468045e-08,
      "loss": 0.87775755,
      "num_input_tokens_seen": 163082505,
      "step": 7543,
      "time_per_iteration": 2.6889524459838867
    },
    {
      "auxiliary_loss_clip": 0.01086025,
      "auxiliary_loss_mlp": 0.01023146,
      "balance_loss_clip": 1.0394665,
      "balance_loss_mlp": 1.01563597,
      "epoch": 0.9071123669813023,
      "flos": 25374731500800.0,
      "grad_norm": 1.3844121840600294,
      "language_loss": 0.75600219,
      "learning_rate": 8.973118106239241e-08,
      "loss": 0.77709389,
      "num_input_tokens_seen": 163105110,
      "step": 7544,
      "time_per_iteration": 2.8879706859588623
    },
    {
      "auxiliary_loss_clip": 0.0111552,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 1.04144931,
      "balance_loss_mlp": 1.02035618,
      "epoch": 0.9072326098719413,
      "flos": 26725798690560.0,
      "grad_norm": 2.034777178906664,
      "language_loss": 0.95015776,
      "learning_rate": 8.95006158363656e-08,
      "loss": 0.97159308,
      "num_input_tokens_seen": 163125295,
      "step": 7545,
      "time_per_iteration": 2.7363805770874023
    },
    {
      "auxiliary_loss_clip": 0.01155042,
      "auxiliary_loss_mlp": 0.00761913,
      "balance_loss_clip": 1.05039382,
      "balance_loss_mlp": 1.00045633,
      "epoch": 0.9073528527625804,
      "flos": 23878800760320.0,
      "grad_norm": 8.089476437974847,
      "language_loss": 0.77765077,
      "learning_rate": 8.9270340431576e-08,
      "loss": 0.79682028,
      "num_input_tokens_seen": 163144385,
      "step": 7546,
      "time_per_iteration": 2.7527356147766113
    },
    {
      "auxiliary_loss_clip": 0.01150484,
      "auxiliary_loss_mlp": 0.01024025,
      "balance_loss_clip": 1.04619789,
      "balance_loss_mlp": 1.01573968,
      "epoch": 0.9074730956532195,
      "flos": 37852144767360.0,
      "grad_norm": 2.3111723966106337,
      "language_loss": 0.73489195,
      "learning_rate": 8.904035488295658e-08,
      "loss": 0.75663704,
      "num_input_tokens_seen": 163163885,
      "step": 7547,
      "time_per_iteration": 2.777773141860962
    },
    {
      "auxiliary_loss_clip": 0.01026031,
      "auxiliary_loss_mlp": 0.01002842,
      "balance_loss_clip": 1.00782084,
      "balance_loss_mlp": 1.00186467,
      "epoch": 0.9075933385438586,
      "flos": 65173307385600.0,
      "grad_norm": 0.6606695376667002,
      "language_loss": 0.53249037,
      "learning_rate": 8.881065922539632e-08,
      "loss": 0.55277908,
      "num_input_tokens_seen": 163224325,
      "step": 7548,
      "time_per_iteration": 3.203157424926758
    },
    {
      "auxiliary_loss_clip": 0.01148455,
      "auxiliary_loss_mlp": 0.01028718,
      "balance_loss_clip": 1.04759336,
      "balance_loss_mlp": 1.0210495,
      "epoch": 0.9077135814344977,
      "flos": 19931571290880.0,
      "grad_norm": 2.794062718046703,
      "language_loss": 0.73333704,
      "learning_rate": 8.85812534937389e-08,
      "loss": 0.75510877,
      "num_input_tokens_seen": 163242425,
      "step": 7549,
      "time_per_iteration": 3.746971607208252
    },
    {
      "auxiliary_loss_clip": 0.01124527,
      "auxiliary_loss_mlp": 0.01031718,
      "balance_loss_clip": 1.04763675,
      "balance_loss_mlp": 1.02355194,
      "epoch": 0.9078338243251368,
      "flos": 17529650784000.0,
      "grad_norm": 4.6848195456688035,
      "language_loss": 0.67869264,
      "learning_rate": 8.835213772278583e-08,
      "loss": 0.7002551,
      "num_input_tokens_seen": 163259280,
      "step": 7550,
      "time_per_iteration": 2.745197296142578
    },
    {
      "auxiliary_loss_clip": 0.01169722,
      "auxiliary_loss_mlp": 0.0102345,
      "balance_loss_clip": 1.04962182,
      "balance_loss_mlp": 1.01635706,
      "epoch": 0.9079540672157759,
      "flos": 28803410277120.0,
      "grad_norm": 2.711665362442525,
      "language_loss": 0.78767645,
      "learning_rate": 8.812331194729373e-08,
      "loss": 0.80960822,
      "num_input_tokens_seen": 163278925,
      "step": 7551,
      "time_per_iteration": 2.6821935176849365
    },
    {
      "auxiliary_loss_clip": 0.01179259,
      "auxiliary_loss_mlp": 0.01043992,
      "balance_loss_clip": 1.0551945,
      "balance_loss_mlp": 1.03576684,
      "epoch": 0.9080743101064149,
      "flos": 23513840622720.0,
      "grad_norm": 2.5344596769169687,
      "language_loss": 0.72141701,
      "learning_rate": 8.789477620197461e-08,
      "loss": 0.74364948,
      "num_input_tokens_seen": 163298450,
      "step": 7552,
      "time_per_iteration": 2.570619821548462
    },
    {
      "auxiliary_loss_clip": 0.01130815,
      "auxiliary_loss_mlp": 0.0103348,
      "balance_loss_clip": 1.04824281,
      "balance_loss_mlp": 1.02543306,
      "epoch": 0.9081945529970541,
      "flos": 22778102344320.0,
      "grad_norm": 2.772946992411759,
      "language_loss": 0.78961122,
      "learning_rate": 8.766653052149831e-08,
      "loss": 0.8112542,
      "num_input_tokens_seen": 163313635,
      "step": 7553,
      "time_per_iteration": 3.6840243339538574
    },
    {
      "auxiliary_loss_clip": 0.01151345,
      "auxiliary_loss_mlp": 0.00762156,
      "balance_loss_clip": 1.0471791,
      "balance_loss_mlp": 1.00052381,
      "epoch": 0.9083147958876931,
      "flos": 18873714821760.0,
      "grad_norm": 2.4570640795416687,
      "language_loss": 0.74841756,
      "learning_rate": 8.743857494048823e-08,
      "loss": 0.76755261,
      "num_input_tokens_seen": 163330450,
      "step": 7554,
      "time_per_iteration": 2.695275068283081
    },
    {
      "auxiliary_loss_clip": 0.01156452,
      "auxiliary_loss_mlp": 0.01024225,
      "balance_loss_clip": 1.05227256,
      "balance_loss_mlp": 1.01686692,
      "epoch": 0.9084350387783322,
      "flos": 18909374048640.0,
      "grad_norm": 2.295095153880438,
      "language_loss": 0.62475502,
      "learning_rate": 8.721090949352605e-08,
      "loss": 0.6465618,
      "num_input_tokens_seen": 163346690,
      "step": 7555,
      "time_per_iteration": 2.6451523303985596
    },
    {
      "auxiliary_loss_clip": 0.01158173,
      "auxiliary_loss_mlp": 0.01032045,
      "balance_loss_clip": 1.05166125,
      "balance_loss_mlp": 1.02333128,
      "epoch": 0.9085552816689714,
      "flos": 20595488325120.0,
      "grad_norm": 1.9646336589062914,
      "language_loss": 0.72812271,
      "learning_rate": 8.698353421514793e-08,
      "loss": 0.75002491,
      "num_input_tokens_seen": 163365065,
      "step": 7556,
      "time_per_iteration": 2.723792314529419
    },
    {
      "auxiliary_loss_clip": 0.01138392,
      "auxiliary_loss_mlp": 0.0103101,
      "balance_loss_clip": 1.04593444,
      "balance_loss_mlp": 1.02289224,
      "epoch": 0.9086755245596104,
      "flos": 18113163223680.0,
      "grad_norm": 3.968934208799892,
      "language_loss": 0.80685854,
      "learning_rate": 8.67564491398467e-08,
      "loss": 0.8285526,
      "num_input_tokens_seen": 163382070,
      "step": 7557,
      "time_per_iteration": 2.621654987335205
    },
    {
      "auxiliary_loss_clip": 0.01138676,
      "auxiliary_loss_mlp": 0.01028096,
      "balance_loss_clip": 1.04782796,
      "balance_loss_mlp": 1.02080894,
      "epoch": 0.9087957674502495,
      "flos": 19129793857920.0,
      "grad_norm": 1.974768933582914,
      "language_loss": 0.73742378,
      "learning_rate": 8.652965430207104e-08,
      "loss": 0.7590915,
      "num_input_tokens_seen": 163399975,
      "step": 7558,
      "time_per_iteration": 2.713420867919922
    },
    {
      "auxiliary_loss_clip": 0.01103522,
      "auxiliary_loss_mlp": 0.01027544,
      "balance_loss_clip": 1.04344749,
      "balance_loss_mlp": 1.01971817,
      "epoch": 0.9089160103408886,
      "flos": 18109930999680.0,
      "grad_norm": 2.2627180403073077,
      "language_loss": 0.65698981,
      "learning_rate": 8.630314973622521e-08,
      "loss": 0.67830044,
      "num_input_tokens_seen": 163417520,
      "step": 7559,
      "time_per_iteration": 2.7395436763763428
    },
    {
      "auxiliary_loss_clip": 0.01125123,
      "auxiliary_loss_mlp": 0.01030698,
      "balance_loss_clip": 1.0449903,
      "balance_loss_mlp": 1.02272856,
      "epoch": 0.9090362532315277,
      "flos": 33364855336320.0,
      "grad_norm": 2.26169244306271,
      "language_loss": 0.71141028,
      "learning_rate": 8.607693547666995e-08,
      "loss": 0.73296845,
      "num_input_tokens_seen": 163440060,
      "step": 7560,
      "time_per_iteration": 2.890937089920044
    },
    {
      "auxiliary_loss_clip": 0.01060824,
      "auxiliary_loss_mlp": 0.01000944,
      "balance_loss_clip": 1.0072962,
      "balance_loss_mlp": 1.00004995,
      "epoch": 0.9091564961221668,
      "flos": 71480585082240.0,
      "grad_norm": 0.8840279251586736,
      "language_loss": 0.57874686,
      "learning_rate": 8.585101155772201e-08,
      "loss": 0.59936458,
      "num_input_tokens_seen": 163502180,
      "step": 7561,
      "time_per_iteration": 3.275007963180542
    },
    {
      "auxiliary_loss_clip": 0.01158292,
      "auxiliary_loss_mlp": 0.01025994,
      "balance_loss_clip": 1.04903901,
      "balance_loss_mlp": 1.01803064,
      "epoch": 0.9092767390128058,
      "flos": 24712574232960.0,
      "grad_norm": 1.7904094371903938,
      "language_loss": 0.68639785,
      "learning_rate": 8.562537801365377e-08,
      "loss": 0.70824075,
      "num_input_tokens_seen": 163521915,
      "step": 7562,
      "time_per_iteration": 3.626014471054077
    },
    {
      "auxiliary_loss_clip": 0.01106449,
      "auxiliary_loss_mlp": 0.00762514,
      "balance_loss_clip": 1.04290581,
      "balance_loss_mlp": 1.00045431,
      "epoch": 0.909396981903445,
      "flos": 23586487879680.0,
      "grad_norm": 1.7038094339820924,
      "language_loss": 0.7019577,
      "learning_rate": 8.540003487869362e-08,
      "loss": 0.72064739,
      "num_input_tokens_seen": 163543585,
      "step": 7563,
      "time_per_iteration": 2.8242440223693848
    },
    {
      "auxiliary_loss_clip": 0.01146822,
      "auxiliary_loss_mlp": 0.01021501,
      "balance_loss_clip": 1.04629219,
      "balance_loss_mlp": 1.01386571,
      "epoch": 0.909517224794084,
      "flos": 23404169422080.0,
      "grad_norm": 2.8939171495388587,
      "language_loss": 0.7962417,
      "learning_rate": 8.517498218702557e-08,
      "loss": 0.81792492,
      "num_input_tokens_seen": 163561515,
      "step": 7564,
      "time_per_iteration": 2.6638848781585693
    },
    {
      "auxiliary_loss_clip": 0.01165057,
      "auxiliary_loss_mlp": 0.01026565,
      "balance_loss_clip": 1.05099893,
      "balance_loss_mlp": 1.01887655,
      "epoch": 0.9096374676847231,
      "flos": 19208618254080.0,
      "grad_norm": 2.4394801598593134,
      "language_loss": 0.69798177,
      "learning_rate": 8.49502199727905e-08,
      "loss": 0.71989799,
      "num_input_tokens_seen": 163579540,
      "step": 7565,
      "time_per_iteration": 2.709554672241211
    },
    {
      "auxiliary_loss_clip": 0.01107334,
      "auxiliary_loss_mlp": 0.01029715,
      "balance_loss_clip": 1.03730023,
      "balance_loss_mlp": 1.02186227,
      "epoch": 0.9097577105753623,
      "flos": 33292495388160.0,
      "grad_norm": 2.4974487004306316,
      "language_loss": 0.66664535,
      "learning_rate": 8.472574827008428e-08,
      "loss": 0.68801588,
      "num_input_tokens_seen": 163600425,
      "step": 7566,
      "time_per_iteration": 2.9084393978118896
    },
    {
      "auxiliary_loss_clip": 0.01143073,
      "auxiliary_loss_mlp": 0.01028256,
      "balance_loss_clip": 1.04577255,
      "balance_loss_mlp": 1.0206207,
      "epoch": 0.9098779534660013,
      "flos": 21906443001600.0,
      "grad_norm": 3.772474220972928,
      "language_loss": 0.83784974,
      "learning_rate": 8.450156711295942e-08,
      "loss": 0.85956299,
      "num_input_tokens_seen": 163620595,
      "step": 7567,
      "time_per_iteration": 2.695103883743286
    },
    {
      "auxiliary_loss_clip": 0.01160445,
      "auxiliary_loss_mlp": 0.01023443,
      "balance_loss_clip": 1.05027878,
      "balance_loss_mlp": 1.01595104,
      "epoch": 0.9099981963566404,
      "flos": 25730354102400.0,
      "grad_norm": 2.1207368104540643,
      "language_loss": 0.87303138,
      "learning_rate": 8.427767653542383e-08,
      "loss": 0.89487028,
      "num_input_tokens_seen": 163635765,
      "step": 7568,
      "time_per_iteration": 2.692340850830078
    },
    {
      "auxiliary_loss_clip": 0.01133106,
      "auxiliary_loss_mlp": 0.01027781,
      "balance_loss_clip": 1.04637504,
      "balance_loss_mlp": 1.02042913,
      "epoch": 0.9101184392472795,
      "flos": 21069437304960.0,
      "grad_norm": 2.7304077120919588,
      "language_loss": 0.70226955,
      "learning_rate": 8.405407657144125e-08,
      "loss": 0.7238785,
      "num_input_tokens_seen": 163654925,
      "step": 7569,
      "time_per_iteration": 2.6747117042541504
    },
    {
      "auxiliary_loss_clip": 0.01140545,
      "auxiliary_loss_mlp": 0.01022124,
      "balance_loss_clip": 1.04747188,
      "balance_loss_mlp": 1.01497698,
      "epoch": 0.9102386821379186,
      "flos": 24752614919040.0,
      "grad_norm": 1.847705861291034,
      "language_loss": 0.72375953,
      "learning_rate": 8.383076725493232e-08,
      "loss": 0.74538618,
      "num_input_tokens_seen": 163672245,
      "step": 7570,
      "time_per_iteration": 2.744030475616455
    },
    {
      "auxiliary_loss_clip": 0.0114241,
      "auxiliary_loss_mlp": 0.01025203,
      "balance_loss_clip": 1.04715586,
      "balance_loss_mlp": 1.01762474,
      "epoch": 0.9103589250285576,
      "flos": 22562818179840.0,
      "grad_norm": 2.088696867665173,
      "language_loss": 0.68184686,
      "learning_rate": 8.360774861977216e-08,
      "loss": 0.70352304,
      "num_input_tokens_seen": 163691365,
      "step": 7571,
      "time_per_iteration": 2.714136838912964
    },
    {
      "auxiliary_loss_clip": 0.01125066,
      "auxiliary_loss_mlp": 0.01030522,
      "balance_loss_clip": 1.04726374,
      "balance_loss_mlp": 1.02346516,
      "epoch": 0.9104791679191968,
      "flos": 25373474524800.0,
      "grad_norm": 2.884811184264634,
      "language_loss": 0.74671423,
      "learning_rate": 8.338502069979281e-08,
      "loss": 0.76827008,
      "num_input_tokens_seen": 163711675,
      "step": 7572,
      "time_per_iteration": 2.7760379314422607
    },
    {
      "auxiliary_loss_clip": 0.01139789,
      "auxiliary_loss_mlp": 0.01028934,
      "balance_loss_clip": 1.04631591,
      "balance_loss_mlp": 1.02084005,
      "epoch": 0.9105994108098359,
      "flos": 14426681558400.0,
      "grad_norm": 4.828012642812359,
      "language_loss": 0.80288869,
      "learning_rate": 8.316258352878214e-08,
      "loss": 0.8245759,
      "num_input_tokens_seen": 163728095,
      "step": 7573,
      "time_per_iteration": 2.654036283493042
    },
    {
      "auxiliary_loss_clip": 0.01136862,
      "auxiliary_loss_mlp": 0.01024968,
      "balance_loss_clip": 1.04770279,
      "balance_loss_mlp": 1.01630712,
      "epoch": 0.9107196537004749,
      "flos": 26718292748160.0,
      "grad_norm": 1.8939326412471322,
      "language_loss": 0.71307182,
      "learning_rate": 8.294043714048338e-08,
      "loss": 0.73469013,
      "num_input_tokens_seen": 163747175,
      "step": 7574,
      "time_per_iteration": 4.063173055648804
    },
    {
      "auxiliary_loss_clip": 0.01022337,
      "auxiliary_loss_mlp": 0.01000204,
      "balance_loss_clip": 1.00663793,
      "balance_loss_mlp": 0.99934626,
      "epoch": 0.9108398965911141,
      "flos": 66532634703360.0,
      "grad_norm": 0.7518556526442952,
      "language_loss": 0.60422218,
      "learning_rate": 8.271858156859624e-08,
      "loss": 0.62444758,
      "num_input_tokens_seen": 163812545,
      "step": 7575,
      "time_per_iteration": 3.404127359390259
    },
    {
      "auxiliary_loss_clip": 0.01126997,
      "auxiliary_loss_mlp": 0.00761548,
      "balance_loss_clip": 1.04756665,
      "balance_loss_mlp": 1.00048137,
      "epoch": 0.9109601394817531,
      "flos": 25411073086080.0,
      "grad_norm": 1.7713953514242977,
      "language_loss": 0.73843408,
      "learning_rate": 8.249701684677557e-08,
      "loss": 0.75731957,
      "num_input_tokens_seen": 163833870,
      "step": 7576,
      "time_per_iteration": 2.7073869705200195
    },
    {
      "auxiliary_loss_clip": 0.01092167,
      "auxiliary_loss_mlp": 0.01027436,
      "balance_loss_clip": 1.03856683,
      "balance_loss_mlp": 1.01973248,
      "epoch": 0.9110803823723922,
      "flos": 22747794243840.0,
      "grad_norm": 1.8081905411825971,
      "language_loss": 0.81543589,
      "learning_rate": 8.227574300863294e-08,
      "loss": 0.83663189,
      "num_input_tokens_seen": 163854040,
      "step": 7577,
      "time_per_iteration": 2.8189361095428467
    },
    {
      "auxiliary_loss_clip": 0.01125798,
      "auxiliary_loss_mlp": 0.01032882,
      "balance_loss_clip": 1.04670215,
      "balance_loss_mlp": 1.02473736,
      "epoch": 0.9112006252630314,
      "flos": 48469924131840.0,
      "grad_norm": 1.5993920764865208,
      "language_loss": 0.69587147,
      "learning_rate": 8.205476008773548e-08,
      "loss": 0.71745831,
      "num_input_tokens_seen": 163878040,
      "step": 7578,
      "time_per_iteration": 3.898014545440674
    },
    {
      "auxiliary_loss_clip": 0.01127479,
      "auxiliary_loss_mlp": 0.01029665,
      "balance_loss_clip": 1.04460931,
      "balance_loss_mlp": 1.02219105,
      "epoch": 0.9113208681536704,
      "flos": 30009649829760.0,
      "grad_norm": 1.9314691283686696,
      "language_loss": 0.82878208,
      "learning_rate": 8.183406811760596e-08,
      "loss": 0.85035348,
      "num_input_tokens_seen": 163897770,
      "step": 7579,
      "time_per_iteration": 2.7902462482452393
    },
    {
      "auxiliary_loss_clip": 0.01146465,
      "auxiliary_loss_mlp": 0.01024434,
      "balance_loss_clip": 1.05004382,
      "balance_loss_mlp": 1.01701951,
      "epoch": 0.9114411110443095,
      "flos": 25594971742080.0,
      "grad_norm": 2.386667344196506,
      "language_loss": 0.74264747,
      "learning_rate": 8.161366713172313e-08,
      "loss": 0.76435649,
      "num_input_tokens_seen": 163920160,
      "step": 7580,
      "time_per_iteration": 2.7606992721557617
    },
    {
      "auxiliary_loss_clip": 0.01161213,
      "auxiliary_loss_mlp": 0.01032048,
      "balance_loss_clip": 1.04916453,
      "balance_loss_mlp": 1.02376842,
      "epoch": 0.9115613539349486,
      "flos": 18399729928320.0,
      "grad_norm": 4.399780463721851,
      "language_loss": 0.84107673,
      "learning_rate": 8.139355716352137e-08,
      "loss": 0.86300933,
      "num_input_tokens_seen": 163935000,
      "step": 7581,
      "time_per_iteration": 2.603933095932007
    },
    {
      "auxiliary_loss_clip": 0.01161309,
      "auxiliary_loss_mlp": 0.00761806,
      "balance_loss_clip": 1.05091178,
      "balance_loss_mlp": 1.00043643,
      "epoch": 0.9116815968255877,
      "flos": 21726171619200.0,
      "grad_norm": 1.7942736437797557,
      "language_loss": 0.69898242,
      "learning_rate": 8.117373824639196e-08,
      "loss": 0.71821356,
      "num_input_tokens_seen": 163955265,
      "step": 7582,
      "time_per_iteration": 2.713892698287964
    },
    {
      "auxiliary_loss_clip": 0.0102595,
      "auxiliary_loss_mlp": 0.00999271,
      "balance_loss_clip": 1.01634777,
      "balance_loss_mlp": 0.99837077,
      "epoch": 0.9118018397162267,
      "flos": 65363526835200.0,
      "grad_norm": 0.7203064248518837,
      "language_loss": 0.59211802,
      "learning_rate": 8.095421041368067e-08,
      "loss": 0.61237025,
      "num_input_tokens_seen": 164014680,
      "step": 7583,
      "time_per_iteration": 3.2290098667144775
    },
    {
      "auxiliary_loss_clip": 0.01146741,
      "auxiliary_loss_mlp": 0.0103302,
      "balance_loss_clip": 1.04994869,
      "balance_loss_mlp": 1.02527785,
      "epoch": 0.9119220826068659,
      "flos": 20922885815040.0,
      "grad_norm": 8.564137125844919,
      "language_loss": 0.70337856,
      "learning_rate": 8.073497369868999e-08,
      "loss": 0.72517616,
      "num_input_tokens_seen": 164033140,
      "step": 7584,
      "time_per_iteration": 2.6787149906158447
    },
    {
      "auxiliary_loss_clip": 0.01151503,
      "auxiliary_loss_mlp": 0.01034703,
      "balance_loss_clip": 1.04697299,
      "balance_loss_mlp": 1.02650118,
      "epoch": 0.912042325497505,
      "flos": 28366449327360.0,
      "grad_norm": 1.6672373494774626,
      "language_loss": 0.75432646,
      "learning_rate": 8.051602813467772e-08,
      "loss": 0.77618861,
      "num_input_tokens_seen": 164054995,
      "step": 7585,
      "time_per_iteration": 2.763896942138672
    },
    {
      "auxiliary_loss_clip": 0.01140645,
      "auxiliary_loss_mlp": 0.01025429,
      "balance_loss_clip": 1.05053627,
      "balance_loss_mlp": 1.01734948,
      "epoch": 0.912162568388144,
      "flos": 17566782468480.0,
      "grad_norm": 1.7339278125823514,
      "language_loss": 0.71177495,
      "learning_rate": 8.029737375485756e-08,
      "loss": 0.73343563,
      "num_input_tokens_seen": 164074225,
      "step": 7586,
      "time_per_iteration": 2.6601409912109375
    },
    {
      "auxiliary_loss_clip": 0.0113289,
      "auxiliary_loss_mlp": 0.00762192,
      "balance_loss_clip": 1.04311025,
      "balance_loss_mlp": 1.00054204,
      "epoch": 0.9122828112787832,
      "flos": 19827897661440.0,
      "grad_norm": 1.6699227508752192,
      "language_loss": 0.72276688,
      "learning_rate": 8.007901059239986e-08,
      "loss": 0.7417177,
      "num_input_tokens_seen": 164093505,
      "step": 7587,
      "time_per_iteration": 2.7020344734191895
    },
    {
      "auxiliary_loss_clip": 0.01128353,
      "auxiliary_loss_mlp": 0.01028554,
      "balance_loss_clip": 1.04631078,
      "balance_loss_mlp": 1.0208497,
      "epoch": 0.9124030541694222,
      "flos": 20813789232000.0,
      "grad_norm": 2.041159382188785,
      "language_loss": 0.80308521,
      "learning_rate": 7.986093868042964e-08,
      "loss": 0.82465434,
      "num_input_tokens_seen": 164113750,
      "step": 7588,
      "time_per_iteration": 2.6579113006591797
    },
    {
      "auxiliary_loss_clip": 0.01137309,
      "auxiliary_loss_mlp": 0.01023827,
      "balance_loss_clip": 1.046278,
      "balance_loss_mlp": 1.01636195,
      "epoch": 0.9125232970600613,
      "flos": 25192305302400.0,
      "grad_norm": 2.034665998112095,
      "language_loss": 0.67890495,
      "learning_rate": 7.964315805202826e-08,
      "loss": 0.70051634,
      "num_input_tokens_seen": 164134330,
      "step": 7589,
      "time_per_iteration": 3.738063097000122
    },
    {
      "auxiliary_loss_clip": 0.01171047,
      "auxiliary_loss_mlp": 0.0102687,
      "balance_loss_clip": 1.0486213,
      "balance_loss_mlp": 1.01944876,
      "epoch": 0.9126435399507005,
      "flos": 19719591177600.0,
      "grad_norm": 2.251025181660057,
      "language_loss": 0.73390698,
      "learning_rate": 7.942566874023304e-08,
      "loss": 0.7558862,
      "num_input_tokens_seen": 164153515,
      "step": 7590,
      "time_per_iteration": 2.5898234844207764
    },
    {
      "auxiliary_loss_clip": 0.01130158,
      "auxiliary_loss_mlp": 0.01031511,
      "balance_loss_clip": 1.04578269,
      "balance_loss_mlp": 1.02306485,
      "epoch": 0.9127637828413395,
      "flos": 19573614305280.0,
      "grad_norm": 2.2285934046036866,
      "language_loss": 0.70111644,
      "learning_rate": 7.920847077803649e-08,
      "loss": 0.72273314,
      "num_input_tokens_seen": 164171305,
      "step": 7591,
      "time_per_iteration": 2.6695680618286133
    },
    {
      "auxiliary_loss_clip": 0.01175102,
      "auxiliary_loss_mlp": 0.01029472,
      "balance_loss_clip": 1.0519042,
      "balance_loss_mlp": 1.02207828,
      "epoch": 0.9128840257319786,
      "flos": 20230635928320.0,
      "grad_norm": 1.7983931296038436,
      "language_loss": 0.82458186,
      "learning_rate": 7.899156419838826e-08,
      "loss": 0.84662765,
      "num_input_tokens_seen": 164190275,
      "step": 7592,
      "time_per_iteration": 2.6424663066864014
    },
    {
      "auxiliary_loss_clip": 0.01143157,
      "auxiliary_loss_mlp": 0.01029155,
      "balance_loss_clip": 1.05018044,
      "balance_loss_mlp": 1.02093506,
      "epoch": 0.9130042686226177,
      "flos": 24858658846080.0,
      "grad_norm": 2.8065085130430374,
      "language_loss": 0.65807378,
      "learning_rate": 7.87749490341918e-08,
      "loss": 0.67979693,
      "num_input_tokens_seen": 164210550,
      "step": 7593,
      "time_per_iteration": 2.7770297527313232
    },
    {
      "auxiliary_loss_clip": 0.01133705,
      "auxiliary_loss_mlp": 0.01025912,
      "balance_loss_clip": 1.0462687,
      "balance_loss_mlp": 1.01799083,
      "epoch": 0.9131245115132568,
      "flos": 23581747284480.0,
      "grad_norm": 1.8353273821178662,
      "language_loss": 0.83423233,
      "learning_rate": 7.855862531830836e-08,
      "loss": 0.8558284,
      "num_input_tokens_seen": 164226660,
      "step": 7594,
      "time_per_iteration": 2.798931121826172
    },
    {
      "auxiliary_loss_clip": 0.01127184,
      "auxiliary_loss_mlp": 0.01030211,
      "balance_loss_clip": 1.04406166,
      "balance_loss_mlp": 1.02191401,
      "epoch": 0.9132447544038959,
      "flos": 19931607204480.0,
      "grad_norm": 1.6446461896483666,
      "language_loss": 0.73047709,
      "learning_rate": 7.834259308355373e-08,
      "loss": 0.75205106,
      "num_input_tokens_seen": 164245425,
      "step": 7595,
      "time_per_iteration": 2.707657814025879
    },
    {
      "auxiliary_loss_clip": 0.01106972,
      "auxiliary_loss_mlp": 0.01032843,
      "balance_loss_clip": 1.0417707,
      "balance_loss_mlp": 1.02507079,
      "epoch": 0.9133649972945349,
      "flos": 21981747864960.0,
      "grad_norm": 4.9665717305404735,
      "language_loss": 0.74917841,
      "learning_rate": 7.812685236269989e-08,
      "loss": 0.7705766,
      "num_input_tokens_seen": 164264085,
      "step": 7596,
      "time_per_iteration": 2.7747058868408203
    },
    {
      "auxiliary_loss_clip": 0.01044576,
      "auxiliary_loss_mlp": 0.01000923,
      "balance_loss_clip": 1.00742912,
      "balance_loss_mlp": 1.00004065,
      "epoch": 0.9134852401851741,
      "flos": 71240523511680.0,
      "grad_norm": 0.8000022357860482,
      "language_loss": 0.58645815,
      "learning_rate": 7.791140318847445e-08,
      "loss": 0.60691315,
      "num_input_tokens_seen": 164322220,
      "step": 7597,
      "time_per_iteration": 3.2338428497314453
    },
    {
      "auxiliary_loss_clip": 0.01106092,
      "auxiliary_loss_mlp": 0.01031291,
      "balance_loss_clip": 1.04087651,
      "balance_loss_mlp": 1.02363217,
      "epoch": 0.9136054830758131,
      "flos": 23626923615360.0,
      "grad_norm": 1.7000413179887526,
      "language_loss": 0.80522734,
      "learning_rate": 7.769624559356081e-08,
      "loss": 0.82660121,
      "num_input_tokens_seen": 164345615,
      "step": 7598,
      "time_per_iteration": 2.8247275352478027
    },
    {
      "auxiliary_loss_clip": 0.01125023,
      "auxiliary_loss_mlp": 0.01030161,
      "balance_loss_clip": 1.04221439,
      "balance_loss_mlp": 1.02184319,
      "epoch": 0.9137257259664522,
      "flos": 23438858981760.0,
      "grad_norm": 2.926636532780072,
      "language_loss": 0.75139362,
      "learning_rate": 7.748137961059842e-08,
      "loss": 0.77294546,
      "num_input_tokens_seen": 164359595,
      "step": 7599,
      "time_per_iteration": 2.7015204429626465
    },
    {
      "auxiliary_loss_clip": 0.01146669,
      "auxiliary_loss_mlp": 0.01027832,
      "balance_loss_clip": 1.04689515,
      "balance_loss_mlp": 1.02008367,
      "epoch": 0.9138459688570914,
      "flos": 19127854523520.0,
      "grad_norm": 3.091603516809275,
      "language_loss": 0.65945482,
      "learning_rate": 7.726680527218211e-08,
      "loss": 0.68119979,
      "num_input_tokens_seen": 164376635,
      "step": 7600,
      "time_per_iteration": 3.581062078475952
    },
    {
      "auxiliary_loss_clip": 0.01155747,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.04967391,
      "balance_loss_mlp": 1.0194366,
      "epoch": 0.9139662117477304,
      "flos": 46281240714240.0,
      "grad_norm": 1.6804126449439551,
      "language_loss": 0.75657153,
      "learning_rate": 7.70525226108627e-08,
      "loss": 0.77839917,
      "num_input_tokens_seen": 164400305,
      "step": 7601,
      "time_per_iteration": 2.8925087451934814
    },
    {
      "auxiliary_loss_clip": 0.01121018,
      "auxiliary_loss_mlp": 0.01026189,
      "balance_loss_clip": 1.04881036,
      "balance_loss_mlp": 1.01866102,
      "epoch": 0.9140864546383695,
      "flos": 22273198819200.0,
      "grad_norm": 1.9288298131924007,
      "language_loss": 0.79849899,
      "learning_rate": 7.683853165914666e-08,
      "loss": 0.81997108,
      "num_input_tokens_seen": 164418075,
      "step": 7602,
      "time_per_iteration": 2.7005584239959717
    },
    {
      "auxiliary_loss_clip": 0.01156146,
      "auxiliary_loss_mlp": 0.01026373,
      "balance_loss_clip": 1.05082011,
      "balance_loss_mlp": 1.01883864,
      "epoch": 0.9142066975290086,
      "flos": 17530009920000.0,
      "grad_norm": 3.6782814408214777,
      "language_loss": 0.77049476,
      "learning_rate": 7.662483244949602e-08,
      "loss": 0.79232001,
      "num_input_tokens_seen": 164435335,
      "step": 7603,
      "time_per_iteration": 2.658341407775879
    },
    {
      "auxiliary_loss_clip": 0.01153654,
      "auxiliary_loss_mlp": 0.01031667,
      "balance_loss_clip": 1.04893947,
      "balance_loss_mlp": 1.02433515,
      "epoch": 0.9143269404196477,
      "flos": 17712148809600.0,
      "grad_norm": 2.543728395523601,
      "language_loss": 0.81351221,
      "learning_rate": 7.641142501432951e-08,
      "loss": 0.83536541,
      "num_input_tokens_seen": 164451530,
      "step": 7604,
      "time_per_iteration": 2.611583948135376
    },
    {
      "auxiliary_loss_clip": 0.01158963,
      "auxiliary_loss_mlp": 0.00761868,
      "balance_loss_clip": 1.05047035,
      "balance_loss_mlp": 1.00050986,
      "epoch": 0.9144471833102867,
      "flos": 33323414019840.0,
      "grad_norm": 2.942078434948713,
      "language_loss": 0.73714387,
      "learning_rate": 7.619830938602013e-08,
      "loss": 0.75635219,
      "num_input_tokens_seen": 164472755,
      "step": 7605,
      "time_per_iteration": 3.7065844535827637
    },
    {
      "auxiliary_loss_clip": 0.01141466,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.04989719,
      "balance_loss_mlp": 1.02159095,
      "epoch": 0.9145674262009259,
      "flos": 21068970428160.0,
      "grad_norm": 1.9294825108178146,
      "language_loss": 0.82573003,
      "learning_rate": 7.598548559689777e-08,
      "loss": 0.84744036,
      "num_input_tokens_seen": 164491155,
      "step": 7606,
      "time_per_iteration": 2.698089122772217
    },
    {
      "auxiliary_loss_clip": 0.010823,
      "auxiliary_loss_mlp": 0.01022961,
      "balance_loss_clip": 1.0380981,
      "balance_loss_mlp": 1.01533186,
      "epoch": 0.914687669091565,
      "flos": 16800269212800.0,
      "grad_norm": 3.3799136182887364,
      "language_loss": 0.81036496,
      "learning_rate": 7.577295367924751e-08,
      "loss": 0.83141762,
      "num_input_tokens_seen": 164507555,
      "step": 7607,
      "time_per_iteration": 2.7796661853790283
    },
    {
      "auxiliary_loss_clip": 0.01156039,
      "auxiliary_loss_mlp": 0.01026141,
      "balance_loss_clip": 1.04966211,
      "balance_loss_mlp": 1.01829743,
      "epoch": 0.914807911982204,
      "flos": 25773627012480.0,
      "grad_norm": 2.471264737138412,
      "language_loss": 0.82380754,
      "learning_rate": 7.556071366531002e-08,
      "loss": 0.84562933,
      "num_input_tokens_seen": 164528525,
      "step": 7608,
      "time_per_iteration": 2.7564878463745117
    },
    {
      "auxiliary_loss_clip": 0.01127458,
      "auxiliary_loss_mlp": 0.0076256,
      "balance_loss_clip": 1.0491091,
      "balance_loss_mlp": 1.00043464,
      "epoch": 0.9149281548728432,
      "flos": 19208043636480.0,
      "grad_norm": 1.8643274233083627,
      "language_loss": 0.78989685,
      "learning_rate": 7.53487655872822e-08,
      "loss": 0.808797,
      "num_input_tokens_seen": 164547695,
      "step": 7609,
      "time_per_iteration": 2.680999994277954
    },
    {
      "auxiliary_loss_clip": 0.01108525,
      "auxiliary_loss_mlp": 0.00761897,
      "balance_loss_clip": 1.03867018,
      "balance_loss_mlp": 1.00044572,
      "epoch": 0.9150483977634822,
      "flos": 26870554500480.0,
      "grad_norm": 1.79293634112879,
      "language_loss": 0.74250966,
      "learning_rate": 7.513710947731656e-08,
      "loss": 0.7612139,
      "num_input_tokens_seen": 164568905,
      "step": 7610,
      "time_per_iteration": 2.7767231464385986
    },
    {
      "auxiliary_loss_clip": 0.01075103,
      "auxiliary_loss_mlp": 0.01025174,
      "balance_loss_clip": 1.03956997,
      "balance_loss_mlp": 1.01775885,
      "epoch": 0.9151686406541213,
      "flos": 21908956953600.0,
      "grad_norm": 1.7655216749001232,
      "language_loss": 0.85164154,
      "learning_rate": 7.492574536752095e-08,
      "loss": 0.87264431,
      "num_input_tokens_seen": 164588895,
      "step": 7611,
      "time_per_iteration": 3.040304183959961
    },
    {
      "auxiliary_loss_clip": 0.01139552,
      "auxiliary_loss_mlp": 0.01024313,
      "balance_loss_clip": 1.04363537,
      "balance_loss_mlp": 1.01692247,
      "epoch": 0.9152888835447605,
      "flos": 27308556944640.0,
      "grad_norm": 2.089075200729387,
      "language_loss": 0.78187859,
      "learning_rate": 7.471467328995907e-08,
      "loss": 0.80351722,
      "num_input_tokens_seen": 164607705,
      "step": 7612,
      "time_per_iteration": 3.004133939743042
    },
    {
      "auxiliary_loss_clip": 0.01148706,
      "auxiliary_loss_mlp": 0.0103176,
      "balance_loss_clip": 1.0491147,
      "balance_loss_mlp": 1.02379417,
      "epoch": 0.9154091264353995,
      "flos": 13370728510080.0,
      "grad_norm": 3.385382977289574,
      "language_loss": 0.60750473,
      "learning_rate": 7.450389327665018e-08,
      "loss": 0.62930942,
      "num_input_tokens_seen": 164625540,
      "step": 7613,
      "time_per_iteration": 2.712035655975342
    },
    {
      "auxiliary_loss_clip": 0.01119293,
      "auxiliary_loss_mlp": 0.01027181,
      "balance_loss_clip": 1.04599464,
      "balance_loss_mlp": 1.01890147,
      "epoch": 0.9155293693260386,
      "flos": 20193037367040.0,
      "grad_norm": 3.7556556394323968,
      "language_loss": 0.67397803,
      "learning_rate": 7.429340535957029e-08,
      "loss": 0.6954428,
      "num_input_tokens_seen": 164640735,
      "step": 7614,
      "time_per_iteration": 3.6477208137512207
    },
    {
      "auxiliary_loss_clip": 0.01141159,
      "auxiliary_loss_mlp": 0.01029768,
      "balance_loss_clip": 1.0488739,
      "balance_loss_mlp": 1.02282071,
      "epoch": 0.9156496122166777,
      "flos": 19354990176000.0,
      "grad_norm": 11.702315805245664,
      "language_loss": 0.70984191,
      "learning_rate": 7.40832095706494e-08,
      "loss": 0.73155117,
      "num_input_tokens_seen": 164657430,
      "step": 7615,
      "time_per_iteration": 2.708977699279785
    },
    {
      "auxiliary_loss_clip": 0.01114791,
      "auxiliary_loss_mlp": 0.01030667,
      "balance_loss_clip": 1.04723144,
      "balance_loss_mlp": 1.0223943,
      "epoch": 0.9157698551073168,
      "flos": 21107287261440.0,
      "grad_norm": 1.8167635735627454,
      "language_loss": 0.80531347,
      "learning_rate": 7.387330594177443e-08,
      "loss": 0.82676804,
      "num_input_tokens_seen": 164679505,
      "step": 7616,
      "time_per_iteration": 2.7370593547821045
    },
    {
      "auxiliary_loss_clip": 0.0110495,
      "auxiliary_loss_mlp": 0.0102358,
      "balance_loss_clip": 1.04123783,
      "balance_loss_mlp": 1.01612973,
      "epoch": 0.9158900979979558,
      "flos": 25193167228800.0,
      "grad_norm": 1.7472493249095091,
      "language_loss": 0.79511881,
      "learning_rate": 7.366369450478749e-08,
      "loss": 0.8164041,
      "num_input_tokens_seen": 164700615,
      "step": 7617,
      "time_per_iteration": 3.377206563949585
    },
    {
      "auxiliary_loss_clip": 0.01124128,
      "auxiliary_loss_mlp": 0.00761707,
      "balance_loss_clip": 1.04459965,
      "balance_loss_mlp": 1.00047112,
      "epoch": 0.916010340888595,
      "flos": 30146648302080.0,
      "grad_norm": 2.0512949746608684,
      "language_loss": 0.6664356,
      "learning_rate": 7.345437529148646e-08,
      "loss": 0.68529391,
      "num_input_tokens_seen": 164719625,
      "step": 7618,
      "time_per_iteration": 3.148916244506836
    },
    {
      "auxiliary_loss_clip": 0.01154075,
      "auxiliary_loss_mlp": 0.01030814,
      "balance_loss_clip": 1.05007672,
      "balance_loss_mlp": 1.02354264,
      "epoch": 0.9161305837792341,
      "flos": 17091827907840.0,
      "grad_norm": 1.961665136928021,
      "language_loss": 0.72432703,
      "learning_rate": 7.324534833362483e-08,
      "loss": 0.74617594,
      "num_input_tokens_seen": 164737200,
      "step": 7619,
      "time_per_iteration": 2.6561641693115234
    },
    {
      "auxiliary_loss_clip": 0.01132324,
      "auxiliary_loss_mlp": 0.0076144,
      "balance_loss_clip": 1.04750514,
      "balance_loss_mlp": 1.00051272,
      "epoch": 0.9162508266698731,
      "flos": 22893699288960.0,
      "grad_norm": 1.7267000340407068,
      "language_loss": 0.68417406,
      "learning_rate": 7.303661366291192e-08,
      "loss": 0.70311165,
      "num_input_tokens_seen": 164757870,
      "step": 7620,
      "time_per_iteration": 2.8032896518707275
    },
    {
      "auxiliary_loss_clip": 0.01132446,
      "auxiliary_loss_mlp": 0.0103317,
      "balance_loss_clip": 1.04945731,
      "balance_loss_mlp": 1.02536798,
      "epoch": 0.9163710695605123,
      "flos": 19974808287360.0,
      "grad_norm": 1.8347532089896148,
      "language_loss": 0.81762624,
      "learning_rate": 7.28281713110126e-08,
      "loss": 0.83928239,
      "num_input_tokens_seen": 164775945,
      "step": 7621,
      "time_per_iteration": 2.7372915744781494
    },
    {
      "auxiliary_loss_clip": 0.01132287,
      "auxiliary_loss_mlp": 0.0102654,
      "balance_loss_clip": 1.04416943,
      "balance_loss_mlp": 1.01903582,
      "epoch": 0.9164913124511513,
      "flos": 22783812606720.0,
      "grad_norm": 2.2289112981128967,
      "language_loss": 0.77405667,
      "learning_rate": 7.262002130954759e-08,
      "loss": 0.79564494,
      "num_input_tokens_seen": 164794400,
      "step": 7622,
      "time_per_iteration": 2.7559101581573486
    },
    {
      "auxiliary_loss_clip": 0.0114487,
      "auxiliary_loss_mlp": 0.01035547,
      "balance_loss_clip": 1.05097103,
      "balance_loss_mlp": 1.02803099,
      "epoch": 0.9166115553417904,
      "flos": 24900854348160.0,
      "grad_norm": 1.739150704512114,
      "language_loss": 0.78977197,
      "learning_rate": 7.241216369009296e-08,
      "loss": 0.81157613,
      "num_input_tokens_seen": 164814585,
      "step": 7623,
      "time_per_iteration": 2.7673721313476562
    },
    {
      "auxiliary_loss_clip": 0.01132389,
      "auxiliary_loss_mlp": 0.01030706,
      "balance_loss_clip": 1.04403734,
      "balance_loss_mlp": 1.02263558,
      "epoch": 0.9167317982324296,
      "flos": 25702919089920.0,
      "grad_norm": 3.0759560376008857,
      "language_loss": 0.66773617,
      "learning_rate": 7.220459848418037e-08,
      "loss": 0.68936712,
      "num_input_tokens_seen": 164834660,
      "step": 7624,
      "time_per_iteration": 2.7724545001983643
    },
    {
      "auxiliary_loss_clip": 0.01113779,
      "auxiliary_loss_mlp": 0.0102832,
      "balance_loss_clip": 1.04411888,
      "balance_loss_mlp": 1.02052355,
      "epoch": 0.9168520411230686,
      "flos": 15632813370240.0,
      "grad_norm": 1.6804562170546902,
      "language_loss": 0.79968035,
      "learning_rate": 7.199732572329708e-08,
      "loss": 0.82110131,
      "num_input_tokens_seen": 164852560,
      "step": 7625,
      "time_per_iteration": 2.912705421447754
    },
    {
      "auxiliary_loss_clip": 0.01164846,
      "auxiliary_loss_mlp": 0.01028298,
      "balance_loss_clip": 1.05145431,
      "balance_loss_mlp": 1.02048945,
      "epoch": 0.9169722840137077,
      "flos": 30258151096320.0,
      "grad_norm": 3.505900179928099,
      "language_loss": 0.76359415,
      "learning_rate": 7.179034543888684e-08,
      "loss": 0.78552556,
      "num_input_tokens_seen": 164872065,
      "step": 7626,
      "time_per_iteration": 4.83204984664917
    },
    {
      "auxiliary_loss_clip": 0.01097743,
      "auxiliary_loss_mlp": 0.01029406,
      "balance_loss_clip": 1.04562819,
      "balance_loss_mlp": 1.02170491,
      "epoch": 0.9170925269043467,
      "flos": 22491643380480.0,
      "grad_norm": 2.9816143793893626,
      "language_loss": 0.78076398,
      "learning_rate": 7.158365766234808e-08,
      "loss": 0.80203545,
      "num_input_tokens_seen": 164890915,
      "step": 7627,
      "time_per_iteration": 2.818603038787842
    },
    {
      "auxiliary_loss_clip": 0.01174401,
      "auxiliary_loss_mlp": 0.00762031,
      "balance_loss_clip": 1.04920578,
      "balance_loss_mlp": 1.00040889,
      "epoch": 0.9172127697949859,
      "flos": 22893914770560.0,
      "grad_norm": 2.386124095119254,
      "language_loss": 0.72920221,
      "learning_rate": 7.137726242503527e-08,
      "loss": 0.74856651,
      "num_input_tokens_seen": 164909835,
      "step": 7628,
      "time_per_iteration": 2.6859402656555176
    },
    {
      "auxiliary_loss_clip": 0.01107502,
      "auxiliary_loss_mlp": 0.01026597,
      "balance_loss_clip": 1.03967476,
      "balance_loss_mlp": 1.01840162,
      "epoch": 0.917333012685625,
      "flos": 17451867882240.0,
      "grad_norm": 2.51228979580675,
      "language_loss": 0.78320432,
      "learning_rate": 7.11711597582585e-08,
      "loss": 0.80454534,
      "num_input_tokens_seen": 164927195,
      "step": 7629,
      "time_per_iteration": 2.753694534301758
    },
    {
      "auxiliary_loss_clip": 0.01115477,
      "auxiliary_loss_mlp": 0.01023087,
      "balance_loss_clip": 1.04327488,
      "balance_loss_mlp": 1.01544595,
      "epoch": 0.917453255576264,
      "flos": 14318949692160.0,
      "grad_norm": 1.789687779870364,
      "language_loss": 0.80247921,
      "learning_rate": 7.096534969328271e-08,
      "loss": 0.82386482,
      "num_input_tokens_seen": 164944640,
      "step": 7630,
      "time_per_iteration": 3.681758403778076
    },
    {
      "auxiliary_loss_clip": 0.01108059,
      "auxiliary_loss_mlp": 0.01031894,
      "balance_loss_clip": 1.04015708,
      "balance_loss_mlp": 1.02439284,
      "epoch": 0.9175734984669032,
      "flos": 20741177888640.0,
      "grad_norm": 2.279546584182616,
      "language_loss": 0.84512341,
      "learning_rate": 7.075983226132987e-08,
      "loss": 0.86652291,
      "num_input_tokens_seen": 164963570,
      "step": 7631,
      "time_per_iteration": 2.7928667068481445
    },
    {
      "auxiliary_loss_clip": 0.01132982,
      "auxiliary_loss_mlp": 0.01026244,
      "balance_loss_clip": 1.04826808,
      "balance_loss_mlp": 1.01730919,
      "epoch": 0.9176937413575422,
      "flos": 14830497233280.0,
      "grad_norm": 2.878719903954061,
      "language_loss": 0.79677242,
      "learning_rate": 7.055460749357656e-08,
      "loss": 0.81836474,
      "num_input_tokens_seen": 164979850,
      "step": 7632,
      "time_per_iteration": 2.6546308994293213
    },
    {
      "auxiliary_loss_clip": 0.01130785,
      "auxiliary_loss_mlp": 0.01025955,
      "balance_loss_clip": 1.04562545,
      "balance_loss_mlp": 1.01850104,
      "epoch": 0.9178139842481813,
      "flos": 18474603828480.0,
      "grad_norm": 1.9679615472455823,
      "language_loss": 0.70435297,
      "learning_rate": 7.034967542115521e-08,
      "loss": 0.72592038,
      "num_input_tokens_seen": 164998115,
      "step": 7633,
      "time_per_iteration": 2.7770681381225586
    },
    {
      "auxiliary_loss_clip": 0.01138627,
      "auxiliary_loss_mlp": 0.01027039,
      "balance_loss_clip": 1.04876602,
      "balance_loss_mlp": 1.01986802,
      "epoch": 0.9179342271388204,
      "flos": 20047455544320.0,
      "grad_norm": 2.2679029850845027,
      "language_loss": 0.7566908,
      "learning_rate": 7.014503607515388e-08,
      "loss": 0.77834743,
      "num_input_tokens_seen": 165017420,
      "step": 7634,
      "time_per_iteration": 2.7365808486938477
    },
    {
      "auxiliary_loss_clip": 0.01149653,
      "auxiliary_loss_mlp": 0.01032539,
      "balance_loss_clip": 1.0497067,
      "balance_loss_mlp": 1.02489173,
      "epoch": 0.9180544700294595,
      "flos": 24676232647680.0,
      "grad_norm": 2.064203732256373,
      "language_loss": 0.68481076,
      "learning_rate": 6.994068948661592e-08,
      "loss": 0.70663267,
      "num_input_tokens_seen": 165035575,
      "step": 7635,
      "time_per_iteration": 2.836554765701294
    },
    {
      "auxiliary_loss_clip": 0.01166509,
      "auxiliary_loss_mlp": 0.01030571,
      "balance_loss_clip": 1.05089808,
      "balance_loss_mlp": 1.02156436,
      "epoch": 0.9181747129200986,
      "flos": 16727478301440.0,
      "grad_norm": 3.0288092085872194,
      "language_loss": 0.76794261,
      "learning_rate": 6.973663568654142e-08,
      "loss": 0.78991342,
      "num_input_tokens_seen": 165053280,
      "step": 7636,
      "time_per_iteration": 2.648531436920166
    },
    {
      "auxiliary_loss_clip": 0.01123261,
      "auxiliary_loss_mlp": 0.01029547,
      "balance_loss_clip": 1.04690504,
      "balance_loss_mlp": 1.02216768,
      "epoch": 0.9182949558107377,
      "flos": 24271626873600.0,
      "grad_norm": 2.5866003315655064,
      "language_loss": 0.65610325,
      "learning_rate": 6.953287470588386e-08,
      "loss": 0.67763138,
      "num_input_tokens_seen": 165071235,
      "step": 7637,
      "time_per_iteration": 2.7516791820526123
    },
    {
      "auxiliary_loss_clip": 0.01107875,
      "auxiliary_loss_mlp": 0.01027915,
      "balance_loss_clip": 1.03912604,
      "balance_loss_mlp": 1.01979041,
      "epoch": 0.9184151987013768,
      "flos": 22082117443200.0,
      "grad_norm": 2.9895272733112606,
      "language_loss": 0.86005479,
      "learning_rate": 6.932940657555452e-08,
      "loss": 0.88141263,
      "num_input_tokens_seen": 165087365,
      "step": 7638,
      "time_per_iteration": 2.728921890258789
    },
    {
      "auxiliary_loss_clip": 0.01169045,
      "auxiliary_loss_mlp": 0.01025376,
      "balance_loss_clip": 1.05118608,
      "balance_loss_mlp": 1.01830745,
      "epoch": 0.9185354415920158,
      "flos": 32166732257280.0,
      "grad_norm": 1.4061272895383492,
      "language_loss": 0.76384819,
      "learning_rate": 6.912623132641938e-08,
      "loss": 0.78579241,
      "num_input_tokens_seen": 165112455,
      "step": 7639,
      "time_per_iteration": 2.8076727390289307
    },
    {
      "auxiliary_loss_clip": 0.01166328,
      "auxiliary_loss_mlp": 0.01031784,
      "balance_loss_clip": 1.05080509,
      "balance_loss_mlp": 1.02384138,
      "epoch": 0.918655684482655,
      "flos": 20997831542400.0,
      "grad_norm": 2.1791094760381853,
      "language_loss": 0.76843148,
      "learning_rate": 6.892334898929952e-08,
      "loss": 0.7904126,
      "num_input_tokens_seen": 165132700,
      "step": 7640,
      "time_per_iteration": 3.5447099208831787
    },
    {
      "auxiliary_loss_clip": 0.01109726,
      "auxiliary_loss_mlp": 0.00761729,
      "balance_loss_clip": 1.04271698,
      "balance_loss_mlp": 1.00056815,
      "epoch": 0.918775927373294,
      "flos": 15560704817280.0,
      "grad_norm": 5.393015213641831,
      "language_loss": 0.85061866,
      "learning_rate": 6.872075959497236e-08,
      "loss": 0.86933315,
      "num_input_tokens_seen": 165151475,
      "step": 7641,
      "time_per_iteration": 2.7734012603759766
    },
    {
      "auxiliary_loss_clip": 0.01105192,
      "auxiliary_loss_mlp": 0.01027529,
      "balance_loss_clip": 1.04235744,
      "balance_loss_mlp": 1.01981008,
      "epoch": 0.9188961702639331,
      "flos": 29934057657600.0,
      "grad_norm": 2.029078194255461,
      "language_loss": 0.83174896,
      "learning_rate": 6.85184631741702e-08,
      "loss": 0.8530761,
      "num_input_tokens_seen": 165172040,
      "step": 7642,
      "time_per_iteration": 2.7941927909851074
    },
    {
      "auxiliary_loss_clip": 0.01137888,
      "auxiliary_loss_mlp": 0.01024967,
      "balance_loss_clip": 1.0450213,
      "balance_loss_mlp": 1.01706386,
      "epoch": 0.9190164131545723,
      "flos": 20701244943360.0,
      "grad_norm": 1.9101643316510302,
      "language_loss": 0.77649367,
      "learning_rate": 6.831645975758161e-08,
      "loss": 0.79812223,
      "num_input_tokens_seen": 165189980,
      "step": 7643,
      "time_per_iteration": 2.670668363571167
    },
    {
      "auxiliary_loss_clip": 0.01116129,
      "auxiliary_loss_mlp": 0.00762113,
      "balance_loss_clip": 1.04403043,
      "balance_loss_mlp": 1.00039947,
      "epoch": 0.9191366560452113,
      "flos": 25629912696960.0,
      "grad_norm": 2.127721295961637,
      "language_loss": 0.67648697,
      "learning_rate": 6.811474937585026e-08,
      "loss": 0.69526941,
      "num_input_tokens_seen": 165209770,
      "step": 7644,
      "time_per_iteration": 2.7680602073669434
    },
    {
      "auxiliary_loss_clip": 0.01121758,
      "auxiliary_loss_mlp": 0.01025704,
      "balance_loss_clip": 1.04823315,
      "balance_loss_mlp": 1.01780057,
      "epoch": 0.9192568989358504,
      "flos": 21434325615360.0,
      "grad_norm": 1.6732017054890598,
      "language_loss": 0.79322672,
      "learning_rate": 6.79133320595755e-08,
      "loss": 0.81470132,
      "num_input_tokens_seen": 165229690,
      "step": 7645,
      "time_per_iteration": 2.6872615814208984
    },
    {
      "auxiliary_loss_clip": 0.01174083,
      "auxiliary_loss_mlp": 0.01031115,
      "balance_loss_clip": 1.05176973,
      "balance_loss_mlp": 1.02352786,
      "epoch": 0.9193771418264896,
      "flos": 23185078416000.0,
      "grad_norm": 3.209463026294622,
      "language_loss": 0.75689435,
      "learning_rate": 6.771220783931198e-08,
      "loss": 0.77894634,
      "num_input_tokens_seen": 165249850,
      "step": 7646,
      "time_per_iteration": 2.7198266983032227
    },
    {
      "auxiliary_loss_clip": 0.0105314,
      "auxiliary_loss_mlp": 0.0099984,
      "balance_loss_clip": 1.00785613,
      "balance_loss_mlp": 0.99895829,
      "epoch": 0.9194973847171286,
      "flos": 70582963184640.0,
      "grad_norm": 0.829487134167322,
      "language_loss": 0.64524472,
      "learning_rate": 6.751137674556994e-08,
      "loss": 0.66577452,
      "num_input_tokens_seen": 165310235,
      "step": 7647,
      "time_per_iteration": 3.304471015930176
    },
    {
      "auxiliary_loss_clip": 0.01146504,
      "auxiliary_loss_mlp": 0.01026233,
      "balance_loss_clip": 1.04568458,
      "balance_loss_mlp": 1.01823997,
      "epoch": 0.9196176276077677,
      "flos": 14720682378240.0,
      "grad_norm": 2.344139968824643,
      "language_loss": 0.77289844,
      "learning_rate": 6.731083880881572e-08,
      "loss": 0.79462588,
      "num_input_tokens_seen": 165326455,
      "step": 7648,
      "time_per_iteration": 2.64803147315979
    },
    {
      "auxiliary_loss_clip": 0.01156238,
      "auxiliary_loss_mlp": 0.01030896,
      "balance_loss_clip": 1.04932737,
      "balance_loss_mlp": 1.02346599,
      "epoch": 0.9197378704984068,
      "flos": 23294893271040.0,
      "grad_norm": 2.345105834944742,
      "language_loss": 0.81154799,
      "learning_rate": 6.711059405947072e-08,
      "loss": 0.83341932,
      "num_input_tokens_seen": 165344645,
      "step": 7649,
      "time_per_iteration": 2.766244411468506
    },
    {
      "auxiliary_loss_clip": 0.01132878,
      "auxiliary_loss_mlp": 0.01030615,
      "balance_loss_clip": 1.04789042,
      "balance_loss_mlp": 1.02322125,
      "epoch": 0.9198581133890459,
      "flos": 20302564913280.0,
      "grad_norm": 2.0947073052273866,
      "language_loss": 0.77251995,
      "learning_rate": 6.691064252791156e-08,
      "loss": 0.79415488,
      "num_input_tokens_seen": 165364120,
      "step": 7650,
      "time_per_iteration": 2.7045905590057373
    },
    {
      "auxiliary_loss_clip": 0.01141964,
      "auxiliary_loss_mlp": 0.01029645,
      "balance_loss_clip": 1.04532695,
      "balance_loss_mlp": 1.02181923,
      "epoch": 0.9199783562796849,
      "flos": 17675663569920.0,
      "grad_norm": 1.6993824971589582,
      "language_loss": 0.78529167,
      "learning_rate": 6.67109842444713e-08,
      "loss": 0.80700779,
      "num_input_tokens_seen": 165383050,
      "step": 7651,
      "time_per_iteration": 2.6964685916900635
    },
    {
      "auxiliary_loss_clip": 0.01159734,
      "auxiliary_loss_mlp": 0.00761947,
      "balance_loss_clip": 1.05031192,
      "balance_loss_mlp": 1.00044644,
      "epoch": 0.9200985991703241,
      "flos": 17676022705920.0,
      "grad_norm": 2.024743360610556,
      "language_loss": 0.76711953,
      "learning_rate": 6.651161923943704e-08,
      "loss": 0.78633624,
      "num_input_tokens_seen": 165400955,
      "step": 7652,
      "time_per_iteration": 4.537797212600708
    },
    {
      "auxiliary_loss_clip": 0.01164268,
      "auxiliary_loss_mlp": 0.01031477,
      "balance_loss_clip": 1.04903471,
      "balance_loss_mlp": 1.023669,
      "epoch": 0.9202188420609632,
      "flos": 20996574566400.0,
      "grad_norm": 2.021521413637887,
      "language_loss": 0.76916444,
      "learning_rate": 6.631254754305326e-08,
      "loss": 0.79112196,
      "num_input_tokens_seen": 165420415,
      "step": 7653,
      "time_per_iteration": 2.6540744304656982
    },
    {
      "auxiliary_loss_clip": 0.01145315,
      "auxiliary_loss_mlp": 0.01026647,
      "balance_loss_clip": 1.04706299,
      "balance_loss_mlp": 1.01861274,
      "epoch": 0.9203390849516022,
      "flos": 13918222586880.0,
      "grad_norm": 2.0209202297131963,
      "language_loss": 0.78799951,
      "learning_rate": 6.611376918551848e-08,
      "loss": 0.80971915,
      "num_input_tokens_seen": 165439200,
      "step": 7654,
      "time_per_iteration": 2.687790632247925
    },
    {
      "auxiliary_loss_clip": 0.01175439,
      "auxiliary_loss_mlp": 0.01024643,
      "balance_loss_clip": 1.0509547,
      "balance_loss_mlp": 1.01674223,
      "epoch": 0.9204593278422414,
      "flos": 21175912195200.0,
      "grad_norm": 2.25466054751698,
      "language_loss": 0.79301071,
      "learning_rate": 6.591528419698744e-08,
      "loss": 0.8150115,
      "num_input_tokens_seen": 165458985,
      "step": 7655,
      "time_per_iteration": 3.555274486541748
    },
    {
      "auxiliary_loss_clip": 0.01118358,
      "auxiliary_loss_mlp": 0.0102601,
      "balance_loss_clip": 1.04122162,
      "balance_loss_mlp": 1.01839244,
      "epoch": 0.9205795707328804,
      "flos": 14501375890560.0,
      "grad_norm": 4.517610805752367,
      "language_loss": 0.83745933,
      "learning_rate": 6.571709260756986e-08,
      "loss": 0.85890299,
      "num_input_tokens_seen": 165475630,
      "step": 7656,
      "time_per_iteration": 2.7002267837524414
    },
    {
      "auxiliary_loss_clip": 0.01136196,
      "auxiliary_loss_mlp": 0.01032101,
      "balance_loss_clip": 1.05075347,
      "balance_loss_mlp": 1.02444732,
      "epoch": 0.9206998136235195,
      "flos": 22417559579520.0,
      "grad_norm": 7.529939869143541,
      "language_loss": 0.76633477,
      "learning_rate": 6.551919444733122e-08,
      "loss": 0.78801775,
      "num_input_tokens_seen": 165493445,
      "step": 7657,
      "time_per_iteration": 2.786367654800415
    },
    {
      "auxiliary_loss_clip": 0.01145857,
      "auxiliary_loss_mlp": 0.01025344,
      "balance_loss_clip": 1.04851043,
      "balance_loss_mlp": 1.01823866,
      "epoch": 0.9208200565141585,
      "flos": 53358407544960.0,
      "grad_norm": 3.2199301015148167,
      "language_loss": 0.65652502,
      "learning_rate": 6.53215897462931e-08,
      "loss": 0.67823702,
      "num_input_tokens_seen": 165517200,
      "step": 7658,
      "time_per_iteration": 2.9384515285491943
    },
    {
      "auxiliary_loss_clip": 0.01135364,
      "auxiliary_loss_mlp": 0.01030343,
      "balance_loss_clip": 1.04945493,
      "balance_loss_mlp": 1.0222491,
      "epoch": 0.9209402994047977,
      "flos": 30589139946240.0,
      "grad_norm": 2.1801537155669104,
      "language_loss": 0.75414383,
      "learning_rate": 6.512427853443103e-08,
      "loss": 0.77580082,
      "num_input_tokens_seen": 165539280,
      "step": 7659,
      "time_per_iteration": 2.780489921569824
    },
    {
      "auxiliary_loss_clip": 0.01157309,
      "auxiliary_loss_mlp": 0.01028619,
      "balance_loss_clip": 1.04639268,
      "balance_loss_mlp": 1.02095389,
      "epoch": 0.9210605422954368,
      "flos": 29132711187840.0,
      "grad_norm": 1.5851057422701718,
      "language_loss": 0.75894094,
      "learning_rate": 6.492726084167799e-08,
      "loss": 0.78080022,
      "num_input_tokens_seen": 165561395,
      "step": 7660,
      "time_per_iteration": 2.6791868209838867
    },
    {
      "auxiliary_loss_clip": 0.01038968,
      "auxiliary_loss_mlp": 0.0100067,
      "balance_loss_clip": 1.00753975,
      "balance_loss_mlp": 0.99977046,
      "epoch": 0.9211807851860758,
      "flos": 54853838472960.0,
      "grad_norm": 0.7788096557254193,
      "language_loss": 0.57469964,
      "learning_rate": 6.473053669792072e-08,
      "loss": 0.59509599,
      "num_input_tokens_seen": 165616085,
      "step": 7661,
      "time_per_iteration": 3.1394858360290527
    },
    {
      "auxiliary_loss_clip": 0.01156862,
      "auxiliary_loss_mlp": 0.01028231,
      "balance_loss_clip": 1.05172527,
      "balance_loss_mlp": 1.02022648,
      "epoch": 0.921301028076715,
      "flos": 19201974238080.0,
      "grad_norm": 2.8240754486625272,
      "language_loss": 0.73212951,
      "learning_rate": 6.453410613300248e-08,
      "loss": 0.7539804,
      "num_input_tokens_seen": 165634015,
      "step": 7662,
      "time_per_iteration": 2.614356756210327
    },
    {
      "auxiliary_loss_clip": 0.01156542,
      "auxiliary_loss_mlp": 0.01032458,
      "balance_loss_clip": 1.05043864,
      "balance_loss_mlp": 1.02509069,
      "epoch": 0.921421270967354,
      "flos": 27526893765120.0,
      "grad_norm": 2.600069455973691,
      "language_loss": 0.58697385,
      "learning_rate": 6.43379691767214e-08,
      "loss": 0.60886383,
      "num_input_tokens_seen": 165653220,
      "step": 7663,
      "time_per_iteration": 2.755405902862549
    },
    {
      "auxiliary_loss_clip": 0.01024031,
      "auxiliary_loss_mlp": 0.01001038,
      "balance_loss_clip": 1.01034069,
      "balance_loss_mlp": 1.00010192,
      "epoch": 0.9215415138579931,
      "flos": 70209311955840.0,
      "grad_norm": 0.7206346324873398,
      "language_loss": 0.5503515,
      "learning_rate": 6.414212585883105e-08,
      "loss": 0.57060218,
      "num_input_tokens_seen": 165715850,
      "step": 7664,
      "time_per_iteration": 3.374865770339966
    },
    {
      "auxiliary_loss_clip": 0.01160491,
      "auxiliary_loss_mlp": 0.01025996,
      "balance_loss_clip": 1.04999709,
      "balance_loss_mlp": 1.01760948,
      "epoch": 0.9216617567486323,
      "flos": 35553107790720.0,
      "grad_norm": 2.0120386482472163,
      "language_loss": 0.69846004,
      "learning_rate": 6.394657620904143e-08,
      "loss": 0.72032487,
      "num_input_tokens_seen": 165738960,
      "step": 7665,
      "time_per_iteration": 2.7584314346313477
    },
    {
      "auxiliary_loss_clip": 0.01146924,
      "auxiliary_loss_mlp": 0.01034791,
      "balance_loss_clip": 1.04871905,
      "balance_loss_mlp": 1.02653575,
      "epoch": 0.9217819996392713,
      "flos": 29533330552320.0,
      "grad_norm": 1.7745688255883274,
      "language_loss": 0.71951216,
      "learning_rate": 6.375132025701657e-08,
      "loss": 0.74132931,
      "num_input_tokens_seen": 165761260,
      "step": 7666,
      "time_per_iteration": 3.763221025466919
    },
    {
      "auxiliary_loss_clip": 0.01111283,
      "auxiliary_loss_mlp": 0.01036312,
      "balance_loss_clip": 1.04852271,
      "balance_loss_mlp": 1.0275383,
      "epoch": 0.9219022425299104,
      "flos": 14574669592320.0,
      "grad_norm": 2.8123609266831724,
      "language_loss": 0.6907813,
      "learning_rate": 6.355635803237724e-08,
      "loss": 0.71225721,
      "num_input_tokens_seen": 165776960,
      "step": 7667,
      "time_per_iteration": 2.761781930923462
    },
    {
      "auxiliary_loss_clip": 0.01117015,
      "auxiliary_loss_mlp": 0.01026122,
      "balance_loss_clip": 1.04288554,
      "balance_loss_mlp": 1.01781893,
      "epoch": 0.9220224854205495,
      "flos": 18077503996800.0,
      "grad_norm": 3.856366258374826,
      "language_loss": 0.7991153,
      "learning_rate": 6.336168956469867e-08,
      "loss": 0.82054675,
      "num_input_tokens_seen": 165795435,
      "step": 7668,
      "time_per_iteration": 2.7543036937713623
    },
    {
      "auxiliary_loss_clip": 0.01120427,
      "auxiliary_loss_mlp": 0.01024895,
      "balance_loss_clip": 1.0465703,
      "balance_loss_mlp": 1.01786208,
      "epoch": 0.9221427283111886,
      "flos": 24790464875520.0,
      "grad_norm": 1.960236290584648,
      "language_loss": 0.7194804,
      "learning_rate": 6.316731488351168e-08,
      "loss": 0.74093366,
      "num_input_tokens_seen": 165816625,
      "step": 7669,
      "time_per_iteration": 2.8210573196411133
    },
    {
      "auxiliary_loss_clip": 0.01130386,
      "auxiliary_loss_mlp": 0.01024486,
      "balance_loss_clip": 1.04618955,
      "balance_loss_mlp": 1.01689863,
      "epoch": 0.9222629712018277,
      "flos": 13845036625920.0,
      "grad_norm": 2.4240002651953922,
      "language_loss": 0.63502526,
      "learning_rate": 6.297323401830334e-08,
      "loss": 0.65657401,
      "num_input_tokens_seen": 165835410,
      "step": 7670,
      "time_per_iteration": 2.7014307975769043
    },
    {
      "auxiliary_loss_clip": 0.01161736,
      "auxiliary_loss_mlp": 0.00762081,
      "balance_loss_clip": 1.04878187,
      "balance_loss_mlp": 1.00047803,
      "epoch": 0.9223832140924668,
      "flos": 21616177196160.0,
      "grad_norm": 2.3322147183025845,
      "language_loss": 0.69118458,
      "learning_rate": 6.277944699851523e-08,
      "loss": 0.71042275,
      "num_input_tokens_seen": 165854930,
      "step": 7671,
      "time_per_iteration": 2.7065203189849854
    },
    {
      "auxiliary_loss_clip": 0.01089753,
      "auxiliary_loss_mlp": 0.01031504,
      "balance_loss_clip": 1.0378933,
      "balance_loss_mlp": 1.0239048,
      "epoch": 0.9225034569831059,
      "flos": 21142084561920.0,
      "grad_norm": 1.7676470472209913,
      "language_loss": 0.73348439,
      "learning_rate": 6.25859538535447e-08,
      "loss": 0.75469691,
      "num_input_tokens_seen": 165875725,
      "step": 7672,
      "time_per_iteration": 2.9139668941497803
    },
    {
      "auxiliary_loss_clip": 0.01150651,
      "auxiliary_loss_mlp": 0.01028137,
      "balance_loss_clip": 1.0461874,
      "balance_loss_mlp": 1.02020347,
      "epoch": 0.9226236998737449,
      "flos": 12495046844160.0,
      "grad_norm": 2.5535339432821567,
      "language_loss": 0.77829242,
      "learning_rate": 6.239275461274474e-08,
      "loss": 0.80008036,
      "num_input_tokens_seen": 165892100,
      "step": 7673,
      "time_per_iteration": 2.8325488567352295
    },
    {
      "auxiliary_loss_clip": 0.0113045,
      "auxiliary_loss_mlp": 0.01026192,
      "balance_loss_clip": 1.0477047,
      "balance_loss_mlp": 1.0181272,
      "epoch": 0.9227439427643841,
      "flos": 26214071581440.0,
      "grad_norm": 2.398499944265665,
      "language_loss": 0.85961354,
      "learning_rate": 6.219984930542299e-08,
      "loss": 0.88117993,
      "num_input_tokens_seen": 165912840,
      "step": 7674,
      "time_per_iteration": 2.734222173690796
    },
    {
      "auxiliary_loss_clip": 0.01155234,
      "auxiliary_loss_mlp": 0.01028028,
      "balance_loss_clip": 1.05078745,
      "balance_loss_mlp": 1.02045798,
      "epoch": 0.9228641856550232,
      "flos": 17967581400960.0,
      "grad_norm": 2.273614876595366,
      "language_loss": 0.75971353,
      "learning_rate": 6.200723796084383e-08,
      "loss": 0.78154618,
      "num_input_tokens_seen": 165930935,
      "step": 7675,
      "time_per_iteration": 2.778852939605713
    },
    {
      "auxiliary_loss_clip": 0.01031489,
      "auxiliary_loss_mlp": 0.01001865,
      "balance_loss_clip": 1.00916409,
      "balance_loss_mlp": 1.00097084,
      "epoch": 0.9229844285456622,
      "flos": 70420609710720.0,
      "grad_norm": 0.7610062030619025,
      "language_loss": 0.62951326,
      "learning_rate": 6.181492060822546e-08,
      "loss": 0.64984679,
      "num_input_tokens_seen": 165991110,
      "step": 7676,
      "time_per_iteration": 3.2207131385803223
    },
    {
      "auxiliary_loss_clip": 0.01133921,
      "auxiliary_loss_mlp": 0.01024163,
      "balance_loss_clip": 1.04662848,
      "balance_loss_mlp": 1.01619971,
      "epoch": 0.9231046714363014,
      "flos": 17967832796160.0,
      "grad_norm": 2.5271367385252703,
      "language_loss": 0.82003695,
      "learning_rate": 6.162289727674274e-08,
      "loss": 0.84161782,
      "num_input_tokens_seen": 166008790,
      "step": 7677,
      "time_per_iteration": 3.9337892532348633
    },
    {
      "auxiliary_loss_clip": 0.01120158,
      "auxiliary_loss_mlp": 0.01027331,
      "balance_loss_clip": 1.04169416,
      "balance_loss_mlp": 1.0202198,
      "epoch": 0.9232249143269404,
      "flos": 17858233422720.0,
      "grad_norm": 2.2065941731418985,
      "language_loss": 0.87988245,
      "learning_rate": 6.143116799552527e-08,
      "loss": 0.90135729,
      "num_input_tokens_seen": 166025035,
      "step": 7678,
      "time_per_iteration": 3.681673526763916
    },
    {
      "auxiliary_loss_clip": 0.0111464,
      "auxiliary_loss_mlp": 0.01027742,
      "balance_loss_clip": 1.04513502,
      "balance_loss_mlp": 1.02017784,
      "epoch": 0.9233451572175795,
      "flos": 23404384903680.0,
      "grad_norm": 3.8050242967449637,
      "language_loss": 0.55821097,
      "learning_rate": 6.123973279365802e-08,
      "loss": 0.57963479,
      "num_input_tokens_seen": 166044010,
      "step": 7679,
      "time_per_iteration": 2.8008058071136475
    },
    {
      "auxiliary_loss_clip": 0.01154158,
      "auxiliary_loss_mlp": 0.00761916,
      "balance_loss_clip": 1.05201435,
      "balance_loss_mlp": 1.00045526,
      "epoch": 0.9234654001082186,
      "flos": 17999326045440.0,
      "grad_norm": 1.789658937520857,
      "language_loss": 0.77923197,
      "learning_rate": 6.10485917001824e-08,
      "loss": 0.79839271,
      "num_input_tokens_seen": 166061865,
      "step": 7680,
      "time_per_iteration": 2.737368106842041
    },
    {
      "auxiliary_loss_clip": 0.01162994,
      "auxiliary_loss_mlp": 0.01032559,
      "balance_loss_clip": 1.05032611,
      "balance_loss_mlp": 1.02496517,
      "epoch": 0.9235856429988577,
      "flos": 24750747411840.0,
      "grad_norm": 1.8209073372206934,
      "language_loss": 0.81044888,
      "learning_rate": 6.085774474409322e-08,
      "loss": 0.83240443,
      "num_input_tokens_seen": 166082425,
      "step": 7681,
      "time_per_iteration": 2.7924792766571045
    },
    {
      "auxiliary_loss_clip": 0.0115614,
      "auxiliary_loss_mlp": 0.01030438,
      "balance_loss_clip": 1.04879427,
      "balance_loss_mlp": 1.02258825,
      "epoch": 0.9237058858894968,
      "flos": 14099894599680.0,
      "grad_norm": 2.2298740605149487,
      "language_loss": 0.70797235,
      "learning_rate": 6.066719195434267e-08,
      "loss": 0.72983813,
      "num_input_tokens_seen": 166100225,
      "step": 7682,
      "time_per_iteration": 3.5971946716308594
    },
    {
      "auxiliary_loss_clip": 0.01136531,
      "auxiliary_loss_mlp": 0.01028061,
      "balance_loss_clip": 1.04685473,
      "balance_loss_mlp": 1.01926923,
      "epoch": 0.9238261287801359,
      "flos": 28694529175680.0,
      "grad_norm": 2.467334696969109,
      "language_loss": 0.66555381,
      "learning_rate": 6.047693335983717e-08,
      "loss": 0.68719971,
      "num_input_tokens_seen": 166122570,
      "step": 7683,
      "time_per_iteration": 2.821002721786499
    },
    {
      "auxiliary_loss_clip": 0.01132283,
      "auxiliary_loss_mlp": 0.00762314,
      "balance_loss_clip": 1.04285431,
      "balance_loss_mlp": 1.00048506,
      "epoch": 0.923946371670775,
      "flos": 23111856541440.0,
      "grad_norm": 2.218462467441709,
      "language_loss": 0.82651281,
      "learning_rate": 6.028696898943853e-08,
      "loss": 0.84545875,
      "num_input_tokens_seen": 166141630,
      "step": 7684,
      "time_per_iteration": 2.7270991802215576
    },
    {
      "auxiliary_loss_clip": 0.01123151,
      "auxiliary_loss_mlp": 0.01024671,
      "balance_loss_clip": 1.04320276,
      "balance_loss_mlp": 1.01596308,
      "epoch": 0.924066614561414,
      "flos": 21867120587520.0,
      "grad_norm": 2.120495953513091,
      "language_loss": 0.7091186,
      "learning_rate": 6.00972988719648e-08,
      "loss": 0.73059678,
      "num_input_tokens_seen": 166159865,
      "step": 7685,
      "time_per_iteration": 2.7774572372436523
    },
    {
      "auxiliary_loss_clip": 0.01137288,
      "auxiliary_loss_mlp": 0.01032297,
      "balance_loss_clip": 1.04577112,
      "balance_loss_mlp": 1.02439332,
      "epoch": 0.9241868574520532,
      "flos": 28511887495680.0,
      "grad_norm": 2.391139177180196,
      "language_loss": 0.70539784,
      "learning_rate": 5.990792303618807e-08,
      "loss": 0.7270937,
      "num_input_tokens_seen": 166179445,
      "step": 7686,
      "time_per_iteration": 2.789064407348633
    },
    {
      "auxiliary_loss_clip": 0.01145226,
      "auxiliary_loss_mlp": 0.01025229,
      "balance_loss_clip": 1.04678583,
      "balance_loss_mlp": 1.01771319,
      "epoch": 0.9243071003426923,
      "flos": 30518324282880.0,
      "grad_norm": 1.8193271873336427,
      "language_loss": 0.69763219,
      "learning_rate": 5.971884151083695e-08,
      "loss": 0.71933675,
      "num_input_tokens_seen": 166201855,
      "step": 7687,
      "time_per_iteration": 2.7426722049713135
    },
    {
      "auxiliary_loss_clip": 0.01148557,
      "auxiliary_loss_mlp": 0.01029206,
      "balance_loss_clip": 1.05010724,
      "balance_loss_mlp": 1.02172005,
      "epoch": 0.9244273432333313,
      "flos": 28658331244800.0,
      "grad_norm": 2.352116698469527,
      "language_loss": 0.74532139,
      "learning_rate": 5.9530054324595124e-08,
      "loss": 0.76709902,
      "num_input_tokens_seen": 166221970,
      "step": 7688,
      "time_per_iteration": 2.781749963760376
    },
    {
      "auxiliary_loss_clip": 0.01061088,
      "auxiliary_loss_mlp": 0.01000352,
      "balance_loss_clip": 1.00737166,
      "balance_loss_mlp": 0.99943411,
      "epoch": 0.9245475861239704,
      "flos": 66230589237120.0,
      "grad_norm": 0.7186531239160004,
      "language_loss": 0.57497776,
      "learning_rate": 5.934156150610103e-08,
      "loss": 0.59559214,
      "num_input_tokens_seen": 166279335,
      "step": 7689,
      "time_per_iteration": 3.2663590908050537
    },
    {
      "auxiliary_loss_clip": 0.01137073,
      "auxiliary_loss_mlp": 0.01026828,
      "balance_loss_clip": 1.04872811,
      "balance_loss_mlp": 1.01956797,
      "epoch": 0.9246678290146095,
      "flos": 24239918142720.0,
      "grad_norm": 5.100741435913334,
      "language_loss": 0.79249328,
      "learning_rate": 5.915336308394914e-08,
      "loss": 0.81413227,
      "num_input_tokens_seen": 166298170,
      "step": 7690,
      "time_per_iteration": 2.7584517002105713
    },
    {
      "auxiliary_loss_clip": 0.01142106,
      "auxiliary_loss_mlp": 0.01022499,
      "balance_loss_clip": 1.04787958,
      "balance_loss_mlp": 1.01521516,
      "epoch": 0.9247880719052486,
      "flos": 18988808976000.0,
      "grad_norm": 2.4516113790133343,
      "language_loss": 0.77485794,
      "learning_rate": 5.89654590866886e-08,
      "loss": 0.79650396,
      "num_input_tokens_seen": 166317670,
      "step": 7691,
      "time_per_iteration": 2.7017502784729004
    },
    {
      "auxiliary_loss_clip": 0.01140196,
      "auxiliary_loss_mlp": 0.01028309,
      "balance_loss_clip": 1.04816556,
      "balance_loss_mlp": 1.02029848,
      "epoch": 0.9249083147958876,
      "flos": 24024095274240.0,
      "grad_norm": 2.4809532330315287,
      "language_loss": 0.88442498,
      "learning_rate": 5.877784954282483e-08,
      "loss": 0.90611005,
      "num_input_tokens_seen": 166337010,
      "step": 7692,
      "time_per_iteration": 3.6329808235168457
    },
    {
      "auxiliary_loss_clip": 0.01102665,
      "auxiliary_loss_mlp": 0.0103156,
      "balance_loss_clip": 1.04373872,
      "balance_loss_mlp": 1.02279234,
      "epoch": 0.9250285576865268,
      "flos": 30773972355840.0,
      "grad_norm": 2.1060306073918813,
      "language_loss": 0.72499561,
      "learning_rate": 5.8590534480817963e-08,
      "loss": 0.74633789,
      "num_input_tokens_seen": 166358735,
      "step": 7693,
      "time_per_iteration": 2.8863003253936768
    },
    {
      "auxiliary_loss_clip": 0.01150121,
      "auxiliary_loss_mlp": 0.01029143,
      "balance_loss_clip": 1.04945517,
      "balance_loss_mlp": 1.02096522,
      "epoch": 0.9251488005771659,
      "flos": 10633581348480.0,
      "grad_norm": 2.488352226874728,
      "language_loss": 0.72321618,
      "learning_rate": 5.840351392908349e-08,
      "loss": 0.74500883,
      "num_input_tokens_seen": 166374455,
      "step": 7694,
      "time_per_iteration": 2.586905002593994
    },
    {
      "auxiliary_loss_clip": 0.01147925,
      "auxiliary_loss_mlp": 0.01030678,
      "balance_loss_clip": 1.05130827,
      "balance_loss_mlp": 1.02218413,
      "epoch": 0.9252690434678049,
      "flos": 23586416052480.0,
      "grad_norm": 5.62292455401385,
      "language_loss": 0.70830166,
      "learning_rate": 5.821678791599205e-08,
      "loss": 0.7300877,
      "num_input_tokens_seen": 166393900,
      "step": 7695,
      "time_per_iteration": 2.661566734313965
    },
    {
      "auxiliary_loss_clip": 0.01131168,
      "auxiliary_loss_mlp": 0.01029305,
      "balance_loss_clip": 1.0488838,
      "balance_loss_mlp": 1.02177703,
      "epoch": 0.9253892863584441,
      "flos": 21469158829440.0,
      "grad_norm": 1.6603844339937692,
      "language_loss": 0.80791533,
      "learning_rate": 5.803035646986965e-08,
      "loss": 0.82952011,
      "num_input_tokens_seen": 166413235,
      "step": 7696,
      "time_per_iteration": 2.625394582748413
    },
    {
      "auxiliary_loss_clip": 0.01130275,
      "auxiliary_loss_mlp": 0.01035277,
      "balance_loss_clip": 1.04399276,
      "balance_loss_mlp": 1.02631819,
      "epoch": 0.9255095292490831,
      "flos": 17456680304640.0,
      "grad_norm": 2.744178699058627,
      "language_loss": 0.67626536,
      "learning_rate": 5.7844219618998766e-08,
      "loss": 0.69792092,
      "num_input_tokens_seen": 166427560,
      "step": 7697,
      "time_per_iteration": 2.688899517059326
    },
    {
      "auxiliary_loss_clip": 0.01131222,
      "auxiliary_loss_mlp": 0.01022306,
      "balance_loss_clip": 1.04567814,
      "balance_loss_mlp": 1.01464665,
      "epoch": 0.9256297721397222,
      "flos": 24750675584640.0,
      "grad_norm": 1.7837922262934598,
      "language_loss": 0.71446353,
      "learning_rate": 5.765837739161505e-08,
      "loss": 0.73599881,
      "num_input_tokens_seen": 166446680,
      "step": 7698,
      "time_per_iteration": 2.7654812335968018
    },
    {
      "auxiliary_loss_clip": 0.01162371,
      "auxiliary_loss_mlp": 0.01023646,
      "balance_loss_clip": 1.04971266,
      "balance_loss_mlp": 1.01624262,
      "epoch": 0.9257500150303614,
      "flos": 23112215677440.0,
      "grad_norm": 4.058355063857064,
      "language_loss": 0.74718952,
      "learning_rate": 5.7472829815911504e-08,
      "loss": 0.76904964,
      "num_input_tokens_seen": 166465505,
      "step": 7699,
      "time_per_iteration": 2.578880548477173
    },
    {
      "auxiliary_loss_clip": 0.01160962,
      "auxiliary_loss_mlp": 0.01026999,
      "balance_loss_clip": 1.05217338,
      "balance_loss_mlp": 1.01922059,
      "epoch": 0.9258702579210004,
      "flos": 22564685687040.0,
      "grad_norm": 2.5537499999245394,
      "language_loss": 0.8143239,
      "learning_rate": 5.7287576920035164e-08,
      "loss": 0.83620358,
      "num_input_tokens_seen": 166484520,
      "step": 7700,
      "time_per_iteration": 2.7082886695861816
    },
    {
      "auxiliary_loss_clip": 0.01135004,
      "auxiliary_loss_mlp": 0.01022138,
      "balance_loss_clip": 1.04653728,
      "balance_loss_mlp": 1.01505947,
      "epoch": 0.9259905008116395,
      "flos": 30004298703360.0,
      "grad_norm": 2.9926502917655764,
      "language_loss": 0.76751804,
      "learning_rate": 5.7102618732088435e-08,
      "loss": 0.78908944,
      "num_input_tokens_seen": 166503850,
      "step": 7701,
      "time_per_iteration": 2.753857135772705
    },
    {
      "auxiliary_loss_clip": 0.0112351,
      "auxiliary_loss_mlp": 0.01024949,
      "balance_loss_clip": 1.04761195,
      "balance_loss_mlp": 1.01770997,
      "epoch": 0.9261107437022786,
      "flos": 24572128055040.0,
      "grad_norm": 2.566008359350624,
      "language_loss": 0.7452274,
      "learning_rate": 5.6917955280130216e-08,
      "loss": 0.76671195,
      "num_input_tokens_seen": 166525330,
      "step": 7702,
      "time_per_iteration": 2.7778072357177734
    },
    {
      "auxiliary_loss_clip": 0.01132892,
      "auxiliary_loss_mlp": 0.01028681,
      "balance_loss_clip": 1.04660082,
      "balance_loss_mlp": 1.02077758,
      "epoch": 0.9262309865929177,
      "flos": 22018448586240.0,
      "grad_norm": 2.1497810260744936,
      "language_loss": 0.71971756,
      "learning_rate": 5.6733586592172755e-08,
      "loss": 0.74133331,
      "num_input_tokens_seen": 166544825,
      "step": 7703,
      "time_per_iteration": 4.532064199447632
    },
    {
      "auxiliary_loss_clip": 0.01117923,
      "auxiliary_loss_mlp": 0.01027825,
      "balance_loss_clip": 1.0455513,
      "balance_loss_mlp": 1.0210216,
      "epoch": 0.9263512294835567,
      "flos": 20339481116160.0,
      "grad_norm": 2.095118135459375,
      "language_loss": 0.80160367,
      "learning_rate": 5.6549512696185244e-08,
      "loss": 0.82306117,
      "num_input_tokens_seen": 166563325,
      "step": 7704,
      "time_per_iteration": 2.719970703125
    },
    {
      "auxiliary_loss_clip": 0.01143142,
      "auxiliary_loss_mlp": 0.01025025,
      "balance_loss_clip": 1.04639602,
      "balance_loss_mlp": 1.01744652,
      "epoch": 0.9264714723741959,
      "flos": 21215378263680.0,
      "grad_norm": 1.8904168509418655,
      "language_loss": 0.68287444,
      "learning_rate": 5.636573362009156e-08,
      "loss": 0.70455611,
      "num_input_tokens_seen": 166583385,
      "step": 7705,
      "time_per_iteration": 2.7578794956207275
    },
    {
      "auxiliary_loss_clip": 0.01114456,
      "auxiliary_loss_mlp": 0.01026528,
      "balance_loss_clip": 1.04223323,
      "balance_loss_mlp": 1.01784635,
      "epoch": 0.926591715264835,
      "flos": 18004964480640.0,
      "grad_norm": 2.711669713176788,
      "language_loss": 0.76822197,
      "learning_rate": 5.618224939177074e-08,
      "loss": 0.78963184,
      "num_input_tokens_seen": 166601290,
      "step": 7706,
      "time_per_iteration": 2.695927858352661
    },
    {
      "auxiliary_loss_clip": 0.01153221,
      "auxiliary_loss_mlp": 0.0103742,
      "balance_loss_clip": 1.0489819,
      "balance_loss_mlp": 1.03005314,
      "epoch": 0.926711958155474,
      "flos": 36167969825280.0,
      "grad_norm": 2.3775593791019753,
      "language_loss": 0.70277393,
      "learning_rate": 5.599906003905719e-08,
      "loss": 0.7246803,
      "num_input_tokens_seen": 166623835,
      "step": 7707,
      "time_per_iteration": 3.738612413406372
    },
    {
      "auxiliary_loss_clip": 0.01144213,
      "auxiliary_loss_mlp": 0.0103351,
      "balance_loss_clip": 1.04979432,
      "balance_loss_mlp": 1.02583838,
      "epoch": 0.9268322010461132,
      "flos": 21032736583680.0,
      "grad_norm": 2.3186741207477843,
      "language_loss": 0.81432724,
      "learning_rate": 5.581616558974023e-08,
      "loss": 0.83610439,
      "num_input_tokens_seen": 166642400,
      "step": 7708,
      "time_per_iteration": 2.775979995727539
    },
    {
      "auxiliary_loss_clip": 0.01129504,
      "auxiliary_loss_mlp": 0.01029874,
      "balance_loss_clip": 1.04366207,
      "balance_loss_mlp": 1.02118361,
      "epoch": 0.9269524439367522,
      "flos": 22964838174720.0,
      "grad_norm": 2.009865774629877,
      "language_loss": 0.78782356,
      "learning_rate": 5.5633566071565444e-08,
      "loss": 0.80941731,
      "num_input_tokens_seen": 166661640,
      "step": 7709,
      "time_per_iteration": 2.726161241531372
    },
    {
      "auxiliary_loss_clip": 0.01154011,
      "auxiliary_loss_mlp": 0.0076128,
      "balance_loss_clip": 1.0491842,
      "balance_loss_mlp": 1.00045204,
      "epoch": 0.9270726868273913,
      "flos": 41975551468800.0,
      "grad_norm": 2.741302106385045,
      "language_loss": 0.71265793,
      "learning_rate": 5.5451261512232896e-08,
      "loss": 0.73181081,
      "num_input_tokens_seen": 166684320,
      "step": 7710,
      "time_per_iteration": 2.838465929031372
    },
    {
      "auxiliary_loss_clip": 0.01147487,
      "auxiliary_loss_mlp": 0.01032047,
      "balance_loss_clip": 1.04669499,
      "balance_loss_mlp": 1.02404845,
      "epoch": 0.9271929297180305,
      "flos": 19791771557760.0,
      "grad_norm": 1.8668749424752669,
      "language_loss": 0.62517011,
      "learning_rate": 5.5269251939397576e-08,
      "loss": 0.64696538,
      "num_input_tokens_seen": 166703835,
      "step": 7711,
      "time_per_iteration": 2.685976266860962
    },
    {
      "auxiliary_loss_clip": 0.01172569,
      "auxiliary_loss_mlp": 0.01025765,
      "balance_loss_clip": 1.05055189,
      "balance_loss_mlp": 1.017748,
      "epoch": 0.9273131726086695,
      "flos": 19968343839360.0,
      "grad_norm": 3.6056940772196953,
      "language_loss": 0.76848364,
      "learning_rate": 5.508753738067073e-08,
      "loss": 0.79046696,
      "num_input_tokens_seen": 166723375,
      "step": 7712,
      "time_per_iteration": 2.7571568489074707
    },
    {
      "auxiliary_loss_clip": 0.01171762,
      "auxiliary_loss_mlp": 0.0102757,
      "balance_loss_clip": 1.0501895,
      "balance_loss_mlp": 1.01990485,
      "epoch": 0.9274334154993086,
      "flos": 23258587599360.0,
      "grad_norm": 4.163770734121761,
      "language_loss": 0.79347181,
      "learning_rate": 5.4906117863617875e-08,
      "loss": 0.81546509,
      "num_input_tokens_seen": 166742760,
      "step": 7713,
      "time_per_iteration": 2.609722852706909
    },
    {
      "auxiliary_loss_clip": 0.01155887,
      "auxiliary_loss_mlp": 0.01025323,
      "balance_loss_clip": 1.04965973,
      "balance_loss_mlp": 1.01803958,
      "epoch": 0.9275536583899477,
      "flos": 31795343585280.0,
      "grad_norm": 1.8700250842284212,
      "language_loss": 0.78118241,
      "learning_rate": 5.4724993415760533e-08,
      "loss": 0.80299449,
      "num_input_tokens_seen": 166761115,
      "step": 7714,
      "time_per_iteration": 2.8179879188537598
    },
    {
      "auxiliary_loss_clip": 0.01150597,
      "auxiliary_loss_mlp": 0.01024477,
      "balance_loss_clip": 1.04840696,
      "balance_loss_mlp": 1.01673484,
      "epoch": 0.9276739012805868,
      "flos": 18696998885760.0,
      "grad_norm": 2.8652999678184528,
      "language_loss": 0.74793255,
      "learning_rate": 5.454416406457496e-08,
      "loss": 0.76968324,
      "num_input_tokens_seen": 166780210,
      "step": 7715,
      "time_per_iteration": 2.6841015815734863
    },
    {
      "auxiliary_loss_clip": 0.01140155,
      "auxiliary_loss_mlp": 0.01026991,
      "balance_loss_clip": 1.04980588,
      "balance_loss_mlp": 1.01958787,
      "epoch": 0.9277941441712259,
      "flos": 13879079740800.0,
      "grad_norm": 3.180786617124237,
      "language_loss": 0.73452556,
      "learning_rate": 5.436362983749299e-08,
      "loss": 0.75619698,
      "num_input_tokens_seen": 166795380,
      "step": 7716,
      "time_per_iteration": 2.7432756423950195
    },
    {
      "auxiliary_loss_clip": 0.01146417,
      "auxiliary_loss_mlp": 0.0102793,
      "balance_loss_clip": 1.05007291,
      "balance_loss_mlp": 1.02059245,
      "epoch": 0.927914387061865,
      "flos": 23258659426560.0,
      "grad_norm": 1.9740094253381932,
      "language_loss": 0.64754999,
      "learning_rate": 5.418339076190137e-08,
      "loss": 0.66929346,
      "num_input_tokens_seen": 166814890,
      "step": 7717,
      "time_per_iteration": 2.747709274291992
    },
    {
      "auxiliary_loss_clip": 0.01114477,
      "auxiliary_loss_mlp": 0.01024309,
      "balance_loss_clip": 1.04402494,
      "balance_loss_mlp": 1.01630449,
      "epoch": 0.9280346299525041,
      "flos": 18073733068800.0,
      "grad_norm": 1.7185253965853247,
      "language_loss": 0.88850862,
      "learning_rate": 5.400344686514202e-08,
      "loss": 0.90989655,
      "num_input_tokens_seen": 166832475,
      "step": 7718,
      "time_per_iteration": 3.7992212772369385
    },
    {
      "auxiliary_loss_clip": 0.01112641,
      "auxiliary_loss_mlp": 0.01026288,
      "balance_loss_clip": 1.04400492,
      "balance_loss_mlp": 1.0189985,
      "epoch": 0.9281548728431431,
      "flos": 22342901160960.0,
      "grad_norm": 2.280639601548766,
      "language_loss": 0.66899574,
      "learning_rate": 5.38237981745131e-08,
      "loss": 0.69038498,
      "num_input_tokens_seen": 166850590,
      "step": 7719,
      "time_per_iteration": 2.7817189693450928
    },
    {
      "auxiliary_loss_clip": 0.0115761,
      "auxiliary_loss_mlp": 0.01028777,
      "balance_loss_clip": 1.04985404,
      "balance_loss_mlp": 1.02124274,
      "epoch": 0.9282751157337822,
      "flos": 18843765857280.0,
      "grad_norm": 1.7300197762807183,
      "language_loss": 0.81502998,
      "learning_rate": 5.364444471726592e-08,
      "loss": 0.8368938,
      "num_input_tokens_seen": 166869795,
      "step": 7720,
      "time_per_iteration": 2.639681816101074
    },
    {
      "auxiliary_loss_clip": 0.01097618,
      "auxiliary_loss_mlp": 0.01023465,
      "balance_loss_clip": 1.03769755,
      "balance_loss_mlp": 1.01567435,
      "epoch": 0.9283953586244214,
      "flos": 25556834476800.0,
      "grad_norm": 1.9075004434508565,
      "language_loss": 0.80093896,
      "learning_rate": 5.346538652060939e-08,
      "loss": 0.82214975,
      "num_input_tokens_seen": 166891150,
      "step": 7721,
      "time_per_iteration": 2.847564220428467
    },
    {
      "auxiliary_loss_clip": 0.01131059,
      "auxiliary_loss_mlp": 0.0102586,
      "balance_loss_clip": 1.04492736,
      "balance_loss_mlp": 1.01802158,
      "epoch": 0.9285156015150604,
      "flos": 18223480869120.0,
      "grad_norm": 1.8265935906391955,
      "language_loss": 0.70304775,
      "learning_rate": 5.3286623611705994e-08,
      "loss": 0.724617,
      "num_input_tokens_seen": 166909195,
      "step": 7722,
      "time_per_iteration": 2.6973166465759277
    },
    {
      "auxiliary_loss_clip": 0.01040136,
      "auxiliary_loss_mlp": 0.01003634,
      "balance_loss_clip": 1.00872135,
      "balance_loss_mlp": 1.00273407,
      "epoch": 0.9286358444056995,
      "flos": 66400017690240.0,
      "grad_norm": 0.8105837251023176,
      "language_loss": 0.60544419,
      "learning_rate": 5.3108156017673824e-08,
      "loss": 0.62588191,
      "num_input_tokens_seen": 166970955,
      "step": 7723,
      "time_per_iteration": 3.4733901023864746
    },
    {
      "auxiliary_loss_clip": 0.01080887,
      "auxiliary_loss_mlp": 0.0103133,
      "balance_loss_clip": 1.03892183,
      "balance_loss_mlp": 1.02300906,
      "epoch": 0.9287560872963386,
      "flos": 22345630594560.0,
      "grad_norm": 1.7032696496685404,
      "language_loss": 0.71669483,
      "learning_rate": 5.2929983765586775e-08,
      "loss": 0.73781699,
      "num_input_tokens_seen": 166989735,
      "step": 7724,
      "time_per_iteration": 3.092362642288208
    },
    {
      "auxiliary_loss_clip": 0.0111916,
      "auxiliary_loss_mlp": 0.01032246,
      "balance_loss_clip": 1.04892302,
      "balance_loss_mlp": 1.02500439,
      "epoch": 0.9288763301869777,
      "flos": 25700225569920.0,
      "grad_norm": 1.8790870852778403,
      "language_loss": 0.62595189,
      "learning_rate": 5.275210688247278e-08,
      "loss": 0.64746594,
      "num_input_tokens_seen": 167010060,
      "step": 7725,
      "time_per_iteration": 2.9536080360412598
    },
    {
      "auxiliary_loss_clip": 0.01134905,
      "auxiliary_loss_mlp": 0.01028252,
      "balance_loss_clip": 1.04644942,
      "balance_loss_mlp": 1.02062273,
      "epoch": 0.9289965730776167,
      "flos": 12312046028160.0,
      "grad_norm": 2.3948683470580243,
      "language_loss": 0.85082626,
      "learning_rate": 5.257452539531604e-08,
      "loss": 0.87245786,
      "num_input_tokens_seen": 167027130,
      "step": 7726,
      "time_per_iteration": 2.74112606048584
    },
    {
      "auxiliary_loss_clip": 0.01121516,
      "auxiliary_loss_mlp": 0.00761519,
      "balance_loss_clip": 1.04250669,
      "balance_loss_mlp": 1.00044191,
      "epoch": 0.9291168159682559,
      "flos": 26685973486080.0,
      "grad_norm": 3.371033468255602,
      "language_loss": 0.68538892,
      "learning_rate": 5.2397239331055445e-08,
      "loss": 0.70421922,
      "num_input_tokens_seen": 167049130,
      "step": 7727,
      "time_per_iteration": 2.7606027126312256
    },
    {
      "auxiliary_loss_clip": 0.01120699,
      "auxiliary_loss_mlp": 0.01026651,
      "balance_loss_clip": 1.04537606,
      "balance_loss_mlp": 1.019045,
      "epoch": 0.929237058858895,
      "flos": 14538256179840.0,
      "grad_norm": 2.052075196478101,
      "language_loss": 0.8117944,
      "learning_rate": 5.2220248716585036e-08,
      "loss": 0.83326787,
      "num_input_tokens_seen": 167066810,
      "step": 7728,
      "time_per_iteration": 2.783764362335205
    },
    {
      "auxiliary_loss_clip": 0.01156688,
      "auxiliary_loss_mlp": 0.01033065,
      "balance_loss_clip": 1.05134559,
      "balance_loss_mlp": 1.02521539,
      "epoch": 0.929357301749534,
      "flos": 23835456023040.0,
      "grad_norm": 2.4136952766620965,
      "language_loss": 0.75482672,
      "learning_rate": 5.204355357875445e-08,
      "loss": 0.77672434,
      "num_input_tokens_seen": 167085155,
      "step": 7729,
      "time_per_iteration": 3.719468116760254
    },
    {
      "auxiliary_loss_clip": 0.01161665,
      "auxiliary_loss_mlp": 0.00762032,
      "balance_loss_clip": 1.04896545,
      "balance_loss_mlp": 1.00051451,
      "epoch": 0.9294775446401732,
      "flos": 12969319046400.0,
      "grad_norm": 2.2569300390669054,
      "language_loss": 0.70408034,
      "learning_rate": 5.1867153944367584e-08,
      "loss": 0.72331738,
      "num_input_tokens_seen": 167101545,
      "step": 7730,
      "time_per_iteration": 2.6366477012634277
    },
    {
      "auxiliary_loss_clip": 0.01125771,
      "auxiliary_loss_mlp": 0.01028127,
      "balance_loss_clip": 1.04449868,
      "balance_loss_mlp": 1.02069139,
      "epoch": 0.9295977875308122,
      "flos": 26211809024640.0,
      "grad_norm": 2.4894508239210653,
      "language_loss": 0.73749435,
      "learning_rate": 5.16910498401848e-08,
      "loss": 0.75903332,
      "num_input_tokens_seen": 167120995,
      "step": 7731,
      "time_per_iteration": 2.797480821609497
    },
    {
      "auxiliary_loss_clip": 0.01175065,
      "auxiliary_loss_mlp": 0.01029294,
      "balance_loss_clip": 1.05369782,
      "balance_loss_mlp": 1.0224309,
      "epoch": 0.9297180304214513,
      "flos": 16472297105280.0,
      "grad_norm": 11.507557484923131,
      "language_loss": 0.83571458,
      "learning_rate": 5.151524129292073e-08,
      "loss": 0.85775816,
      "num_input_tokens_seen": 167138890,
      "step": 7732,
      "time_per_iteration": 2.624285936355591
    },
    {
      "auxiliary_loss_clip": 0.01174016,
      "auxiliary_loss_mlp": 0.01029009,
      "balance_loss_clip": 1.05096054,
      "balance_loss_mlp": 1.02124238,
      "epoch": 0.9298382733120905,
      "flos": 24060436859520.0,
      "grad_norm": 2.375624719022283,
      "language_loss": 0.66936922,
      "learning_rate": 5.1339728329245155e-08,
      "loss": 0.69139946,
      "num_input_tokens_seen": 167159455,
      "step": 7733,
      "time_per_iteration": 3.598973274230957
    },
    {
      "auxiliary_loss_clip": 0.01177483,
      "auxiliary_loss_mlp": 0.01036524,
      "balance_loss_clip": 1.05187058,
      "balance_loss_mlp": 1.02808356,
      "epoch": 0.9299585162027295,
      "flos": 22127652910080.0,
      "grad_norm": 2.479123083332457,
      "language_loss": 0.79414582,
      "learning_rate": 5.116451097578367e-08,
      "loss": 0.81628585,
      "num_input_tokens_seen": 167178495,
      "step": 7734,
      "time_per_iteration": 2.692650556564331
    },
    {
      "auxiliary_loss_clip": 0.01158783,
      "auxiliary_loss_mlp": 0.01028856,
      "balance_loss_clip": 1.05498123,
      "balance_loss_mlp": 1.02179551,
      "epoch": 0.9300787590933686,
      "flos": 21471780522240.0,
      "grad_norm": 3.217842093839665,
      "language_loss": 0.74269795,
      "learning_rate": 5.0989589259115895e-08,
      "loss": 0.76457429,
      "num_input_tokens_seen": 167199380,
      "step": 7735,
      "time_per_iteration": 2.665314197540283
    },
    {
      "auxiliary_loss_clip": 0.01163361,
      "auxiliary_loss_mlp": 0.01029343,
      "balance_loss_clip": 1.05198956,
      "balance_loss_mlp": 1.02104604,
      "epoch": 0.9301990019840077,
      "flos": 17779588594560.0,
      "grad_norm": 2.340307851852093,
      "language_loss": 0.71898675,
      "learning_rate": 5.081496320577816e-08,
      "loss": 0.74091375,
      "num_input_tokens_seen": 167216500,
      "step": 7736,
      "time_per_iteration": 2.664912462234497
    },
    {
      "auxiliary_loss_clip": 0.01042614,
      "auxiliary_loss_mlp": 0.00998791,
      "balance_loss_clip": 1.0082438,
      "balance_loss_mlp": 0.99790841,
      "epoch": 0.9303192448746468,
      "flos": 58896122307840.0,
      "grad_norm": 0.918446271712147,
      "language_loss": 0.61126482,
      "learning_rate": 5.0640632842260835e-08,
      "loss": 0.63167894,
      "num_input_tokens_seen": 167276760,
      "step": 7737,
      "time_per_iteration": 3.287834405899048
    },
    {
      "auxiliary_loss_clip": 0.01173165,
      "auxiliary_loss_mlp": 0.01029604,
      "balance_loss_clip": 1.05120993,
      "balance_loss_mlp": 1.02188563,
      "epoch": 0.9304394877652858,
      "flos": 57663522172800.0,
      "grad_norm": 1.578242708684971,
      "language_loss": 0.73056751,
      "learning_rate": 5.0466598195009426e-08,
      "loss": 0.75259525,
      "num_input_tokens_seen": 167303630,
      "step": 7738,
      "time_per_iteration": 2.931312322616577
    },
    {
      "auxiliary_loss_clip": 0.01114026,
      "auxiliary_loss_mlp": 0.0102903,
      "balance_loss_clip": 1.04310369,
      "balance_loss_mlp": 1.02112067,
      "epoch": 0.930559730655925,
      "flos": 20996143603200.0,
      "grad_norm": 1.9001665861903902,
      "language_loss": 0.70092297,
      "learning_rate": 5.0292859290425036e-08,
      "loss": 0.72235358,
      "num_input_tokens_seen": 167321500,
      "step": 7739,
      "time_per_iteration": 2.7582476139068604
    },
    {
      "auxiliary_loss_clip": 0.01145016,
      "auxiliary_loss_mlp": 0.01024641,
      "balance_loss_clip": 1.05074012,
      "balance_loss_mlp": 1.01690412,
      "epoch": 0.9306799735465641,
      "flos": 23258264376960.0,
      "grad_norm": 4.241210951347444,
      "language_loss": 0.77864391,
      "learning_rate": 5.011941615486348e-08,
      "loss": 0.80034047,
      "num_input_tokens_seen": 167340615,
      "step": 7740,
      "time_per_iteration": 2.680640935897827
    },
    {
      "auxiliary_loss_clip": 0.01147801,
      "auxiliary_loss_mlp": 0.01023249,
      "balance_loss_clip": 1.04937553,
      "balance_loss_mlp": 1.01605475,
      "epoch": 0.9308002164372031,
      "flos": 15231547560960.0,
      "grad_norm": 2.7874037246185734,
      "language_loss": 0.84596813,
      "learning_rate": 4.994626881463659e-08,
      "loss": 0.86767864,
      "num_input_tokens_seen": 167356870,
      "step": 7741,
      "time_per_iteration": 2.675292491912842
    },
    {
      "auxiliary_loss_clip": 0.01152684,
      "auxiliary_loss_mlp": 0.01033804,
      "balance_loss_clip": 1.04846966,
      "balance_loss_mlp": 1.02643657,
      "epoch": 0.9309204593278423,
      "flos": 30847481539200.0,
      "grad_norm": 2.812931922226318,
      "language_loss": 0.71371365,
      "learning_rate": 4.9773417296009814e-08,
      "loss": 0.73557854,
      "num_input_tokens_seen": 167378390,
      "step": 7742,
      "time_per_iteration": 2.741659164428711
    },
    {
      "auxiliary_loss_clip": 0.01147049,
      "auxiliary_loss_mlp": 0.01032316,
      "balance_loss_clip": 1.05159211,
      "balance_loss_mlp": 1.02326226,
      "epoch": 0.9310407022184813,
      "flos": 23037269950080.0,
      "grad_norm": 2.116104443710169,
      "language_loss": 0.65661871,
      "learning_rate": 4.960086162520527e-08,
      "loss": 0.67841244,
      "num_input_tokens_seen": 167398480,
      "step": 7743,
      "time_per_iteration": 2.745938301086426
    },
    {
      "auxiliary_loss_clip": 0.01133024,
      "auxiliary_loss_mlp": 0.01025273,
      "balance_loss_clip": 1.04562938,
      "balance_loss_mlp": 1.01806736,
      "epoch": 0.9311609451091204,
      "flos": 22127976132480.0,
      "grad_norm": 2.2820792249102397,
      "language_loss": 0.82066023,
      "learning_rate": 4.942860182839936e-08,
      "loss": 0.84224319,
      "num_input_tokens_seen": 167416825,
      "step": 7744,
      "time_per_iteration": 2.7984580993652344
    },
    {
      "auxiliary_loss_clip": 0.01123273,
      "auxiliary_loss_mlp": 0.0102975,
      "balance_loss_clip": 1.04714489,
      "balance_loss_mlp": 1.02181053,
      "epoch": 0.9312811879997596,
      "flos": 21099206701440.0,
      "grad_norm": 2.0195026411338852,
      "language_loss": 0.79986489,
      "learning_rate": 4.925663793172341e-08,
      "loss": 0.8213951,
      "num_input_tokens_seen": 167434785,
      "step": 7745,
      "time_per_iteration": 3.8585829734802246
    },
    {
      "auxiliary_loss_clip": 0.01043914,
      "auxiliary_loss_mlp": 0.00999755,
      "balance_loss_clip": 1.0157187,
      "balance_loss_mlp": 0.99880737,
      "epoch": 0.9314014308903986,
      "flos": 67148179096320.0,
      "grad_norm": 0.7863956602957382,
      "language_loss": 0.56443924,
      "learning_rate": 4.908496996126477e-08,
      "loss": 0.58487588,
      "num_input_tokens_seen": 167498245,
      "step": 7746,
      "time_per_iteration": 3.3164029121398926
    },
    {
      "auxiliary_loss_clip": 0.0110493,
      "auxiliary_loss_mlp": 0.01026975,
      "balance_loss_clip": 1.04678535,
      "balance_loss_mlp": 1.01844549,
      "epoch": 0.9315216737810377,
      "flos": 22565583527040.0,
      "grad_norm": 1.7680706961847341,
      "language_loss": 0.76574504,
      "learning_rate": 4.89135979430646e-08,
      "loss": 0.78706408,
      "num_input_tokens_seen": 167518290,
      "step": 7747,
      "time_per_iteration": 2.7904064655303955
    },
    {
      "auxiliary_loss_clip": 0.01158961,
      "auxiliary_loss_mlp": 0.01029915,
      "balance_loss_clip": 1.05054069,
      "balance_loss_mlp": 1.02212477,
      "epoch": 0.9316419166716768,
      "flos": 23984054588160.0,
      "grad_norm": 2.0930757585134514,
      "language_loss": 0.85665894,
      "learning_rate": 4.874252190312078e-08,
      "loss": 0.87854767,
      "num_input_tokens_seen": 167538675,
      "step": 7748,
      "time_per_iteration": 2.6778225898742676
    },
    {
      "auxiliary_loss_clip": 0.01141605,
      "auxiliary_loss_mlp": 0.01026213,
      "balance_loss_clip": 1.04733491,
      "balance_loss_mlp": 1.01807761,
      "epoch": 0.9317621595623159,
      "flos": 30230464688640.0,
      "grad_norm": 2.7584037732778484,
      "language_loss": 0.64950055,
      "learning_rate": 4.857174186738477e-08,
      "loss": 0.6711787,
      "num_input_tokens_seen": 167562025,
      "step": 7749,
      "time_per_iteration": 2.8064727783203125
    },
    {
      "auxiliary_loss_clip": 0.01151425,
      "auxiliary_loss_mlp": 0.01029112,
      "balance_loss_clip": 1.05248237,
      "balance_loss_mlp": 1.02055836,
      "epoch": 0.931882402452955,
      "flos": 15742735966080.0,
      "grad_norm": 2.713740432117461,
      "language_loss": 0.73578525,
      "learning_rate": 4.840125786176408e-08,
      "loss": 0.75759065,
      "num_input_tokens_seen": 167578230,
      "step": 7750,
      "time_per_iteration": 2.7160162925720215
    },
    {
      "auxiliary_loss_clip": 0.01112335,
      "auxiliary_loss_mlp": 0.01028974,
      "balance_loss_clip": 1.04392099,
      "balance_loss_mlp": 1.02162445,
      "epoch": 0.932002645343594,
      "flos": 28366521154560.0,
      "grad_norm": 2.443140471560337,
      "language_loss": 0.7784096,
      "learning_rate": 4.823106991212067e-08,
      "loss": 0.79982269,
      "num_input_tokens_seen": 167597470,
      "step": 7751,
      "time_per_iteration": 2.7667770385742188
    },
    {
      "auxiliary_loss_clip": 0.01128231,
      "auxiliary_loss_mlp": 0.01028239,
      "balance_loss_clip": 1.04455376,
      "balance_loss_mlp": 1.0209527,
      "epoch": 0.9321228882342332,
      "flos": 15341146934400.0,
      "grad_norm": 2.2875099596591975,
      "language_loss": 0.83540165,
      "learning_rate": 4.806117804427212e-08,
      "loss": 0.85696632,
      "num_input_tokens_seen": 167615405,
      "step": 7752,
      "time_per_iteration": 2.777665853500366
    },
    {
      "auxiliary_loss_clip": 0.01133728,
      "auxiliary_loss_mlp": 0.0102848,
      "balance_loss_clip": 1.04626715,
      "balance_loss_mlp": 1.02049947,
      "epoch": 0.9322431311248722,
      "flos": 17895365107200.0,
      "grad_norm": 1.8636524389884692,
      "language_loss": 0.64496994,
      "learning_rate": 4.7891582283990926e-08,
      "loss": 0.666592,
      "num_input_tokens_seen": 167634130,
      "step": 7753,
      "time_per_iteration": 2.6925957202911377
    },
    {
      "auxiliary_loss_clip": 0.01106536,
      "auxiliary_loss_mlp": 0.01028703,
      "balance_loss_clip": 1.04121435,
      "balance_loss_mlp": 1.02119255,
      "epoch": 0.9323633740155113,
      "flos": 24169713010560.0,
      "grad_norm": 1.5800492701988829,
      "language_loss": 0.73020375,
      "learning_rate": 4.772228265700473e-08,
      "loss": 0.75155616,
      "num_input_tokens_seen": 167654990,
      "step": 7754,
      "time_per_iteration": 2.8441405296325684
    },
    {
      "auxiliary_loss_clip": 0.01145995,
      "auxiliary_loss_mlp": 0.01032615,
      "balance_loss_clip": 1.04769373,
      "balance_loss_mlp": 1.02462196,
      "epoch": 0.9324836169061504,
      "flos": 15043482927360.0,
      "grad_norm": 3.3632476663926725,
      "language_loss": 0.75772202,
      "learning_rate": 4.75532791889961e-08,
      "loss": 0.77950811,
      "num_input_tokens_seen": 167671690,
      "step": 7755,
      "time_per_iteration": 3.6489369869232178
    },
    {
      "auxiliary_loss_clip": 0.01136975,
      "auxiliary_loss_mlp": 0.00761633,
      "balance_loss_clip": 1.04816532,
      "balance_loss_mlp": 1.0005455,
      "epoch": 0.9326038597967895,
      "flos": 18624890332800.0,
      "grad_norm": 1.8186853229189952,
      "language_loss": 0.66016793,
      "learning_rate": 4.738457190560252e-08,
      "loss": 0.67915392,
      "num_input_tokens_seen": 167690800,
      "step": 7756,
      "time_per_iteration": 2.68955659866333
    },
    {
      "auxiliary_loss_clip": 0.01091195,
      "auxiliary_loss_mlp": 0.01038986,
      "balance_loss_clip": 1.03922343,
      "balance_loss_mlp": 1.03077853,
      "epoch": 0.9327241026874286,
      "flos": 18952646958720.0,
      "grad_norm": 2.3708024856842305,
      "language_loss": 0.78712046,
      "learning_rate": 4.721616083241664e-08,
      "loss": 0.80842221,
      "num_input_tokens_seen": 167709055,
      "step": 7757,
      "time_per_iteration": 2.8617124557495117
    },
    {
      "auxiliary_loss_clip": 0.01120325,
      "auxiliary_loss_mlp": 0.01029848,
      "balance_loss_clip": 1.04883695,
      "balance_loss_mlp": 1.02209306,
      "epoch": 0.9328443455780677,
      "flos": 29570282668800.0,
      "grad_norm": 1.6412394817964309,
      "language_loss": 0.77715862,
      "learning_rate": 4.7048045994986684e-08,
      "loss": 0.79866034,
      "num_input_tokens_seen": 167729915,
      "step": 7758,
      "time_per_iteration": 2.75067138671875
    },
    {
      "auxiliary_loss_clip": 0.01115818,
      "auxiliary_loss_mlp": 0.01027683,
      "balance_loss_clip": 1.04501367,
      "balance_loss_mlp": 1.01973462,
      "epoch": 0.9329645884687068,
      "flos": 30081722469120.0,
      "grad_norm": 2.664923732654687,
      "language_loss": 0.91213644,
      "learning_rate": 4.688022741881559e-08,
      "loss": 0.93357146,
      "num_input_tokens_seen": 167750440,
      "step": 7759,
      "time_per_iteration": 3.6714155673980713
    },
    {
      "auxiliary_loss_clip": 0.01137903,
      "auxiliary_loss_mlp": 0.01027382,
      "balance_loss_clip": 1.04787016,
      "balance_loss_mlp": 1.02022624,
      "epoch": 0.9330848313593458,
      "flos": 21867982513920.0,
      "grad_norm": 1.7581715020763,
      "language_loss": 0.75267273,
      "learning_rate": 4.671270512936076e-08,
      "loss": 0.77432561,
      "num_input_tokens_seen": 167769600,
      "step": 7760,
      "time_per_iteration": 2.714118242263794
    },
    {
      "auxiliary_loss_clip": 0.01130181,
      "auxiliary_loss_mlp": 0.0102445,
      "balance_loss_clip": 1.043082,
      "balance_loss_mlp": 1.01746726,
      "epoch": 0.933205074249985,
      "flos": 22127221946880.0,
      "grad_norm": 1.912085089596499,
      "language_loss": 0.82952839,
      "learning_rate": 4.6545479152035884e-08,
      "loss": 0.8510747,
      "num_input_tokens_seen": 167788770,
      "step": 7761,
      "time_per_iteration": 2.6907224655151367
    },
    {
      "auxiliary_loss_clip": 0.01140831,
      "auxiliary_loss_mlp": 0.01024591,
      "balance_loss_clip": 1.04594779,
      "balance_loss_mlp": 1.01644278,
      "epoch": 0.9333253171406241,
      "flos": 15341254675200.0,
      "grad_norm": 3.181947708916566,
      "language_loss": 0.76241291,
      "learning_rate": 4.637854951220821e-08,
      "loss": 0.78406715,
      "num_input_tokens_seen": 167805555,
      "step": 7762,
      "time_per_iteration": 2.6875784397125244
    },
    {
      "auxiliary_loss_clip": 0.01155098,
      "auxiliary_loss_mlp": 0.01028348,
      "balance_loss_clip": 1.0450232,
      "balance_loss_mlp": 1.02112424,
      "epoch": 0.9334455600312631,
      "flos": 15706142985600.0,
      "grad_norm": 1.7406301491367515,
      "language_loss": 0.75193453,
      "learning_rate": 4.621191623520171e-08,
      "loss": 0.77376902,
      "num_input_tokens_seen": 167823985,
      "step": 7763,
      "time_per_iteration": 2.563272476196289
    },
    {
      "auxiliary_loss_clip": 0.01107066,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.04065895,
      "balance_loss_mlp": 1.01935005,
      "epoch": 0.9335658029219023,
      "flos": 22163563532160.0,
      "grad_norm": 2.5647728714693976,
      "language_loss": 0.84772825,
      "learning_rate": 4.604557934629372e-08,
      "loss": 0.86907083,
      "num_input_tokens_seen": 167843060,
      "step": 7764,
      "time_per_iteration": 2.6598339080810547
    },
    {
      "auxiliary_loss_clip": 0.01128177,
      "auxiliary_loss_mlp": 0.0102947,
      "balance_loss_clip": 1.04295373,
      "balance_loss_mlp": 1.02154541,
      "epoch": 0.9336860458125413,
      "flos": 20266833859200.0,
      "grad_norm": 2.0045000991503694,
      "language_loss": 0.80719888,
      "learning_rate": 4.587953887071805e-08,
      "loss": 0.82877535,
      "num_input_tokens_seen": 167862880,
      "step": 7765,
      "time_per_iteration": 2.593036651611328
    },
    {
      "auxiliary_loss_clip": 0.01137822,
      "auxiliary_loss_mlp": 0.01027751,
      "balance_loss_clip": 1.04505587,
      "balance_loss_mlp": 1.02009749,
      "epoch": 0.9338062887031804,
      "flos": 20919689504640.0,
      "grad_norm": 1.8155136042295876,
      "language_loss": 0.85925233,
      "learning_rate": 4.5713794833662554e-08,
      "loss": 0.88090801,
      "num_input_tokens_seen": 167882095,
      "step": 7766,
      "time_per_iteration": 2.7367260456085205
    },
    {
      "auxiliary_loss_clip": 0.01108512,
      "auxiliary_loss_mlp": 0.01024708,
      "balance_loss_clip": 1.04216766,
      "balance_loss_mlp": 1.01641083,
      "epoch": 0.9339265315938196,
      "flos": 23221635482880.0,
      "grad_norm": 1.7532743855997066,
      "language_loss": 0.63255268,
      "learning_rate": 4.5548347260270236e-08,
      "loss": 0.65388489,
      "num_input_tokens_seen": 167901385,
      "step": 7767,
      "time_per_iteration": 2.7131073474884033
    },
    {
      "auxiliary_loss_clip": 0.01111945,
      "auxiliary_loss_mlp": 0.01026466,
      "balance_loss_clip": 1.04029441,
      "balance_loss_mlp": 1.01923871,
      "epoch": 0.9340467744844586,
      "flos": 22820261932800.0,
      "grad_norm": 1.8451788808554552,
      "language_loss": 0.69881153,
      "learning_rate": 4.538319617564012e-08,
      "loss": 0.72019565,
      "num_input_tokens_seen": 167920405,
      "step": 7768,
      "time_per_iteration": 2.7226076126098633
    },
    {
      "auxiliary_loss_clip": 0.01123362,
      "auxiliary_loss_mlp": 0.01027418,
      "balance_loss_clip": 1.04566836,
      "balance_loss_mlp": 1.01889503,
      "epoch": 0.9341670173750977,
      "flos": 23660428026240.0,
      "grad_norm": 1.967223131462489,
      "language_loss": 0.74342287,
      "learning_rate": 4.521834160482485e-08,
      "loss": 0.76493067,
      "num_input_tokens_seen": 167939145,
      "step": 7769,
      "time_per_iteration": 2.7803380489349365
    },
    {
      "auxiliary_loss_clip": 0.01148949,
      "auxiliary_loss_mlp": 0.01027208,
      "balance_loss_clip": 1.04861879,
      "balance_loss_mlp": 1.01948953,
      "epoch": 0.9342872602657368,
      "flos": 24824256595200.0,
      "grad_norm": 1.5766162284662524,
      "language_loss": 0.8242079,
      "learning_rate": 4.5053783572832846e-08,
      "loss": 0.84596944,
      "num_input_tokens_seen": 167959325,
      "step": 7770,
      "time_per_iteration": 2.648489475250244
    },
    {
      "auxiliary_loss_clip": 0.01118846,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 1.04372895,
      "balance_loss_mlp": 1.01799822,
      "epoch": 0.9344075031563759,
      "flos": 25771831332480.0,
      "grad_norm": 2.107377009818977,
      "language_loss": 0.764081,
      "learning_rate": 4.488952210462771e-08,
      "loss": 0.7855255,
      "num_input_tokens_seen": 167979530,
      "step": 7771,
      "time_per_iteration": 3.801746129989624
    },
    {
      "auxiliary_loss_clip": 0.01092122,
      "auxiliary_loss_mlp": 0.0102778,
      "balance_loss_clip": 1.04608107,
      "balance_loss_mlp": 1.02042174,
      "epoch": 0.9345277460470149,
      "flos": 25551303782400.0,
      "grad_norm": 2.081068577141348,
      "language_loss": 0.86231124,
      "learning_rate": 4.4725557225127495e-08,
      "loss": 0.88351029,
      "num_input_tokens_seen": 167997870,
      "step": 7772,
      "time_per_iteration": 2.8115713596343994
    },
    {
      "auxiliary_loss_clip": 0.01126321,
      "auxiliary_loss_mlp": 0.01036249,
      "balance_loss_clip": 1.04703426,
      "balance_loss_mlp": 1.02890015,
      "epoch": 0.9346479889376541,
      "flos": 34313112432000.0,
      "grad_norm": 1.7823665813790377,
      "language_loss": 0.79594725,
      "learning_rate": 4.456188895920565e-08,
      "loss": 0.81757295,
      "num_input_tokens_seen": 168019625,
      "step": 7773,
      "time_per_iteration": 2.861311435699463
    },
    {
      "auxiliary_loss_clip": 0.0115108,
      "auxiliary_loss_mlp": 0.01029002,
      "balance_loss_clip": 1.04819751,
      "balance_loss_mlp": 1.02022231,
      "epoch": 0.9347682318282932,
      "flos": 19093739581440.0,
      "grad_norm": 1.9461213774490114,
      "language_loss": 0.85327244,
      "learning_rate": 4.439851733169031e-08,
      "loss": 0.87507319,
      "num_input_tokens_seen": 168037415,
      "step": 7774,
      "time_per_iteration": 2.7319788932800293
    },
    {
      "auxiliary_loss_clip": 0.01154238,
      "auxiliary_loss_mlp": 0.01026914,
      "balance_loss_clip": 1.04950285,
      "balance_loss_mlp": 1.01926684,
      "epoch": 0.9348884747189322,
      "flos": 26249587153920.0,
      "grad_norm": 3.67128844751241,
      "language_loss": 0.7014395,
      "learning_rate": 4.4235442367365204e-08,
      "loss": 0.72325104,
      "num_input_tokens_seen": 168057725,
      "step": 7775,
      "time_per_iteration": 2.687026023864746
    },
    {
      "auxiliary_loss_clip": 0.01094207,
      "auxiliary_loss_mlp": 0.00762035,
      "balance_loss_clip": 1.0418601,
      "balance_loss_mlp": 1.00055957,
      "epoch": 0.9350087176095714,
      "flos": 18333080242560.0,
      "grad_norm": 1.9336657422748038,
      "language_loss": 0.79557544,
      "learning_rate": 4.4072664090968545e-08,
      "loss": 0.81413782,
      "num_input_tokens_seen": 168076110,
      "step": 7776,
      "time_per_iteration": 2.9034531116485596
    },
    {
      "auxiliary_loss_clip": 0.011588,
      "auxiliary_loss_mlp": 0.01028335,
      "balance_loss_clip": 1.04779208,
      "balance_loss_mlp": 1.02095556,
      "epoch": 0.9351289605002104,
      "flos": 19318253541120.0,
      "grad_norm": 1.858317202001108,
      "language_loss": 0.84860373,
      "learning_rate": 4.391018252719347e-08,
      "loss": 0.87047511,
      "num_input_tokens_seen": 168095905,
      "step": 7777,
      "time_per_iteration": 3.0308732986450195
    },
    {
      "auxiliary_loss_clip": 0.01141006,
      "auxiliary_loss_mlp": 0.0102929,
      "balance_loss_clip": 1.04639101,
      "balance_loss_mlp": 1.02171445,
      "epoch": 0.9352492033908495,
      "flos": 18799990156800.0,
      "grad_norm": 1.8335107436225409,
      "language_loss": 0.69317943,
      "learning_rate": 4.374799770068849e-08,
      "loss": 0.71488237,
      "num_input_tokens_seen": 168112580,
      "step": 7778,
      "time_per_iteration": 2.67267107963562
    },
    {
      "auxiliary_loss_clip": 0.01148714,
      "auxiliary_loss_mlp": 0.01025308,
      "balance_loss_clip": 1.04676521,
      "balance_loss_mlp": 1.01789951,
      "epoch": 0.9353694462814887,
      "flos": 29530134241920.0,
      "grad_norm": 1.9503437746890508,
      "language_loss": 0.75204313,
      "learning_rate": 4.358610963605658e-08,
      "loss": 0.77378333,
      "num_input_tokens_seen": 168133030,
      "step": 7779,
      "time_per_iteration": 2.700812578201294
    },
    {
      "auxiliary_loss_clip": 0.01129435,
      "auxiliary_loss_mlp": 0.01027146,
      "balance_loss_clip": 1.04562366,
      "balance_loss_mlp": 1.01900434,
      "epoch": 0.9354896891721277,
      "flos": 30665450390400.0,
      "grad_norm": 6.344425825935142,
      "language_loss": 0.68802774,
      "learning_rate": 4.342451835785677e-08,
      "loss": 0.70959359,
      "num_input_tokens_seen": 168153940,
      "step": 7780,
      "time_per_iteration": 2.780731678009033
    },
    {
      "auxiliary_loss_clip": 0.01111033,
      "auxiliary_loss_mlp": 0.01023199,
      "balance_loss_clip": 1.04420376,
      "balance_loss_mlp": 1.0160346,
      "epoch": 0.9356099320627668,
      "flos": 19463907191040.0,
      "grad_norm": 4.170084616624682,
      "language_loss": 0.75389373,
      "learning_rate": 4.3263223890601665e-08,
      "loss": 0.77523601,
      "num_input_tokens_seen": 168172650,
      "step": 7781,
      "time_per_iteration": 3.685087203979492
    },
    {
      "auxiliary_loss_clip": 0.01123995,
      "auxiliary_loss_mlp": 0.01026331,
      "balance_loss_clip": 1.04412508,
      "balance_loss_mlp": 1.01880848,
      "epoch": 0.9357301749534058,
      "flos": 19098156954240.0,
      "grad_norm": 2.947498563955271,
      "language_loss": 0.7954936,
      "learning_rate": 4.31022262587597e-08,
      "loss": 0.81699681,
      "num_input_tokens_seen": 168191325,
      "step": 7782,
      "time_per_iteration": 2.8237500190734863
    },
    {
      "auxiliary_loss_clip": 0.0116265,
      "auxiliary_loss_mlp": 0.01031794,
      "balance_loss_clip": 1.05177331,
      "balance_loss_mlp": 1.02299607,
      "epoch": 0.935850417844045,
      "flos": 23550361776000.0,
      "grad_norm": 1.927373261173024,
      "language_loss": 0.66273034,
      "learning_rate": 4.2941525486754225e-08,
      "loss": 0.68467474,
      "num_input_tokens_seen": 168211645,
      "step": 7783,
      "time_per_iteration": 2.6472232341766357
    },
    {
      "auxiliary_loss_clip": 0.01140066,
      "auxiliary_loss_mlp": 0.0102534,
      "balance_loss_clip": 1.04912269,
      "balance_loss_mlp": 1.0181576,
      "epoch": 0.935970660734684,
      "flos": 18588333265920.0,
      "grad_norm": 1.809819212556387,
      "language_loss": 0.79524207,
      "learning_rate": 4.278112159896286e-08,
      "loss": 0.8168962,
      "num_input_tokens_seen": 168229485,
      "step": 7784,
      "time_per_iteration": 2.6152210235595703
    },
    {
      "auxiliary_loss_clip": 0.01101275,
      "auxiliary_loss_mlp": 0.01028464,
      "balance_loss_clip": 1.04347849,
      "balance_loss_mlp": 1.0212338,
      "epoch": 0.9360909036253231,
      "flos": 20631255292800.0,
      "grad_norm": 3.571825562290519,
      "language_loss": 0.67862415,
      "learning_rate": 4.2621014619719896e-08,
      "loss": 0.69992155,
      "num_input_tokens_seen": 168247250,
      "step": 7785,
      "time_per_iteration": 3.8390462398529053
    },
    {
      "auxiliary_loss_clip": 0.0101649,
      "auxiliary_loss_mlp": 0.01001574,
      "balance_loss_clip": 1.00810599,
      "balance_loss_mlp": 1.0005722,
      "epoch": 0.9362111465159623,
      "flos": 61791421052160.0,
      "grad_norm": 0.7262549495755544,
      "language_loss": 0.58590794,
      "learning_rate": 4.246120457331215e-08,
      "loss": 0.60608852,
      "num_input_tokens_seen": 168309425,
      "step": 7786,
      "time_per_iteration": 3.422482490539551
    },
    {
      "auxiliary_loss_clip": 0.01149222,
      "auxiliary_loss_mlp": 0.01026555,
      "balance_loss_clip": 1.05244327,
      "balance_loss_mlp": 1.01923823,
      "epoch": 0.9363313894066013,
      "flos": 24170395368960.0,
      "grad_norm": 2.8194899634684467,
      "language_loss": 0.72337633,
      "learning_rate": 4.2301691483983325e-08,
      "loss": 0.74513412,
      "num_input_tokens_seen": 168329545,
      "step": 7787,
      "time_per_iteration": 2.9481911659240723
    },
    {
      "auxiliary_loss_clip": 0.01163367,
      "auxiliary_loss_mlp": 0.01029476,
      "balance_loss_clip": 1.04914057,
      "balance_loss_mlp": 1.02147746,
      "epoch": 0.9364516322972404,
      "flos": 20120354196480.0,
      "grad_norm": 2.1040902902792493,
      "language_loss": 0.75891066,
      "learning_rate": 4.214247537593163e-08,
      "loss": 0.78083909,
      "num_input_tokens_seen": 168348795,
      "step": 7788,
      "time_per_iteration": 2.6878175735473633
    },
    {
      "auxiliary_loss_clip": 0.01165483,
      "auxiliary_loss_mlp": 0.01034266,
      "balance_loss_clip": 1.05208588,
      "balance_loss_mlp": 1.02650869,
      "epoch": 0.9365718751878795,
      "flos": 20703758895360.0,
      "grad_norm": 2.9424386644659033,
      "language_loss": 0.80369979,
      "learning_rate": 4.1983556273309293e-08,
      "loss": 0.82569724,
      "num_input_tokens_seen": 168367545,
      "step": 7789,
      "time_per_iteration": 2.643639326095581
    },
    {
      "auxiliary_loss_clip": 0.01133857,
      "auxiliary_loss_mlp": 0.01023576,
      "balance_loss_clip": 1.04964268,
      "balance_loss_mlp": 1.01496935,
      "epoch": 0.9366921180785186,
      "flos": 18655270260480.0,
      "grad_norm": 2.510209714281658,
      "language_loss": 0.69528437,
      "learning_rate": 4.182493420022526e-08,
      "loss": 0.71685874,
      "num_input_tokens_seen": 168383215,
      "step": 7790,
      "time_per_iteration": 2.795485496520996
    },
    {
      "auxiliary_loss_clip": 0.01154771,
      "auxiliary_loss_mlp": 0.01026688,
      "balance_loss_clip": 1.04842293,
      "balance_loss_mlp": 1.0194757,
      "epoch": 0.9368123609691577,
      "flos": 25774955815680.0,
      "grad_norm": 1.690445326866869,
      "language_loss": 0.78625977,
      "learning_rate": 4.166660918074139e-08,
      "loss": 0.80807441,
      "num_input_tokens_seen": 168403120,
      "step": 7791,
      "time_per_iteration": 2.6871907711029053
    },
    {
      "auxiliary_loss_clip": 0.01155099,
      "auxiliary_loss_mlp": 0.00761921,
      "balance_loss_clip": 1.05002856,
      "balance_loss_mlp": 1.00050592,
      "epoch": 0.9369326038597968,
      "flos": 25553386771200.0,
      "grad_norm": 1.50658755509366,
      "language_loss": 0.73493052,
      "learning_rate": 4.15085812388758e-08,
      "loss": 0.7541008,
      "num_input_tokens_seen": 168425340,
      "step": 7792,
      "time_per_iteration": 2.67838978767395
    },
    {
      "auxiliary_loss_clip": 0.01143324,
      "auxiliary_loss_mlp": 0.01032034,
      "balance_loss_clip": 1.0487982,
      "balance_loss_mlp": 1.02402866,
      "epoch": 0.9370528467504359,
      "flos": 23220019370880.0,
      "grad_norm": 1.7005939672557218,
      "language_loss": 0.78772867,
      "learning_rate": 4.135085039860153e-08,
      "loss": 0.80948228,
      "num_input_tokens_seen": 168444740,
      "step": 7793,
      "time_per_iteration": 2.7194838523864746
    },
    {
      "auxiliary_loss_clip": 0.01106295,
      "auxiliary_loss_mlp": 0.01025366,
      "balance_loss_clip": 1.04157662,
      "balance_loss_mlp": 1.01682484,
      "epoch": 0.9371730896410749,
      "flos": 24967468120320.0,
      "grad_norm": 2.3278209932066964,
      "language_loss": 0.78543514,
      "learning_rate": 4.1193416683845906e-08,
      "loss": 0.80675179,
      "num_input_tokens_seen": 168463670,
      "step": 7794,
      "time_per_iteration": 2.814384698867798
    },
    {
      "auxiliary_loss_clip": 0.01155528,
      "auxiliary_loss_mlp": 0.01023319,
      "balance_loss_clip": 1.05244291,
      "balance_loss_mlp": 1.01650953,
      "epoch": 0.9372933325317141,
      "flos": 15553091134080.0,
      "grad_norm": 3.0324889052923867,
      "language_loss": 0.83360779,
      "learning_rate": 4.103628011849136e-08,
      "loss": 0.85539627,
      "num_input_tokens_seen": 168479030,
      "step": 7795,
      "time_per_iteration": 2.7206475734710693
    },
    {
      "auxiliary_loss_clip": 0.01154965,
      "auxiliary_loss_mlp": 0.01029504,
      "balance_loss_clip": 1.0477761,
      "balance_loss_mlp": 1.02145696,
      "epoch": 0.9374135754223532,
      "flos": 21871861182720.0,
      "grad_norm": 2.9785548420042107,
      "language_loss": 0.76097018,
      "learning_rate": 4.0879440726375506e-08,
      "loss": 0.78281492,
      "num_input_tokens_seen": 168496815,
      "step": 7796,
      "time_per_iteration": 3.7350540161132812
    },
    {
      "auxiliary_loss_clip": 0.01149177,
      "auxiliary_loss_mlp": 0.0102748,
      "balance_loss_clip": 1.04806042,
      "balance_loss_mlp": 1.01856351,
      "epoch": 0.9375338183129922,
      "flos": 22631048064000.0,
      "grad_norm": 2.819074954085271,
      "language_loss": 0.55705082,
      "learning_rate": 4.0722898531291074e-08,
      "loss": 0.57881737,
      "num_input_tokens_seen": 168514055,
      "step": 7797,
      "time_per_iteration": 2.748931884765625
    },
    {
      "auxiliary_loss_clip": 0.01153504,
      "auxiliary_loss_mlp": 0.00762229,
      "balance_loss_clip": 1.04823196,
      "balance_loss_mlp": 1.00048804,
      "epoch": 0.9376540612036314,
      "flos": 26104292640000.0,
      "grad_norm": 2.354752318159447,
      "language_loss": 0.77456856,
      "learning_rate": 4.0566653556985295e-08,
      "loss": 0.79372585,
      "num_input_tokens_seen": 168534600,
      "step": 7798,
      "time_per_iteration": 2.7798213958740234
    },
    {
      "auxiliary_loss_clip": 0.01165207,
      "auxiliary_loss_mlp": 0.01029641,
      "balance_loss_clip": 1.05225372,
      "balance_loss_mlp": 1.02055097,
      "epoch": 0.9377743040942704,
      "flos": 19717580016000.0,
      "grad_norm": 2.1020205102781713,
      "language_loss": 0.81718659,
      "learning_rate": 4.0410705827159886e-08,
      "loss": 0.83913505,
      "num_input_tokens_seen": 168551895,
      "step": 7799,
      "time_per_iteration": 2.8759870529174805
    },
    {
      "auxiliary_loss_clip": 0.0116371,
      "auxiliary_loss_mlp": 0.0102535,
      "balance_loss_clip": 1.05076051,
      "balance_loss_mlp": 1.01760173,
      "epoch": 0.9378945469849095,
      "flos": 15267530010240.0,
      "grad_norm": 2.209933444592369,
      "language_loss": 0.70978057,
      "learning_rate": 4.0255055365472356e-08,
      "loss": 0.73167121,
      "num_input_tokens_seen": 168569990,
      "step": 7800,
      "time_per_iteration": 2.711812734603882
    },
    {
      "auxiliary_loss_clip": 0.01145427,
      "auxiliary_loss_mlp": 0.01024482,
      "balance_loss_clip": 1.04514563,
      "balance_loss_mlp": 1.01725817,
      "epoch": 0.9380147898755486,
      "flos": 20591394174720.0,
      "grad_norm": 2.383534555248821,
      "language_loss": 0.74948299,
      "learning_rate": 4.009970219553471e-08,
      "loss": 0.77118218,
      "num_input_tokens_seen": 168586940,
      "step": 7801,
      "time_per_iteration": 2.7312896251678467
    },
    {
      "auxiliary_loss_clip": 0.01119464,
      "auxiliary_loss_mlp": 0.01032507,
      "balance_loss_clip": 1.04660273,
      "balance_loss_mlp": 1.02423429,
      "epoch": 0.9381350327661877,
      "flos": 26281116316800.0,
      "grad_norm": 2.536421465758481,
      "language_loss": 0.77151299,
      "learning_rate": 3.99446463409141e-08,
      "loss": 0.79303271,
      "num_input_tokens_seen": 168604795,
      "step": 7802,
      "time_per_iteration": 2.8513379096984863
    },
    {
      "auxiliary_loss_clip": 0.01142205,
      "auxiliary_loss_mlp": 0.01028462,
      "balance_loss_clip": 1.04638958,
      "balance_loss_mlp": 1.02055836,
      "epoch": 0.9382552756568268,
      "flos": 23586344225280.0,
      "grad_norm": 2.4852200243251588,
      "language_loss": 0.69428754,
      "learning_rate": 3.978988782513215e-08,
      "loss": 0.71599424,
      "num_input_tokens_seen": 168622290,
      "step": 7803,
      "time_per_iteration": 2.7553088665008545
    },
    {
      "auxiliary_loss_clip": 0.01127567,
      "auxiliary_loss_mlp": 0.01026519,
      "balance_loss_clip": 1.04372525,
      "balance_loss_mlp": 1.01843965,
      "epoch": 0.9383755185474659,
      "flos": 28438809275520.0,
      "grad_norm": 1.9217247506571695,
      "language_loss": 0.7656337,
      "learning_rate": 3.963542667166586e-08,
      "loss": 0.78717458,
      "num_input_tokens_seen": 168642395,
      "step": 7804,
      "time_per_iteration": 2.8788247108459473
    },
    {
      "auxiliary_loss_clip": 0.01173112,
      "auxiliary_loss_mlp": 0.01030901,
      "balance_loss_clip": 1.05261707,
      "balance_loss_mlp": 1.02352154,
      "epoch": 0.938495761438105,
      "flos": 20449583280000.0,
      "grad_norm": 2.678302156858265,
      "language_loss": 0.68628836,
      "learning_rate": 3.9481262903946486e-08,
      "loss": 0.70832849,
      "num_input_tokens_seen": 168661840,
      "step": 7805,
      "time_per_iteration": 2.630807638168335
    },
    {
      "auxiliary_loss_clip": 0.01061284,
      "auxiliary_loss_mlp": 0.01001649,
      "balance_loss_clip": 1.00774181,
      "balance_loss_mlp": 1.00072527,
      "epoch": 0.938616004328744,
      "flos": 69302711658240.0,
      "grad_norm": 0.7654876932988505,
      "language_loss": 0.54415929,
      "learning_rate": 3.932739654536066e-08,
      "loss": 0.56478858,
      "num_input_tokens_seen": 168724540,
      "step": 7806,
      "time_per_iteration": 4.189447402954102
    },
    {
      "auxiliary_loss_clip": 0.01139383,
      "auxiliary_loss_mlp": 0.01028733,
      "balance_loss_clip": 1.04651344,
      "balance_loss_mlp": 1.02104664,
      "epoch": 0.9387362472193832,
      "flos": 18911636605440.0,
      "grad_norm": 2.555158103863432,
      "language_loss": 0.74556851,
      "learning_rate": 3.917382761925014e-08,
      "loss": 0.76724964,
      "num_input_tokens_seen": 168740375,
      "step": 7807,
      "time_per_iteration": 2.820596694946289
    },
    {
      "auxiliary_loss_clip": 0.01173092,
      "auxiliary_loss_mlp": 0.00761701,
      "balance_loss_clip": 1.05208147,
      "balance_loss_mlp": 1.0004487,
      "epoch": 0.9388564901100223,
      "flos": 26501967089280.0,
      "grad_norm": 1.7175257897204859,
      "language_loss": 0.79414672,
      "learning_rate": 3.9020556148910754e-08,
      "loss": 0.81349462,
      "num_input_tokens_seen": 168759730,
      "step": 7808,
      "time_per_iteration": 2.6762428283691406
    },
    {
      "auxiliary_loss_clip": 0.0102672,
      "auxiliary_loss_mlp": 0.01002472,
      "balance_loss_clip": 1.00877333,
      "balance_loss_mlp": 1.00164926,
      "epoch": 0.9389767330006613,
      "flos": 58941083157120.0,
      "grad_norm": 0.7036557539093109,
      "language_loss": 0.56577408,
      "learning_rate": 3.8867582157593895e-08,
      "loss": 0.58606601,
      "num_input_tokens_seen": 168813935,
      "step": 7809,
      "time_per_iteration": 3.1704177856445312
    },
    {
      "auxiliary_loss_clip": 0.01120757,
      "auxiliary_loss_mlp": 0.01028436,
      "balance_loss_clip": 1.04608512,
      "balance_loss_mlp": 1.02052653,
      "epoch": 0.9390969758913005,
      "flos": 31102554994560.0,
      "grad_norm": 1.6915942318742454,
      "language_loss": 0.76584691,
      "learning_rate": 3.871490566850544e-08,
      "loss": 0.78733885,
      "num_input_tokens_seen": 168838145,
      "step": 7810,
      "time_per_iteration": 2.982133388519287
    },
    {
      "auxiliary_loss_clip": 0.01144391,
      "auxiliary_loss_mlp": 0.01024722,
      "balance_loss_clip": 1.04846668,
      "balance_loss_mlp": 1.0173732,
      "epoch": 0.9392172187819395,
      "flos": 22419391173120.0,
      "grad_norm": 3.877455588144004,
      "language_loss": 0.70745921,
      "learning_rate": 3.856252670480642e-08,
      "loss": 0.7291503,
      "num_input_tokens_seen": 168856805,
      "step": 7811,
      "time_per_iteration": 3.6890196800231934
    },
    {
      "auxiliary_loss_clip": 0.01144501,
      "auxiliary_loss_mlp": 0.01026688,
      "balance_loss_clip": 1.04672575,
      "balance_loss_mlp": 1.01933241,
      "epoch": 0.9393374616725786,
      "flos": 19719483436800.0,
      "grad_norm": 2.077587341665534,
      "language_loss": 0.8147049,
      "learning_rate": 3.841044528961279e-08,
      "loss": 0.83641678,
      "num_input_tokens_seen": 168874600,
      "step": 7812,
      "time_per_iteration": 2.807209014892578
    },
    {
      "auxiliary_loss_clip": 0.01148033,
      "auxiliary_loss_mlp": 0.01025775,
      "balance_loss_clip": 1.04697788,
      "balance_loss_mlp": 1.01761484,
      "epoch": 0.9394577045632178,
      "flos": 24170215800960.0,
      "grad_norm": 1.8567952107580736,
      "language_loss": 0.79244614,
      "learning_rate": 3.825866144599477e-08,
      "loss": 0.81418419,
      "num_input_tokens_seen": 168893655,
      "step": 7813,
      "time_per_iteration": 2.7720260620117188
    },
    {
      "auxiliary_loss_clip": 0.01094624,
      "auxiliary_loss_mlp": 0.01026224,
      "balance_loss_clip": 1.03965068,
      "balance_loss_mlp": 1.01836765,
      "epoch": 0.9395779474538568,
      "flos": 19023929498880.0,
      "grad_norm": 2.6507465915932316,
      "language_loss": 0.75876284,
      "learning_rate": 3.8107175196978145e-08,
      "loss": 0.77997136,
      "num_input_tokens_seen": 168909960,
      "step": 7814,
      "time_per_iteration": 2.790031671524048
    },
    {
      "auxiliary_loss_clip": 0.01131276,
      "auxiliary_loss_mlp": 0.00761468,
      "balance_loss_clip": 1.05012786,
      "balance_loss_mlp": 1.00054955,
      "epoch": 0.9396981903444959,
      "flos": 14319129260160.0,
      "grad_norm": 2.099233648581207,
      "language_loss": 0.7684406,
      "learning_rate": 3.7955986565542996e-08,
      "loss": 0.78736806,
      "num_input_tokens_seen": 168928040,
      "step": 7815,
      "time_per_iteration": 2.8148953914642334
    },
    {
      "auxiliary_loss_clip": 0.01124247,
      "auxiliary_loss_mlp": 0.01026689,
      "balance_loss_clip": 1.04670763,
      "balance_loss_mlp": 1.01886296,
      "epoch": 0.9398184332351349,
      "flos": 34787564202240.0,
      "grad_norm": 2.2692067471403004,
      "language_loss": 0.68596971,
      "learning_rate": 3.780509557462497e-08,
      "loss": 0.70747906,
      "num_input_tokens_seen": 168948240,
      "step": 7816,
      "time_per_iteration": 2.9019551277160645
    },
    {
      "auxiliary_loss_clip": 0.011121,
      "auxiliary_loss_mlp": 0.01029537,
      "balance_loss_clip": 1.04511058,
      "balance_loss_mlp": 1.02155614,
      "epoch": 0.9399386761257741,
      "flos": 25372253462400.0,
      "grad_norm": 1.8586751410290876,
      "language_loss": 0.7583313,
      "learning_rate": 3.765450224711375e-08,
      "loss": 0.77974772,
      "num_input_tokens_seen": 168968745,
      "step": 7817,
      "time_per_iteration": 2.8721938133239746
    },
    {
      "auxiliary_loss_clip": 0.01117775,
      "auxiliary_loss_mlp": 0.01027172,
      "balance_loss_clip": 1.04335821,
      "balance_loss_mlp": 1.01985216,
      "epoch": 0.9400589190164131,
      "flos": 27304965584640.0,
      "grad_norm": 1.8361560822255254,
      "language_loss": 0.80021775,
      "learning_rate": 3.750420660585396e-08,
      "loss": 0.82166719,
      "num_input_tokens_seen": 168990685,
      "step": 7818,
      "time_per_iteration": 2.7548935413360596
    },
    {
      "auxiliary_loss_clip": 0.01163198,
      "auxiliary_loss_mlp": 0.01023703,
      "balance_loss_clip": 1.05143344,
      "balance_loss_mlp": 1.01621068,
      "epoch": 0.9401791619070522,
      "flos": 23399859790080.0,
      "grad_norm": 11.545223616459992,
      "language_loss": 0.79759747,
      "learning_rate": 3.735420867364603e-08,
      "loss": 0.81946647,
      "num_input_tokens_seen": 169011665,
      "step": 7819,
      "time_per_iteration": 2.7408576011657715
    },
    {
      "auxiliary_loss_clip": 0.01102956,
      "auxiliary_loss_mlp": 0.01022621,
      "balance_loss_clip": 1.04037094,
      "balance_loss_mlp": 1.01552224,
      "epoch": 0.9402994047976914,
      "flos": 35881403120640.0,
      "grad_norm": 1.7107192969825833,
      "language_loss": 0.61686528,
      "learning_rate": 3.7204508473244186e-08,
      "loss": 0.63812101,
      "num_input_tokens_seen": 169035290,
      "step": 7820,
      "time_per_iteration": 3.052544116973877
    },
    {
      "auxiliary_loss_clip": 0.01149843,
      "auxiliary_loss_mlp": 0.01026119,
      "balance_loss_clip": 1.04710698,
      "balance_loss_mlp": 1.01839995,
      "epoch": 0.9404196476883304,
      "flos": 22236821320320.0,
      "grad_norm": 1.7734684853083023,
      "language_loss": 0.69332385,
      "learning_rate": 3.7055106027357395e-08,
      "loss": 0.71508354,
      "num_input_tokens_seen": 169055155,
      "step": 7821,
      "time_per_iteration": 2.7370734214782715
    },
    {
      "auxiliary_loss_clip": 0.01114018,
      "auxiliary_loss_mlp": 0.01026178,
      "balance_loss_clip": 1.04164875,
      "balance_loss_mlp": 1.01899552,
      "epoch": 0.9405398905789695,
      "flos": 18915802583040.0,
      "grad_norm": 3.451593104859411,
      "language_loss": 0.71757579,
      "learning_rate": 3.690600135865063e-08,
      "loss": 0.73897779,
      "num_input_tokens_seen": 169072080,
      "step": 7822,
      "time_per_iteration": 3.758070945739746
    },
    {
      "auxiliary_loss_clip": 0.0102401,
      "auxiliary_loss_mlp": 0.0100056,
      "balance_loss_clip": 1.0085609,
      "balance_loss_mlp": 0.99963611,
      "epoch": 0.9406601334696086,
      "flos": 70274130048000.0,
      "grad_norm": 0.7852769349056675,
      "language_loss": 0.57989275,
      "learning_rate": 3.675719448974246e-08,
      "loss": 0.60013843,
      "num_input_tokens_seen": 169137170,
      "step": 7823,
      "time_per_iteration": 3.3221983909606934
    },
    {
      "auxiliary_loss_clip": 0.01095528,
      "auxiliary_loss_mlp": 0.01028718,
      "balance_loss_clip": 1.04047632,
      "balance_loss_mlp": 1.02104104,
      "epoch": 0.9407803763602477,
      "flos": 22165071903360.0,
      "grad_norm": 1.889545907135428,
      "language_loss": 0.60006207,
      "learning_rate": 3.6608685443207054e-08,
      "loss": 0.62130451,
      "num_input_tokens_seen": 169156320,
      "step": 7824,
      "time_per_iteration": 2.8433403968811035
    },
    {
      "auxiliary_loss_clip": 0.01140775,
      "auxiliary_loss_mlp": 0.01029618,
      "balance_loss_clip": 1.04715478,
      "balance_loss_mlp": 1.02221239,
      "epoch": 0.9409006192508867,
      "flos": 18879496911360.0,
      "grad_norm": 2.6743856492893068,
      "language_loss": 0.6681807,
      "learning_rate": 3.646047424157306e-08,
      "loss": 0.6898846,
      "num_input_tokens_seen": 169173295,
      "step": 7825,
      "time_per_iteration": 2.7445638179779053
    },
    {
      "auxiliary_loss_clip": 0.01148452,
      "auxiliary_loss_mlp": 0.01031244,
      "balance_loss_clip": 1.04838729,
      "balance_loss_mlp": 1.02232122,
      "epoch": 0.9410208621415259,
      "flos": 23368258800000.0,
      "grad_norm": 3.086869978144927,
      "language_loss": 0.68965757,
      "learning_rate": 3.631256090732382e-08,
      "loss": 0.71145451,
      "num_input_tokens_seen": 169193755,
      "step": 7826,
      "time_per_iteration": 2.6921234130859375
    },
    {
      "auxiliary_loss_clip": 0.01156406,
      "auxiliary_loss_mlp": 0.01030116,
      "balance_loss_clip": 1.05136299,
      "balance_loss_mlp": 1.02319002,
      "epoch": 0.941141105032165,
      "flos": 22742227635840.0,
      "grad_norm": 2.153138533169675,
      "language_loss": 0.82925034,
      "learning_rate": 3.6164945462897833e-08,
      "loss": 0.85111558,
      "num_input_tokens_seen": 169213045,
      "step": 7827,
      "time_per_iteration": 2.7304675579071045
    },
    {
      "auxiliary_loss_clip": 0.011416,
      "auxiliary_loss_mlp": 0.00761363,
      "balance_loss_clip": 1.05055285,
      "balance_loss_mlp": 1.00046015,
      "epoch": 0.941261347922804,
      "flos": 20704908130560.0,
      "grad_norm": 2.480435036991579,
      "language_loss": 0.76029575,
      "learning_rate": 3.6017627930687856e-08,
      "loss": 0.77932543,
      "num_input_tokens_seen": 169232870,
      "step": 7828,
      "time_per_iteration": 2.680967330932617
    },
    {
      "auxiliary_loss_clip": 0.01127494,
      "auxiliary_loss_mlp": 0.01030676,
      "balance_loss_clip": 1.04413331,
      "balance_loss_mlp": 1.02319872,
      "epoch": 0.9413815908134432,
      "flos": 19421998997760.0,
      "grad_norm": 2.2095279754779735,
      "language_loss": 0.7698859,
      "learning_rate": 3.587060833304267e-08,
      "loss": 0.79146761,
      "num_input_tokens_seen": 169251060,
      "step": 7829,
      "time_per_iteration": 2.7471344470977783
    },
    {
      "auxiliary_loss_clip": 0.01136043,
      "auxiliary_loss_mlp": 0.01029498,
      "balance_loss_clip": 1.04688203,
      "balance_loss_mlp": 1.02142763,
      "epoch": 0.9415018337040822,
      "flos": 17493452853120.0,
      "grad_norm": 2.0593060548866595,
      "language_loss": 0.6434353,
      "learning_rate": 3.5723886692264225e-08,
      "loss": 0.66509068,
      "num_input_tokens_seen": 169268600,
      "step": 7830,
      "time_per_iteration": 2.6824166774749756
    },
    {
      "auxiliary_loss_clip": 0.0114575,
      "auxiliary_loss_mlp": 0.0102528,
      "balance_loss_clip": 1.04744422,
      "balance_loss_mlp": 1.01764464,
      "epoch": 0.9416220765947213,
      "flos": 31831613343360.0,
      "grad_norm": 2.126611013719133,
      "language_loss": 0.61926675,
      "learning_rate": 3.557746303061071e-08,
      "loss": 0.64097703,
      "num_input_tokens_seen": 169290355,
      "step": 7831,
      "time_per_iteration": 2.8107705116271973
    },
    {
      "auxiliary_loss_clip": 0.01125804,
      "auxiliary_loss_mlp": 0.01026001,
      "balance_loss_clip": 1.04661858,
      "balance_loss_mlp": 1.01833308,
      "epoch": 0.9417423194853605,
      "flos": 23511973115520.0,
      "grad_norm": 1.7848207500123476,
      "language_loss": 0.72580767,
      "learning_rate": 3.543133737029391e-08,
      "loss": 0.74732572,
      "num_input_tokens_seen": 169310865,
      "step": 7832,
      "time_per_iteration": 3.7547755241394043
    },
    {
      "auxiliary_loss_clip": 0.01109676,
      "auxiliary_loss_mlp": 0.01029053,
      "balance_loss_clip": 1.04435396,
      "balance_loss_mlp": 1.02061892,
      "epoch": 0.9418625623759995,
      "flos": 23915106432000.0,
      "grad_norm": 2.2264216363133182,
      "language_loss": 0.68843424,
      "learning_rate": 3.5285509733481214e-08,
      "loss": 0.70982158,
      "num_input_tokens_seen": 169330590,
      "step": 7833,
      "time_per_iteration": 2.79248046875
    },
    {
      "auxiliary_loss_clip": 0.01104206,
      "auxiliary_loss_mlp": 0.01022274,
      "balance_loss_clip": 1.04626811,
      "balance_loss_mlp": 1.01419806,
      "epoch": 0.9419828052666386,
      "flos": 18076965292800.0,
      "grad_norm": 1.798012975495053,
      "language_loss": 0.766114,
      "learning_rate": 3.513998014229469e-08,
      "loss": 0.78737879,
      "num_input_tokens_seen": 169349540,
      "step": 7834,
      "time_per_iteration": 2.773623466491699
    },
    {
      "auxiliary_loss_clip": 0.01106666,
      "auxiliary_loss_mlp": 0.01028232,
      "balance_loss_clip": 1.04282939,
      "balance_loss_mlp": 1.02048898,
      "epoch": 0.9421030481572777,
      "flos": 17712328377600.0,
      "grad_norm": 2.593197775086667,
      "language_loss": 0.86569047,
      "learning_rate": 3.499474861881069e-08,
      "loss": 0.88703942,
      "num_input_tokens_seen": 169366765,
      "step": 7835,
      "time_per_iteration": 2.726022243499756
    },
    {
      "auxiliary_loss_clip": 0.01173637,
      "auxiliary_loss_mlp": 0.01028666,
      "balance_loss_clip": 1.0517,
      "balance_loss_mlp": 1.02081037,
      "epoch": 0.9422232910479168,
      "flos": 20194114775040.0,
      "grad_norm": 1.9555080054303462,
      "language_loss": 0.67875123,
      "learning_rate": 3.4849815185061136e-08,
      "loss": 0.70077419,
      "num_input_tokens_seen": 169386655,
      "step": 7836,
      "time_per_iteration": 2.6923842430114746
    },
    {
      "auxiliary_loss_clip": 0.01154908,
      "auxiliary_loss_mlp": 0.01026521,
      "balance_loss_clip": 1.04726696,
      "balance_loss_mlp": 1.01992846,
      "epoch": 0.9423435339385559,
      "flos": 18442571875200.0,
      "grad_norm": 2.678327371793994,
      "language_loss": 0.76070982,
      "learning_rate": 3.470517986303223e-08,
      "loss": 0.78252411,
      "num_input_tokens_seen": 169405640,
      "step": 7837,
      "time_per_iteration": 3.515756845474243
    },
    {
      "auxiliary_loss_clip": 0.01109166,
      "auxiliary_loss_mlp": 0.010339,
      "balance_loss_clip": 1.04603565,
      "balance_loss_mlp": 1.02612209,
      "epoch": 0.942463776829195,
      "flos": 20080636732800.0,
      "grad_norm": 1.7072128054249456,
      "language_loss": 0.7936846,
      "learning_rate": 3.4560842674664856e-08,
      "loss": 0.81511527,
      "num_input_tokens_seen": 169424155,
      "step": 7838,
      "time_per_iteration": 2.7865819931030273
    },
    {
      "auxiliary_loss_clip": 0.01129473,
      "auxiliary_loss_mlp": 0.01028643,
      "balance_loss_clip": 1.04485416,
      "balance_loss_mlp": 1.02106094,
      "epoch": 0.9425840197198341,
      "flos": 22636255536000.0,
      "grad_norm": 1.7756165509934174,
      "language_loss": 0.75632024,
      "learning_rate": 3.441680364185506e-08,
      "loss": 0.77790141,
      "num_input_tokens_seen": 169444025,
      "step": 7839,
      "time_per_iteration": 2.7366037368774414
    },
    {
      "auxiliary_loss_clip": 0.01154301,
      "auxiliary_loss_mlp": 0.01028614,
      "balance_loss_clip": 1.04719687,
      "balance_loss_mlp": 1.02092457,
      "epoch": 0.9427042626104731,
      "flos": 19937892084480.0,
      "grad_norm": 2.556016980016409,
      "language_loss": 0.74684966,
      "learning_rate": 3.427306278645314e-08,
      "loss": 0.76867878,
      "num_input_tokens_seen": 169462480,
      "step": 7840,
      "time_per_iteration": 2.631070137023926
    },
    {
      "auxiliary_loss_clip": 0.0113977,
      "auxiliary_loss_mlp": 0.01026596,
      "balance_loss_clip": 1.04771698,
      "balance_loss_mlp": 1.01849914,
      "epoch": 0.9428245055011123,
      "flos": 22856998567680.0,
      "grad_norm": 1.8110074438666333,
      "language_loss": 0.73007679,
      "learning_rate": 3.4129620130264767e-08,
      "loss": 0.75174046,
      "num_input_tokens_seen": 169480840,
      "step": 7841,
      "time_per_iteration": 2.7592968940734863
    },
    {
      "auxiliary_loss_clip": 0.01145616,
      "auxiliary_loss_mlp": 0.01029769,
      "balance_loss_clip": 1.05004191,
      "balance_loss_mlp": 1.02232409,
      "epoch": 0.9429447483917514,
      "flos": 20951757371520.0,
      "grad_norm": 3.31650730817934,
      "language_loss": 0.78124869,
      "learning_rate": 3.398647569505009e-08,
      "loss": 0.8030026,
      "num_input_tokens_seen": 169498265,
      "step": 7842,
      "time_per_iteration": 2.6636462211608887
    },
    {
      "auxiliary_loss_clip": 0.01148959,
      "auxiliary_loss_mlp": 0.01027957,
      "balance_loss_clip": 1.04747486,
      "balance_loss_mlp": 1.0202055,
      "epoch": 0.9430649912823904,
      "flos": 18843658116480.0,
      "grad_norm": 2.586872968900566,
      "language_loss": 0.75070453,
      "learning_rate": 3.384362950252373e-08,
      "loss": 0.77247369,
      "num_input_tokens_seen": 169515235,
      "step": 7843,
      "time_per_iteration": 2.6866037845611572
    },
    {
      "auxiliary_loss_clip": 0.01157193,
      "auxiliary_loss_mlp": 0.01029073,
      "balance_loss_clip": 1.05034983,
      "balance_loss_mlp": 1.02119327,
      "epoch": 0.9431852341730296,
      "flos": 32556038837760.0,
      "grad_norm": 4.162850986440326,
      "language_loss": 0.57274455,
      "learning_rate": 3.3701081574355473e-08,
      "loss": 0.59460723,
      "num_input_tokens_seen": 169537195,
      "step": 7844,
      "time_per_iteration": 2.791534423828125
    },
    {
      "auxiliary_loss_clip": 0.01038568,
      "auxiliary_loss_mlp": 0.01001909,
      "balance_loss_clip": 1.00802827,
      "balance_loss_mlp": 1.00103879,
      "epoch": 0.9433054770636686,
      "flos": 66904490252160.0,
      "grad_norm": 0.6404345770453078,
      "language_loss": 0.51646137,
      "learning_rate": 3.3558831932169796e-08,
      "loss": 0.53686613,
      "num_input_tokens_seen": 169605865,
      "step": 7845,
      "time_per_iteration": 3.3625874519348145
    },
    {
      "auxiliary_loss_clip": 0.01151921,
      "auxiliary_loss_mlp": 0.00762247,
      "balance_loss_clip": 1.04771447,
      "balance_loss_mlp": 1.00049853,
      "epoch": 0.9434257199543077,
      "flos": 26140346916480.0,
      "grad_norm": 2.1030055802869922,
      "language_loss": 0.88769406,
      "learning_rate": 3.341688059754588e-08,
      "loss": 0.90683579,
      "num_input_tokens_seen": 169621520,
      "step": 7846,
      "time_per_iteration": 2.711029529571533
    },
    {
      "auxiliary_loss_clip": 0.01161462,
      "auxiliary_loss_mlp": 0.01029302,
      "balance_loss_clip": 1.0504508,
      "balance_loss_mlp": 1.0218575,
      "epoch": 0.9435459628449467,
      "flos": 25003486483200.0,
      "grad_norm": 2.650030522197129,
      "language_loss": 0.78059161,
      "learning_rate": 3.327522759201762e-08,
      "loss": 0.80249918,
      "num_input_tokens_seen": 169641390,
      "step": 7847,
      "time_per_iteration": 3.6530466079711914
    },
    {
      "auxiliary_loss_clip": 0.01172648,
      "auxiliary_loss_mlp": 0.01027963,
      "balance_loss_clip": 1.05096257,
      "balance_loss_mlp": 1.02000558,
      "epoch": 0.9436662057355859,
      "flos": 22163240309760.0,
      "grad_norm": 2.585346247617591,
      "language_loss": 0.66649145,
      "learning_rate": 3.313387293707359e-08,
      "loss": 0.68849754,
      "num_input_tokens_seen": 169660095,
      "step": 7848,
      "time_per_iteration": 2.68231463432312
    },
    {
      "auxiliary_loss_clip": 0.01117698,
      "auxiliary_loss_mlp": 0.01030548,
      "balance_loss_clip": 1.04497528,
      "balance_loss_mlp": 1.0221498,
      "epoch": 0.943786448626225,
      "flos": 20118522602880.0,
      "grad_norm": 1.973314410827076,
      "language_loss": 0.68328297,
      "learning_rate": 3.29928166541571e-08,
      "loss": 0.70476544,
      "num_input_tokens_seen": 169679050,
      "step": 7849,
      "time_per_iteration": 2.743269920349121
    },
    {
      "auxiliary_loss_clip": 0.01143383,
      "auxiliary_loss_mlp": 0.01024071,
      "balance_loss_clip": 1.04735637,
      "balance_loss_mlp": 1.01660228,
      "epoch": 0.943906691516864,
      "flos": 22090808534400.0,
      "grad_norm": 2.897322034957466,
      "language_loss": 0.80374151,
      "learning_rate": 3.2852058764666346e-08,
      "loss": 0.82541609,
      "num_input_tokens_seen": 169698150,
      "step": 7850,
      "time_per_iteration": 2.7285921573638916
    },
    {
      "auxiliary_loss_clip": 0.01158119,
      "auxiliary_loss_mlp": 0.01031161,
      "balance_loss_clip": 1.04885626,
      "balance_loss_mlp": 1.02320349,
      "epoch": 0.9440269344075032,
      "flos": 35298501212160.0,
      "grad_norm": 2.0257216269784806,
      "language_loss": 0.68656403,
      "learning_rate": 3.2711599289954264e-08,
      "loss": 0.70845681,
      "num_input_tokens_seen": 169722185,
      "step": 7851,
      "time_per_iteration": 2.737046480178833
    },
    {
      "auxiliary_loss_clip": 0.01123201,
      "auxiliary_loss_mlp": 0.01031643,
      "balance_loss_clip": 1.04452026,
      "balance_loss_mlp": 1.02322114,
      "epoch": 0.9441471772981422,
      "flos": 19238136255360.0,
      "grad_norm": 2.0261351100872327,
      "language_loss": 0.77686918,
      "learning_rate": 3.257143825132847e-08,
      "loss": 0.79841763,
      "num_input_tokens_seen": 169740355,
      "step": 7852,
      "time_per_iteration": 2.7286624908447266
    },
    {
      "auxiliary_loss_clip": 0.01130929,
      "auxiliary_loss_mlp": 0.01034736,
      "balance_loss_clip": 1.04269528,
      "balance_loss_mlp": 1.02740431,
      "epoch": 0.9442674201887813,
      "flos": 25739799379200.0,
      "grad_norm": 1.7470167872566034,
      "language_loss": 0.75996089,
      "learning_rate": 3.243157567005106e-08,
      "loss": 0.78161752,
      "num_input_tokens_seen": 169758535,
      "step": 7853,
      "time_per_iteration": 2.7783186435699463
    },
    {
      "auxiliary_loss_clip": 0.01150461,
      "auxiliary_loss_mlp": 0.01027338,
      "balance_loss_clip": 1.05073714,
      "balance_loss_mlp": 1.01917183,
      "epoch": 0.9443876630794205,
      "flos": 15523321737600.0,
      "grad_norm": 1.969042950151432,
      "language_loss": 0.63976717,
      "learning_rate": 3.2292011567339296e-08,
      "loss": 0.66154516,
      "num_input_tokens_seen": 169776340,
      "step": 7854,
      "time_per_iteration": 2.676964282989502
    },
    {
      "auxiliary_loss_clip": 0.0114261,
      "auxiliary_loss_mlp": 0.01022863,
      "balance_loss_clip": 1.04918158,
      "balance_loss_mlp": 1.01560044,
      "epoch": 0.9445079059700595,
      "flos": 13400821128960.0,
      "grad_norm": 2.0603118609388553,
      "language_loss": 0.5585314,
      "learning_rate": 3.21527459643649e-08,
      "loss": 0.58018613,
      "num_input_tokens_seen": 169793225,
      "step": 7855,
      "time_per_iteration": 2.65775465965271
    },
    {
      "auxiliary_loss_clip": 0.0115135,
      "auxiliary_loss_mlp": 0.00761662,
      "balance_loss_clip": 1.04886281,
      "balance_loss_mlp": 1.00049186,
      "epoch": 0.9446281488606986,
      "flos": 23659242877440.0,
      "grad_norm": 2.040145680148041,
      "language_loss": 0.74122107,
      "learning_rate": 3.2013778882254536e-08,
      "loss": 0.76035118,
      "num_input_tokens_seen": 169812020,
      "step": 7856,
      "time_per_iteration": 2.679495334625244
    },
    {
      "auxiliary_loss_clip": 0.01118596,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 1.04729867,
      "balance_loss_mlp": 1.02179098,
      "epoch": 0.9447483917513377,
      "flos": 25557337267200.0,
      "grad_norm": 2.3455698613584857,
      "language_loss": 0.76057965,
      "learning_rate": 3.1875110342088676e-08,
      "loss": 0.78205311,
      "num_input_tokens_seen": 169833470,
      "step": 7857,
      "time_per_iteration": 3.7731618881225586
    },
    {
      "auxiliary_loss_clip": 0.01115518,
      "auxiliary_loss_mlp": 0.01030517,
      "balance_loss_clip": 1.04061055,
      "balance_loss_mlp": 1.02294719,
      "epoch": 0.9448686346419768,
      "flos": 24535463247360.0,
      "grad_norm": 1.9054725294475463,
      "language_loss": 0.65535307,
      "learning_rate": 3.1736740364904035e-08,
      "loss": 0.67681348,
      "num_input_tokens_seen": 169854000,
      "step": 7858,
      "time_per_iteration": 3.6974499225616455
    },
    {
      "auxiliary_loss_clip": 0.01175268,
      "auxiliary_loss_mlp": 0.01023136,
      "balance_loss_clip": 1.05194378,
      "balance_loss_mlp": 1.01548862,
      "epoch": 0.9449888775326158,
      "flos": 14721256995840.0,
      "grad_norm": 2.139658181803255,
      "language_loss": 0.77000791,
      "learning_rate": 3.159866897169094e-08,
      "loss": 0.79199195,
      "num_input_tokens_seen": 169872200,
      "step": 7859,
      "time_per_iteration": 2.607753038406372
    },
    {
      "auxiliary_loss_clip": 0.0111385,
      "auxiliary_loss_mlp": 0.01032554,
      "balance_loss_clip": 1.0428493,
      "balance_loss_mlp": 1.02401876,
      "epoch": 0.945109120423255,
      "flos": 15447873219840.0,
      "grad_norm": 1.9545627215831773,
      "language_loss": 0.76047927,
      "learning_rate": 3.146089618339487e-08,
      "loss": 0.78194332,
      "num_input_tokens_seen": 169889055,
      "step": 7860,
      "time_per_iteration": 2.816004514694214
    },
    {
      "auxiliary_loss_clip": 0.01114093,
      "auxiliary_loss_mlp": 0.0103114,
      "balance_loss_clip": 1.04123878,
      "balance_loss_mlp": 1.02325428,
      "epoch": 0.9452293633138941,
      "flos": 25448097029760.0,
      "grad_norm": 1.7341950726282722,
      "language_loss": 0.68063247,
      "learning_rate": 3.132342202091554e-08,
      "loss": 0.70208478,
      "num_input_tokens_seen": 169909280,
      "step": 7861,
      "time_per_iteration": 2.8069169521331787
    },
    {
      "auxiliary_loss_clip": 0.01093202,
      "auxiliary_loss_mlp": 0.01035985,
      "balance_loss_clip": 1.04198945,
      "balance_loss_mlp": 1.02776575,
      "epoch": 0.9453496062045331,
      "flos": 21215342350080.0,
      "grad_norm": 2.033976960175136,
      "language_loss": 0.68502647,
      "learning_rate": 3.1186246505107595e-08,
      "loss": 0.70631838,
      "num_input_tokens_seen": 169928420,
      "step": 7862,
      "time_per_iteration": 2.934431314468384
    },
    {
      "auxiliary_loss_clip": 0.01118795,
      "auxiliary_loss_mlp": 0.01033605,
      "balance_loss_clip": 1.04615104,
      "balance_loss_mlp": 1.0257374,
      "epoch": 0.9454698490951723,
      "flos": 20010898477440.0,
      "grad_norm": 1.8522884927046983,
      "language_loss": 0.83768618,
      "learning_rate": 3.104936965678084e-08,
      "loss": 0.85921019,
      "num_input_tokens_seen": 169946750,
      "step": 7863,
      "time_per_iteration": 4.037428140640259
    },
    {
      "auxiliary_loss_clip": 0.01123546,
      "auxiliary_loss_mlp": 0.01024882,
      "balance_loss_clip": 1.04511511,
      "balance_loss_mlp": 1.01767588,
      "epoch": 0.9455900919858113,
      "flos": 21069652786560.0,
      "grad_norm": 2.3008980354662336,
      "language_loss": 0.82132632,
      "learning_rate": 3.091279149669956e-08,
      "loss": 0.84281063,
      "num_input_tokens_seen": 169965540,
      "step": 7864,
      "time_per_iteration": 2.767274856567383
    },
    {
      "auxiliary_loss_clip": 0.01153752,
      "auxiliary_loss_mlp": 0.01026932,
      "balance_loss_clip": 1.04609895,
      "balance_loss_mlp": 1.01952291,
      "epoch": 0.9457103348764504,
      "flos": 20740854666240.0,
      "grad_norm": 2.2640518445238524,
      "language_loss": 0.74072397,
      "learning_rate": 3.0776512045581624e-08,
      "loss": 0.7625308,
      "num_input_tokens_seen": 169984330,
      "step": 7865,
      "time_per_iteration": 2.6842312812805176
    },
    {
      "auxiliary_loss_clip": 0.01132433,
      "auxiliary_loss_mlp": 0.00762165,
      "balance_loss_clip": 1.04964685,
      "balance_loss_mlp": 1.0004108,
      "epoch": 0.9458305777670896,
      "flos": 21428363957760.0,
      "grad_norm": 2.574337404380048,
      "language_loss": 0.77619487,
      "learning_rate": 3.0640531324101384e-08,
      "loss": 0.79514086,
      "num_input_tokens_seen": 170002095,
      "step": 7866,
      "time_per_iteration": 2.7128374576568604
    },
    {
      "auxiliary_loss_clip": 0.01168917,
      "auxiliary_loss_mlp": 0.01033295,
      "balance_loss_clip": 1.05396581,
      "balance_loss_mlp": 1.02480149,
      "epoch": 0.9459508206577286,
      "flos": 20011185786240.0,
      "grad_norm": 2.069616505648207,
      "language_loss": 0.76054776,
      "learning_rate": 3.0504849352886554e-08,
      "loss": 0.78256989,
      "num_input_tokens_seen": 170020240,
      "step": 7867,
      "time_per_iteration": 2.8179211616516113
    },
    {
      "auxiliary_loss_clip": 0.01141923,
      "auxiliary_loss_mlp": 0.01032149,
      "balance_loss_clip": 1.04804754,
      "balance_loss_mlp": 1.02358413,
      "epoch": 0.9460710635483677,
      "flos": 12166428291840.0,
      "grad_norm": 3.0544249127326433,
      "language_loss": 0.71405733,
      "learning_rate": 3.036946615252023e-08,
      "loss": 0.735798,
      "num_input_tokens_seen": 170035770,
      "step": 7868,
      "time_per_iteration": 2.625878095626831
    },
    {
      "auxiliary_loss_clip": 0.01078551,
      "auxiliary_loss_mlp": 0.01028941,
      "balance_loss_clip": 1.04046011,
      "balance_loss_mlp": 1.02145219,
      "epoch": 0.9461913064390068,
      "flos": 34276196229120.0,
      "grad_norm": 2.248246903236765,
      "language_loss": 0.66828346,
      "learning_rate": 3.0234381743539984e-08,
      "loss": 0.68935835,
      "num_input_tokens_seen": 170053385,
      "step": 7869,
      "time_per_iteration": 2.902240753173828
    },
    {
      "auxiliary_loss_clip": 0.01147545,
      "auxiliary_loss_mlp": 0.01023471,
      "balance_loss_clip": 1.04697347,
      "balance_loss_mlp": 1.01636004,
      "epoch": 0.9463115493296459,
      "flos": 19463763536640.0,
      "grad_norm": 2.8299975000715714,
      "language_loss": 0.80313623,
      "learning_rate": 3.0099596146437863e-08,
      "loss": 0.82484639,
      "num_input_tokens_seen": 170070490,
      "step": 7870,
      "time_per_iteration": 2.66120982170105
    },
    {
      "auxiliary_loss_clip": 0.01052508,
      "auxiliary_loss_mlp": 0.01001856,
      "balance_loss_clip": 1.00845551,
      "balance_loss_mlp": 1.00095594,
      "epoch": 0.946431792220285,
      "flos": 70570824387840.0,
      "grad_norm": 0.8814141774647044,
      "language_loss": 0.59987092,
      "learning_rate": 2.996510938166086e-08,
      "loss": 0.62041461,
      "num_input_tokens_seen": 170133465,
      "step": 7871,
      "time_per_iteration": 3.3372788429260254
    },
    {
      "auxiliary_loss_clip": 0.01129744,
      "auxiliary_loss_mlp": 0.01029544,
      "balance_loss_clip": 1.04823482,
      "balance_loss_mlp": 1.02188468,
      "epoch": 0.9465520351109241,
      "flos": 18947906363520.0,
      "grad_norm": 1.9704099491644833,
      "language_loss": 0.7349385,
      "learning_rate": 2.983092146960997e-08,
      "loss": 0.75653136,
      "num_input_tokens_seen": 170150810,
      "step": 7872,
      "time_per_iteration": 2.721015453338623
    },
    {
      "auxiliary_loss_clip": 0.01148323,
      "auxiliary_loss_mlp": 0.01031931,
      "balance_loss_clip": 1.0484283,
      "balance_loss_mlp": 1.02391481,
      "epoch": 0.9466722780015632,
      "flos": 19135647774720.0,
      "grad_norm": 2.2011161091459615,
      "language_loss": 0.80122924,
      "learning_rate": 2.9697032430642256e-08,
      "loss": 0.82303172,
      "num_input_tokens_seen": 170169025,
      "step": 7873,
      "time_per_iteration": 2.689573049545288
    },
    {
      "auxiliary_loss_clip": 0.01125099,
      "auxiliary_loss_mlp": 0.01022706,
      "balance_loss_clip": 1.04411435,
      "balance_loss_mlp": 1.01545513,
      "epoch": 0.9467925208922022,
      "flos": 17237912520960.0,
      "grad_norm": 2.4156232772572896,
      "language_loss": 0.73844612,
      "learning_rate": 2.9563442285067906e-08,
      "loss": 0.75992417,
      "num_input_tokens_seen": 170186070,
      "step": 7874,
      "time_per_iteration": 3.6981234550476074
    },
    {
      "auxiliary_loss_clip": 0.01147924,
      "auxiliary_loss_mlp": 0.01026101,
      "balance_loss_clip": 1.04752219,
      "balance_loss_mlp": 1.01782155,
      "epoch": 0.9469127637828414,
      "flos": 29169016859520.0,
      "grad_norm": 3.0642123739844025,
      "language_loss": 0.79709578,
      "learning_rate": 2.943015105315294e-08,
      "loss": 0.81883597,
      "num_input_tokens_seen": 170206265,
      "step": 7875,
      "time_per_iteration": 2.8066041469573975
    },
    {
      "auxiliary_loss_clip": 0.01140851,
      "auxiliary_loss_mlp": 0.00762727,
      "balance_loss_clip": 1.04615045,
      "balance_loss_mlp": 1.00047946,
      "epoch": 0.9470330066734804,
      "flos": 26030460234240.0,
      "grad_norm": 2.7765916635939374,
      "language_loss": 0.67034364,
      "learning_rate": 2.929715875511718e-08,
      "loss": 0.68937945,
      "num_input_tokens_seen": 170225300,
      "step": 7876,
      "time_per_iteration": 2.746303081512451
    },
    {
      "auxiliary_loss_clip": 0.01144288,
      "auxiliary_loss_mlp": 0.01027187,
      "balance_loss_clip": 1.04338598,
      "balance_loss_mlp": 1.01992083,
      "epoch": 0.9471532495641195,
      "flos": 23440906056960.0,
      "grad_norm": 1.7970693755857712,
      "language_loss": 0.70069373,
      "learning_rate": 2.9164465411135375e-08,
      "loss": 0.72240847,
      "num_input_tokens_seen": 170245070,
      "step": 7877,
      "time_per_iteration": 2.730160713195801
    },
    {
      "auxiliary_loss_clip": 0.01147562,
      "auxiliary_loss_mlp": 0.01031659,
      "balance_loss_clip": 1.05143523,
      "balance_loss_mlp": 1.02404118,
      "epoch": 0.9472734924547586,
      "flos": 15815850099840.0,
      "grad_norm": 1.9413604960527528,
      "language_loss": 0.80778491,
      "learning_rate": 2.9032071041337426e-08,
      "loss": 0.82957709,
      "num_input_tokens_seen": 170263305,
      "step": 7878,
      "time_per_iteration": 2.700671911239624
    },
    {
      "auxiliary_loss_clip": 0.01122699,
      "auxiliary_loss_mlp": 0.01027302,
      "balance_loss_clip": 1.04643297,
      "balance_loss_mlp": 1.0198102,
      "epoch": 0.9473937353453977,
      "flos": 11181793697280.0,
      "grad_norm": 1.68660166038838,
      "language_loss": 0.72525001,
      "learning_rate": 2.889997566580704e-08,
      "loss": 0.74675,
      "num_input_tokens_seen": 170281460,
      "step": 7879,
      "time_per_iteration": 2.7702488899230957
    },
    {
      "auxiliary_loss_clip": 0.01176291,
      "auxiliary_loss_mlp": 0.01026178,
      "balance_loss_clip": 1.05131233,
      "balance_loss_mlp": 1.01834011,
      "epoch": 0.9475139782360368,
      "flos": 25775530433280.0,
      "grad_norm": 1.8036620270602284,
      "language_loss": 0.70356119,
      "learning_rate": 2.8768179304583086e-08,
      "loss": 0.72558594,
      "num_input_tokens_seen": 170303515,
      "step": 7880,
      "time_per_iteration": 2.663229465484619
    },
    {
      "auxiliary_loss_clip": 0.01113456,
      "auxiliary_loss_mlp": 0.01032303,
      "balance_loss_clip": 1.04204261,
      "balance_loss_mlp": 1.0245961,
      "epoch": 0.9476342211266758,
      "flos": 22820046451200.0,
      "grad_norm": 4.416835794815859,
      "language_loss": 0.73584378,
      "learning_rate": 2.8636681977659117e-08,
      "loss": 0.75730139,
      "num_input_tokens_seen": 170323165,
      "step": 7881,
      "time_per_iteration": 2.8016340732574463
    },
    {
      "auxiliary_loss_clip": 0.01166066,
      "auxiliary_loss_mlp": 0.01027173,
      "balance_loss_clip": 1.05353785,
      "balance_loss_mlp": 1.01931405,
      "epoch": 0.947754464017315,
      "flos": 20193611984640.0,
      "grad_norm": 2.023247338311431,
      "language_loss": 0.77932322,
      "learning_rate": 2.850548370498318e-08,
      "loss": 0.80125558,
      "num_input_tokens_seen": 170341005,
      "step": 7882,
      "time_per_iteration": 2.61847186088562
    },
    {
      "auxiliary_loss_clip": 0.01125414,
      "auxiliary_loss_mlp": 0.01025649,
      "balance_loss_clip": 1.04491305,
      "balance_loss_mlp": 1.01812148,
      "epoch": 0.9478747069079541,
      "flos": 24717925359360.0,
      "grad_norm": 1.7930308990005945,
      "language_loss": 0.71585423,
      "learning_rate": 2.8374584506457798e-08,
      "loss": 0.73736489,
      "num_input_tokens_seen": 170362280,
      "step": 7883,
      "time_per_iteration": 3.647916078567505
    },
    {
      "auxiliary_loss_clip": 0.0115477,
      "auxiliary_loss_mlp": 0.01027535,
      "balance_loss_clip": 1.04723454,
      "balance_loss_mlp": 1.01927924,
      "epoch": 0.9479949497985931,
      "flos": 21361355136000.0,
      "grad_norm": 2.280347306495216,
      "language_loss": 0.67730451,
      "learning_rate": 2.824398440193998e-08,
      "loss": 0.69912755,
      "num_input_tokens_seen": 170381080,
      "step": 7884,
      "time_per_iteration": 3.6985361576080322
    },
    {
      "auxiliary_loss_clip": 0.01152898,
      "auxiliary_loss_mlp": 0.01023627,
      "balance_loss_clip": 1.0468179,
      "balance_loss_mlp": 1.01547885,
      "epoch": 0.9481151926892323,
      "flos": 18148606968960.0,
      "grad_norm": 2.8059685065444113,
      "language_loss": 0.71545553,
      "learning_rate": 2.811368341124232e-08,
      "loss": 0.73722076,
      "num_input_tokens_seen": 170400150,
      "step": 7885,
      "time_per_iteration": 2.6590003967285156
    },
    {
      "auxiliary_loss_clip": 0.0115878,
      "auxiliary_loss_mlp": 0.01025322,
      "balance_loss_clip": 1.05175364,
      "balance_loss_mlp": 1.01807749,
      "epoch": 0.9482354355798713,
      "flos": 22128012046080.0,
      "grad_norm": 3.202594712143516,
      "language_loss": 0.68074811,
      "learning_rate": 2.7983681554131222e-08,
      "loss": 0.70258915,
      "num_input_tokens_seen": 170420410,
      "step": 7886,
      "time_per_iteration": 2.6751816272735596
    },
    {
      "auxiliary_loss_clip": 0.01162808,
      "auxiliary_loss_mlp": 0.00762006,
      "balance_loss_clip": 1.05249357,
      "balance_loss_mlp": 1.00044274,
      "epoch": 0.9483556784705104,
      "flos": 19063072344960.0,
      "grad_norm": 2.235821141661353,
      "language_loss": 0.7089873,
      "learning_rate": 2.7853978850327365e-08,
      "loss": 0.72823536,
      "num_input_tokens_seen": 170439580,
      "step": 7887,
      "time_per_iteration": 2.6268279552459717
    },
    {
      "auxiliary_loss_clip": 0.01125441,
      "auxiliary_loss_mlp": 0.01025362,
      "balance_loss_clip": 1.04731369,
      "balance_loss_mlp": 1.0174768,
      "epoch": 0.9484759213611496,
      "flos": 25777110631680.0,
      "grad_norm": 2.5505000606312467,
      "language_loss": 0.87353301,
      "learning_rate": 2.7724575319507225e-08,
      "loss": 0.89504099,
      "num_input_tokens_seen": 170459290,
      "step": 7888,
      "time_per_iteration": 2.7795090675354004
    },
    {
      "auxiliary_loss_clip": 0.01143238,
      "auxiliary_loss_mlp": 0.01023342,
      "balance_loss_clip": 1.04627061,
      "balance_loss_mlp": 1.01672602,
      "epoch": 0.9485961642517886,
      "flos": 20667740532480.0,
      "grad_norm": 1.8174478851452804,
      "language_loss": 0.77007711,
      "learning_rate": 2.759547098130044e-08,
      "loss": 0.79174292,
      "num_input_tokens_seen": 170478020,
      "step": 7889,
      "time_per_iteration": 3.7194106578826904
    },
    {
      "auxiliary_loss_clip": 0.0113142,
      "auxiliary_loss_mlp": 0.01029293,
      "balance_loss_clip": 1.04850245,
      "balance_loss_mlp": 1.02155662,
      "epoch": 0.9487164071424277,
      "flos": 22674069578880.0,
      "grad_norm": 1.8081687342868318,
      "language_loss": 0.76484072,
      "learning_rate": 2.746666585529267e-08,
      "loss": 0.78644788,
      "num_input_tokens_seen": 170498295,
      "step": 7890,
      "time_per_iteration": 2.727163314819336
    },
    {
      "auxiliary_loss_clip": 0.01099696,
      "auxiliary_loss_mlp": 0.01027243,
      "balance_loss_clip": 1.04489088,
      "balance_loss_mlp": 1.0195297,
      "epoch": 0.9488366500330668,
      "flos": 38726461716480.0,
      "grad_norm": 2.242183421107104,
      "language_loss": 0.74276155,
      "learning_rate": 2.73381599610234e-08,
      "loss": 0.76403087,
      "num_input_tokens_seen": 170518695,
      "step": 7891,
      "time_per_iteration": 2.944267988204956
    },
    {
      "auxiliary_loss_clip": 0.01146395,
      "auxiliary_loss_mlp": 0.01033823,
      "balance_loss_clip": 1.04818344,
      "balance_loss_mlp": 1.02572274,
      "epoch": 0.9489568929237059,
      "flos": 27890920149120.0,
      "grad_norm": 1.8066689523609945,
      "language_loss": 0.71655297,
      "learning_rate": 2.7209953317987033e-08,
      "loss": 0.7383551,
      "num_input_tokens_seen": 170539735,
      "step": 7892,
      "time_per_iteration": 2.7415857315063477
    },
    {
      "auxiliary_loss_clip": 0.01171292,
      "auxiliary_loss_mlp": 0.01021394,
      "balance_loss_clip": 1.05034351,
      "balance_loss_mlp": 1.01397359,
      "epoch": 0.9490771358143449,
      "flos": 33580642291200.0,
      "grad_norm": 7.354462168293882,
      "language_loss": 0.77981639,
      "learning_rate": 2.7082045945631793e-08,
      "loss": 0.80174321,
      "num_input_tokens_seen": 170561950,
      "step": 7893,
      "time_per_iteration": 2.7179813385009766
    },
    {
      "auxiliary_loss_clip": 0.01157169,
      "auxiliary_loss_mlp": 0.01021654,
      "balance_loss_clip": 1.04938757,
      "balance_loss_mlp": 1.01409054,
      "epoch": 0.9491973787049841,
      "flos": 14793796512000.0,
      "grad_norm": 4.396998152791707,
      "language_loss": 0.69709533,
      "learning_rate": 2.6954437863361712e-08,
      "loss": 0.71888351,
      "num_input_tokens_seen": 170579865,
      "step": 7894,
      "time_per_iteration": 2.613396644592285
    },
    {
      "auxiliary_loss_clip": 0.01091772,
      "auxiliary_loss_mlp": 0.01029828,
      "balance_loss_clip": 1.03911173,
      "balance_loss_mlp": 1.02232432,
      "epoch": 0.9493176215956232,
      "flos": 25332535998720.0,
      "grad_norm": 2.0699902030750263,
      "language_loss": 0.71054578,
      "learning_rate": 2.6827129090534862e-08,
      "loss": 0.73176181,
      "num_input_tokens_seen": 170600165,
      "step": 7895,
      "time_per_iteration": 2.8728816509246826
    },
    {
      "auxiliary_loss_clip": 0.01157493,
      "auxiliary_loss_mlp": 0.01026097,
      "balance_loss_clip": 1.04677093,
      "balance_loss_mlp": 1.01829529,
      "epoch": 0.9494378644862622,
      "flos": 21029971236480.0,
      "grad_norm": 1.9372464664501934,
      "language_loss": 0.78438413,
      "learning_rate": 2.670011964646335e-08,
      "loss": 0.80622,
      "num_input_tokens_seen": 170618845,
      "step": 7896,
      "time_per_iteration": 2.8500680923461914
    },
    {
      "auxiliary_loss_clip": 0.01143596,
      "auxiliary_loss_mlp": 0.01027937,
      "balance_loss_clip": 1.04774714,
      "balance_loss_mlp": 1.01981282,
      "epoch": 0.9495581073769014,
      "flos": 15195134148480.0,
      "grad_norm": 1.9377701516996801,
      "language_loss": 0.68273467,
      "learning_rate": 2.657340955041487e-08,
      "loss": 0.70445001,
      "num_input_tokens_seen": 170637620,
      "step": 7897,
      "time_per_iteration": 2.6113903522491455
    },
    {
      "auxiliary_loss_clip": 0.01150789,
      "auxiliary_loss_mlp": 0.01029438,
      "balance_loss_clip": 1.05026269,
      "balance_loss_mlp": 1.02136731,
      "epoch": 0.9496783502675404,
      "flos": 28616566705920.0,
      "grad_norm": 2.0975548315500854,
      "language_loss": 0.71836597,
      "learning_rate": 2.6446998821611167e-08,
      "loss": 0.74016821,
      "num_input_tokens_seen": 170657815,
      "step": 7898,
      "time_per_iteration": 2.7617578506469727
    },
    {
      "auxiliary_loss_clip": 0.01165937,
      "auxiliary_loss_mlp": 0.01028409,
      "balance_loss_clip": 1.05352402,
      "balance_loss_mlp": 1.02048182,
      "epoch": 0.9497985931581795,
      "flos": 14866874732160.0,
      "grad_norm": 2.915298504312722,
      "language_loss": 0.72292423,
      "learning_rate": 2.6320887479228228e-08,
      "loss": 0.74486768,
      "num_input_tokens_seen": 170674415,
      "step": 7899,
      "time_per_iteration": 3.5443267822265625
    },
    {
      "auxiliary_loss_clip": 0.0110337,
      "auxiliary_loss_mlp": 0.01026488,
      "balance_loss_clip": 1.04377949,
      "balance_loss_mlp": 1.01877546,
      "epoch": 0.9499188360488187,
      "flos": 27193319136000.0,
      "grad_norm": 2.1287164439821447,
      "language_loss": 0.72671032,
      "learning_rate": 2.619507554239786e-08,
      "loss": 0.74800891,
      "num_input_tokens_seen": 170692975,
      "step": 7900,
      "time_per_iteration": 2.854843854904175
    },
    {
      "auxiliary_loss_clip": 0.01157533,
      "auxiliary_loss_mlp": 0.00762027,
      "balance_loss_clip": 1.0494194,
      "balance_loss_mlp": 1.00044644,
      "epoch": 0.9500390789394577,
      "flos": 24316479982080.0,
      "grad_norm": 1.6216264763753354,
      "language_loss": 0.69661558,
      "learning_rate": 2.606956303020502e-08,
      "loss": 0.71581119,
      "num_input_tokens_seen": 170713780,
      "step": 7901,
      "time_per_iteration": 2.704202890396118
    },
    {
      "auxiliary_loss_clip": 0.01116229,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 1.04514909,
      "balance_loss_mlp": 1.01999736,
      "epoch": 0.9501593218300968,
      "flos": 14354752573440.0,
      "grad_norm": 1.8913396483089338,
      "language_loss": 0.844226,
      "learning_rate": 2.5944349961690036e-08,
      "loss": 0.86566025,
      "num_input_tokens_seen": 170730800,
      "step": 7902,
      "time_per_iteration": 2.729989767074585
    },
    {
      "auxiliary_loss_clip": 0.01132151,
      "auxiliary_loss_mlp": 0.0102738,
      "balance_loss_clip": 1.0440352,
      "balance_loss_mlp": 1.01976836,
      "epoch": 0.9502795647207359,
      "flos": 38728113742080.0,
      "grad_norm": 1.6117712070849117,
      "language_loss": 0.73160064,
      "learning_rate": 2.581943635584749e-08,
      "loss": 0.753196,
      "num_input_tokens_seen": 170753630,
      "step": 7903,
      "time_per_iteration": 2.8964462280273438
    },
    {
      "auxiliary_loss_clip": 0.0110542,
      "auxiliary_loss_mlp": 0.01027835,
      "balance_loss_clip": 1.04530549,
      "balance_loss_mlp": 1.02090907,
      "epoch": 0.950399807611375,
      "flos": 40808023799040.0,
      "grad_norm": 1.584742337673387,
      "language_loss": 0.65294504,
      "learning_rate": 2.569482223162689e-08,
      "loss": 0.6742776,
      "num_input_tokens_seen": 170777605,
      "step": 7904,
      "time_per_iteration": 2.9041080474853516
    },
    {
      "auxiliary_loss_clip": 0.01172445,
      "auxiliary_loss_mlp": 0.01029302,
      "balance_loss_clip": 1.04781091,
      "balance_loss_mlp": 1.02152991,
      "epoch": 0.950520050502014,
      "flos": 23440403266560.0,
      "grad_norm": 1.8312090843514388,
      "language_loss": 0.72606933,
      "learning_rate": 2.5570507607932e-08,
      "loss": 0.74808681,
      "num_input_tokens_seen": 170797520,
      "step": 7905,
      "time_per_iteration": 2.6748998165130615
    },
    {
      "auxiliary_loss_clip": 0.0111732,
      "auxiliary_loss_mlp": 0.0103069,
      "balance_loss_clip": 1.04478621,
      "balance_loss_mlp": 1.02246737,
      "epoch": 0.9506402933926532,
      "flos": 17783718658560.0,
      "grad_norm": 3.372719458025101,
      "language_loss": 0.64300287,
      "learning_rate": 2.54464925036213e-08,
      "loss": 0.66448295,
      "num_input_tokens_seen": 170814810,
      "step": 7906,
      "time_per_iteration": 2.6949732303619385
    },
    {
      "auxiliary_loss_clip": 0.01128536,
      "auxiliary_loss_mlp": 0.01025819,
      "balance_loss_clip": 1.04643226,
      "balance_loss_mlp": 1.01835978,
      "epoch": 0.9507605362832923,
      "flos": 32561928668160.0,
      "grad_norm": 1.9534810604402075,
      "language_loss": 0.60946798,
      "learning_rate": 2.532277693750773e-08,
      "loss": 0.63101149,
      "num_input_tokens_seen": 170835735,
      "step": 7907,
      "time_per_iteration": 2.840325355529785
    },
    {
      "auxiliary_loss_clip": 0.01164538,
      "auxiliary_loss_mlp": 0.01030082,
      "balance_loss_clip": 1.05427957,
      "balance_loss_mlp": 1.02170134,
      "epoch": 0.9508807791739313,
      "flos": 19602054898560.0,
      "grad_norm": 3.0303338141364566,
      "language_loss": 0.76057541,
      "learning_rate": 2.5199360928358948e-08,
      "loss": 0.78252161,
      "num_input_tokens_seen": 170852970,
      "step": 7908,
      "time_per_iteration": 2.587822198867798
    },
    {
      "auxiliary_loss_clip": 0.0116621,
      "auxiliary_loss_mlp": 0.00761306,
      "balance_loss_clip": 1.04998732,
      "balance_loss_mlp": 1.000525,
      "epoch": 0.9510010220645704,
      "flos": 21471852349440.0,
      "grad_norm": 2.0827902117261448,
      "language_loss": 0.87402695,
      "learning_rate": 2.507624449489665e-08,
      "loss": 0.89330214,
      "num_input_tokens_seen": 170871600,
      "step": 7909,
      "time_per_iteration": 3.585780143737793
    },
    {
      "auxiliary_loss_clip": 0.01133473,
      "auxiliary_loss_mlp": 0.01029442,
      "balance_loss_clip": 1.04903197,
      "balance_loss_mlp": 1.02208114,
      "epoch": 0.9511212649552095,
      "flos": 18879999701760.0,
      "grad_norm": 1.9126064442024053,
      "language_loss": 0.65465939,
      "learning_rate": 2.495342765579811e-08,
      "loss": 0.6762886,
      "num_input_tokens_seen": 170890260,
      "step": 7910,
      "time_per_iteration": 3.666285514831543
    },
    {
      "auxiliary_loss_clip": 0.0114603,
      "auxiliary_loss_mlp": 0.01025244,
      "balance_loss_clip": 1.04778469,
      "balance_loss_mlp": 1.01820135,
      "epoch": 0.9512415078458486,
      "flos": 20810521094400.0,
      "grad_norm": 1.8562988788156949,
      "language_loss": 0.71333528,
      "learning_rate": 2.4830910429693984e-08,
      "loss": 0.73504794,
      "num_input_tokens_seen": 170910220,
      "step": 7911,
      "time_per_iteration": 2.659508466720581
    },
    {
      "auxiliary_loss_clip": 0.01131933,
      "auxiliary_loss_mlp": 0.0102743,
      "balance_loss_clip": 1.04135013,
      "balance_loss_mlp": 1.01943088,
      "epoch": 0.9513617507364877,
      "flos": 18369565482240.0,
      "grad_norm": 1.9515148432232938,
      "language_loss": 0.79784441,
      "learning_rate": 2.470869283517052e-08,
      "loss": 0.8194381,
      "num_input_tokens_seen": 170928255,
      "step": 7912,
      "time_per_iteration": 2.8196120262145996
    },
    {
      "auxiliary_loss_clip": 0.01172215,
      "auxiliary_loss_mlp": 0.00761687,
      "balance_loss_clip": 1.05196953,
      "balance_loss_mlp": 1.00054944,
      "epoch": 0.9514819936271268,
      "flos": 25010166412800.0,
      "grad_norm": 1.9702635649673759,
      "language_loss": 0.7724725,
      "learning_rate": 2.458677489076777e-08,
      "loss": 0.79181153,
      "num_input_tokens_seen": 170949265,
      "step": 7913,
      "time_per_iteration": 2.6227927207946777
    },
    {
      "auxiliary_loss_clip": 0.01172725,
      "auxiliary_loss_mlp": 0.010296,
      "balance_loss_clip": 1.05170703,
      "balance_loss_mlp": 1.02174461,
      "epoch": 0.9516022365177659,
      "flos": 18662129758080.0,
      "grad_norm": 2.0359729964443614,
      "language_loss": 0.83080637,
      "learning_rate": 2.446515661498072e-08,
      "loss": 0.85282964,
      "num_input_tokens_seen": 170968595,
      "step": 7914,
      "time_per_iteration": 2.6470422744750977
    },
    {
      "auxiliary_loss_clip": 0.01105842,
      "auxiliary_loss_mlp": 0.01026545,
      "balance_loss_clip": 1.04397273,
      "balance_loss_mlp": 1.0191834,
      "epoch": 0.9517224794084049,
      "flos": 25372109808000.0,
      "grad_norm": 3.5283921183663685,
      "language_loss": 0.74373686,
      "learning_rate": 2.434383802625861e-08,
      "loss": 0.76506066,
      "num_input_tokens_seen": 170987550,
      "step": 7915,
      "time_per_iteration": 3.807798385620117
    },
    {
      "auxiliary_loss_clip": 0.01138366,
      "auxiliary_loss_mlp": 0.01028148,
      "balance_loss_clip": 1.04413116,
      "balance_loss_mlp": 1.02086186,
      "epoch": 0.9518427222990441,
      "flos": 21470918595840.0,
      "grad_norm": 1.967029848980665,
      "language_loss": 0.74311507,
      "learning_rate": 2.4222819143005168e-08,
      "loss": 0.76478016,
      "num_input_tokens_seen": 171007145,
      "step": 7916,
      "time_per_iteration": 2.6519429683685303
    },
    {
      "auxiliary_loss_clip": 0.01157418,
      "auxiliary_loss_mlp": 0.00761527,
      "balance_loss_clip": 1.05058134,
      "balance_loss_mlp": 1.00048804,
      "epoch": 0.9519629651896832,
      "flos": 21033634423680.0,
      "grad_norm": 1.8499310873715193,
      "language_loss": 0.8100782,
      "learning_rate": 2.4102099983579706e-08,
      "loss": 0.82926762,
      "num_input_tokens_seen": 171026295,
      "step": 7917,
      "time_per_iteration": 2.869807481765747
    },
    {
      "auxiliary_loss_clip": 0.01143226,
      "auxiliary_loss_mlp": 0.01027472,
      "balance_loss_clip": 1.05029988,
      "balance_loss_mlp": 1.01952362,
      "epoch": 0.9520832080803222,
      "flos": 21689219502720.0,
      "grad_norm": 1.7066842805509541,
      "language_loss": 0.77623326,
      "learning_rate": 2.3981680566294236e-08,
      "loss": 0.79794025,
      "num_input_tokens_seen": 171045895,
      "step": 7918,
      "time_per_iteration": 2.701976776123047
    },
    {
      "auxiliary_loss_clip": 0.01133806,
      "auxiliary_loss_mlp": 0.01029207,
      "balance_loss_clip": 1.05056572,
      "balance_loss_mlp": 1.02207279,
      "epoch": 0.9522034509709614,
      "flos": 23145289125120.0,
      "grad_norm": 1.9817329803855714,
      "language_loss": 0.73719651,
      "learning_rate": 2.3861560909416822e-08,
      "loss": 0.75882661,
      "num_input_tokens_seen": 171065445,
      "step": 7919,
      "time_per_iteration": 2.765209913253784
    },
    {
      "auxiliary_loss_clip": 0.01134218,
      "auxiliary_loss_mlp": 0.01026334,
      "balance_loss_clip": 1.04944682,
      "balance_loss_mlp": 1.01897907,
      "epoch": 0.9523236938616004,
      "flos": 24679428958080.0,
      "grad_norm": 1.6638853264966273,
      "language_loss": 0.82498002,
      "learning_rate": 2.3741741031169325e-08,
      "loss": 0.84658551,
      "num_input_tokens_seen": 171085015,
      "step": 7920,
      "time_per_iteration": 2.7337136268615723
    },
    {
      "auxiliary_loss_clip": 0.01130581,
      "auxiliary_loss_mlp": 0.01027625,
      "balance_loss_clip": 1.04708576,
      "balance_loss_mlp": 1.02003765,
      "epoch": 0.9524439367522395,
      "flos": 22672309812480.0,
      "grad_norm": 1.9546077249513785,
      "language_loss": 0.71642059,
      "learning_rate": 2.3622220949728544e-08,
      "loss": 0.73800266,
      "num_input_tokens_seen": 171103900,
      "step": 7921,
      "time_per_iteration": 2.802765369415283
    },
    {
      "auxiliary_loss_clip": 0.01108686,
      "auxiliary_loss_mlp": 0.01027327,
      "balance_loss_clip": 1.03989947,
      "balance_loss_mlp": 1.01936424,
      "epoch": 0.9525641796428787,
      "flos": 34055525024640.0,
      "grad_norm": 2.581680910851165,
      "language_loss": 0.61241937,
      "learning_rate": 2.3503000683225526e-08,
      "loss": 0.63377953,
      "num_input_tokens_seen": 171121615,
      "step": 7922,
      "time_per_iteration": 2.8500468730926514
    },
    {
      "auxiliary_loss_clip": 0.01112951,
      "auxiliary_loss_mlp": 0.01029534,
      "balance_loss_clip": 1.04094243,
      "balance_loss_mlp": 1.02180052,
      "epoch": 0.9526844225335177,
      "flos": 16727083251840.0,
      "grad_norm": 2.824294020500618,
      "language_loss": 0.84732372,
      "learning_rate": 2.3384080249745585e-08,
      "loss": 0.86874855,
      "num_input_tokens_seen": 171139505,
      "step": 7923,
      "time_per_iteration": 2.6835529804229736
    },
    {
      "auxiliary_loss_clip": 0.01161712,
      "auxiliary_loss_mlp": 0.01023701,
      "balance_loss_clip": 1.05050135,
      "balance_loss_mlp": 1.01618445,
      "epoch": 0.9528046654241568,
      "flos": 36939367330560.0,
      "grad_norm": 2.5280666467500055,
      "language_loss": 0.82606083,
      "learning_rate": 2.3265459667329178e-08,
      "loss": 0.84791493,
      "num_input_tokens_seen": 171158995,
      "step": 7924,
      "time_per_iteration": 2.8555803298950195
    },
    {
      "auxiliary_loss_clip": 0.01139391,
      "auxiliary_loss_mlp": 0.0102056,
      "balance_loss_clip": 1.04614806,
      "balance_loss_mlp": 1.01259124,
      "epoch": 0.9529249083147959,
      "flos": 18255010032000.0,
      "grad_norm": 2.32841935406248,
      "language_loss": 0.86612052,
      "learning_rate": 2.31471389539708e-08,
      "loss": 0.88771999,
      "num_input_tokens_seen": 171176120,
      "step": 7925,
      "time_per_iteration": 3.630161762237549
    },
    {
      "auxiliary_loss_clip": 0.01152147,
      "auxiliary_loss_mlp": 0.01028264,
      "balance_loss_clip": 1.04888248,
      "balance_loss_mlp": 1.02085781,
      "epoch": 0.953045151205435,
      "flos": 28658438985600.0,
      "grad_norm": 1.9919906057838712,
      "language_loss": 0.72471642,
      "learning_rate": 2.3029118127619872e-08,
      "loss": 0.74652058,
      "num_input_tokens_seen": 171195835,
      "step": 7926,
      "time_per_iteration": 2.7380993366241455
    },
    {
      "auxiliary_loss_clip": 0.01135339,
      "auxiliary_loss_mlp": 0.01030124,
      "balance_loss_clip": 1.04825664,
      "balance_loss_mlp": 1.02186298,
      "epoch": 0.953165394096074,
      "flos": 21835232288640.0,
      "grad_norm": 2.0289280897774686,
      "language_loss": 0.86996663,
      "learning_rate": 2.2911397206179628e-08,
      "loss": 0.89162123,
      "num_input_tokens_seen": 171212585,
      "step": 7927,
      "time_per_iteration": 2.7841110229492188
    },
    {
      "auxiliary_loss_clip": 0.01147673,
      "auxiliary_loss_mlp": 0.01025971,
      "balance_loss_clip": 1.05059361,
      "balance_loss_mlp": 1.0180912,
      "epoch": 0.9532856369867132,
      "flos": 19975059682560.0,
      "grad_norm": 1.7436136703822571,
      "language_loss": 0.6279158,
      "learning_rate": 2.279397620750845e-08,
      "loss": 0.64965224,
      "num_input_tokens_seen": 171231630,
      "step": 7928,
      "time_per_iteration": 2.6938588619232178
    },
    {
      "auxiliary_loss_clip": 0.01152471,
      "auxiliary_loss_mlp": 0.01026561,
      "balance_loss_clip": 1.05075741,
      "balance_loss_mlp": 1.01924157,
      "epoch": 0.9534058798773523,
      "flos": 15049588239360.0,
      "grad_norm": 2.035388712873834,
      "language_loss": 0.78898287,
      "learning_rate": 2.2676855149419195e-08,
      "loss": 0.81077325,
      "num_input_tokens_seen": 171248800,
      "step": 7929,
      "time_per_iteration": 2.7459826469421387
    },
    {
      "auxiliary_loss_clip": 0.01147559,
      "auxiliary_loss_mlp": 0.01025978,
      "balance_loss_clip": 1.05043912,
      "balance_loss_mlp": 1.01894426,
      "epoch": 0.9535261227679913,
      "flos": 17602800831360.0,
      "grad_norm": 2.57766960628019,
      "language_loss": 0.75406104,
      "learning_rate": 2.2560034049678988e-08,
      "loss": 0.77579641,
      "num_input_tokens_seen": 171263150,
      "step": 7930,
      "time_per_iteration": 2.674793243408203
    },
    {
      "auxiliary_loss_clip": 0.01150375,
      "auxiliary_loss_mlp": 0.01030263,
      "balance_loss_clip": 1.04862404,
      "balance_loss_mlp": 1.02265739,
      "epoch": 0.9536463656586305,
      "flos": 23142954741120.0,
      "grad_norm": 1.884127554176705,
      "language_loss": 0.75424266,
      "learning_rate": 2.2443512926008988e-08,
      "loss": 0.77604902,
      "num_input_tokens_seen": 171282480,
      "step": 7931,
      "time_per_iteration": 2.7501511573791504
    },
    {
      "auxiliary_loss_clip": 0.01146788,
      "auxiliary_loss_mlp": 0.01026689,
      "balance_loss_clip": 1.04756761,
      "balance_loss_mlp": 1.01942968,
      "epoch": 0.9537666085492695,
      "flos": 18625033987200.0,
      "grad_norm": 3.1642612055007224,
      "language_loss": 0.70227545,
      "learning_rate": 2.2327291796085946e-08,
      "loss": 0.72401023,
      "num_input_tokens_seen": 171300840,
      "step": 7932,
      "time_per_iteration": 2.6639602184295654
    },
    {
      "auxiliary_loss_clip": 0.01159128,
      "auxiliary_loss_mlp": 0.01027365,
      "balance_loss_clip": 1.04831791,
      "balance_loss_mlp": 1.01891923,
      "epoch": 0.9538868514399086,
      "flos": 18989347680000.0,
      "grad_norm": 2.4370667764074354,
      "language_loss": 0.76837474,
      "learning_rate": 2.2211370677540197e-08,
      "loss": 0.79023969,
      "num_input_tokens_seen": 171317365,
      "step": 7933,
      "time_per_iteration": 2.741206169128418
    },
    {
      "auxiliary_loss_clip": 0.01153979,
      "auxiliary_loss_mlp": 0.01027185,
      "balance_loss_clip": 1.0497036,
      "balance_loss_mlp": 1.01946604,
      "epoch": 0.9540070943305478,
      "flos": 16800556521600.0,
      "grad_norm": 2.7721348119894875,
      "language_loss": 0.78745115,
      "learning_rate": 2.2095749587957012e-08,
      "loss": 0.80926281,
      "num_input_tokens_seen": 171335270,
      "step": 7934,
      "time_per_iteration": 3.512990713119507
    },
    {
      "auxiliary_loss_clip": 0.01122894,
      "auxiliary_loss_mlp": 0.01025935,
      "balance_loss_clip": 1.0457561,
      "balance_loss_mlp": 1.01836526,
      "epoch": 0.9541273372211868,
      "flos": 20156911263360.0,
      "grad_norm": 1.9985392784548348,
      "language_loss": 0.69552332,
      "learning_rate": 2.1980428544876138e-08,
      "loss": 0.71701157,
      "num_input_tokens_seen": 171353910,
      "step": 7935,
      "time_per_iteration": 2.8407247066497803
    },
    {
      "auxiliary_loss_clip": 0.01137805,
      "auxiliary_loss_mlp": 0.01028,
      "balance_loss_clip": 1.04779029,
      "balance_loss_mlp": 1.02041233,
      "epoch": 0.9542475801118259,
      "flos": 26725511381760.0,
      "grad_norm": 2.2734322225209063,
      "language_loss": 0.74478567,
      "learning_rate": 2.1865407565791584e-08,
      "loss": 0.76644373,
      "num_input_tokens_seen": 171375480,
      "step": 7936,
      "time_per_iteration": 3.758927345275879
    },
    {
      "auxiliary_loss_clip": 0.01145347,
      "auxiliary_loss_mlp": 0.00761913,
      "balance_loss_clip": 1.04624236,
      "balance_loss_mlp": 1.00051856,
      "epoch": 0.954367823002465,
      "flos": 23330911633920.0,
      "grad_norm": 3.177211050969698,
      "language_loss": 0.7748124,
      "learning_rate": 2.175068666815183e-08,
      "loss": 0.79388499,
      "num_input_tokens_seen": 171396320,
      "step": 7937,
      "time_per_iteration": 2.7262563705444336
    },
    {
      "auxiliary_loss_clip": 0.01119274,
      "auxiliary_loss_mlp": 0.01027899,
      "balance_loss_clip": 1.04808903,
      "balance_loss_mlp": 1.01963234,
      "epoch": 0.9544880658931041,
      "flos": 14902713527040.0,
      "grad_norm": 2.0101103238098474,
      "language_loss": 0.79089081,
      "learning_rate": 2.163626586935985e-08,
      "loss": 0.81236261,
      "num_input_tokens_seen": 171412860,
      "step": 7938,
      "time_per_iteration": 2.7659990787506104
    },
    {
      "auxiliary_loss_clip": 0.01088625,
      "auxiliary_loss_mlp": 0.01027338,
      "balance_loss_clip": 1.04600275,
      "balance_loss_mlp": 1.01948261,
      "epoch": 0.9546083087837431,
      "flos": 29095902725760.0,
      "grad_norm": 1.8440489260069277,
      "language_loss": 0.63236028,
      "learning_rate": 2.1522145186773755e-08,
      "loss": 0.65351987,
      "num_input_tokens_seen": 171431780,
      "step": 7939,
      "time_per_iteration": 2.924673318862915
    },
    {
      "auxiliary_loss_clip": 0.01094618,
      "auxiliary_loss_mlp": 0.01024558,
      "balance_loss_clip": 1.04496694,
      "balance_loss_mlp": 1.01722622,
      "epoch": 0.9547285516743822,
      "flos": 21142335957120.0,
      "grad_norm": 1.6373956216477294,
      "language_loss": 0.85581374,
      "learning_rate": 2.140832463770481e-08,
      "loss": 0.87700552,
      "num_input_tokens_seen": 171450975,
      "step": 7940,
      "time_per_iteration": 2.8535406589508057
    },
    {
      "auxiliary_loss_clip": 0.01143998,
      "auxiliary_loss_mlp": 0.01027658,
      "balance_loss_clip": 1.04770899,
      "balance_loss_mlp": 1.01931918,
      "epoch": 0.9548487945650214,
      "flos": 27490157130240.0,
      "grad_norm": 2.0841379220063927,
      "language_loss": 0.76410335,
      "learning_rate": 2.129480423941987e-08,
      "loss": 0.78581989,
      "num_input_tokens_seen": 171467645,
      "step": 7941,
      "time_per_iteration": 3.8127529621124268
    },
    {
      "auxiliary_loss_clip": 0.01119257,
      "auxiliary_loss_mlp": 0.00761131,
      "balance_loss_clip": 1.04842496,
      "balance_loss_mlp": 1.00046086,
      "epoch": 0.9549690374556604,
      "flos": 22273198819200.0,
      "grad_norm": 1.9509386302377376,
      "language_loss": 0.8039577,
      "learning_rate": 2.1181584009140052e-08,
      "loss": 0.82276165,
      "num_input_tokens_seen": 171487185,
      "step": 7942,
      "time_per_iteration": 2.758244752883911
    },
    {
      "auxiliary_loss_clip": 0.01146685,
      "auxiliary_loss_mlp": 0.01027472,
      "balance_loss_clip": 1.05040121,
      "balance_loss_mlp": 1.02015853,
      "epoch": 0.9550892803462995,
      "flos": 17595294888960.0,
      "grad_norm": 2.025148076756248,
      "language_loss": 0.84272301,
      "learning_rate": 2.10686639640405e-08,
      "loss": 0.86446464,
      "num_input_tokens_seen": 171501275,
      "step": 7943,
      "time_per_iteration": 2.73677134513855
    },
    {
      "auxiliary_loss_clip": 0.01130683,
      "auxiliary_loss_mlp": 0.01026666,
      "balance_loss_clip": 1.04556441,
      "balance_loss_mlp": 1.0188067,
      "epoch": 0.9552095232369386,
      "flos": 24353144789760.0,
      "grad_norm": 1.649194293945545,
      "language_loss": 0.81278229,
      "learning_rate": 2.0956044121251294e-08,
      "loss": 0.83435577,
      "num_input_tokens_seen": 171520060,
      "step": 7944,
      "time_per_iteration": 2.782238006591797
    },
    {
      "auxiliary_loss_clip": 0.01158892,
      "auxiliary_loss_mlp": 0.01033337,
      "balance_loss_clip": 1.05159593,
      "balance_loss_mlp": 1.02501631,
      "epoch": 0.9553297661275777,
      "flos": 22746860490240.0,
      "grad_norm": 1.7094091734751973,
      "language_loss": 0.81030595,
      "learning_rate": 2.084372449785654e-08,
      "loss": 0.83222818,
      "num_input_tokens_seen": 171539895,
      "step": 7945,
      "time_per_iteration": 2.730043888092041
    },
    {
      "auxiliary_loss_clip": 0.01116473,
      "auxiliary_loss_mlp": 0.01029505,
      "balance_loss_clip": 1.04455531,
      "balance_loss_mlp": 1.02183378,
      "epoch": 0.9554500090182168,
      "flos": 15413866018560.0,
      "grad_norm": 3.8261509627202375,
      "language_loss": 0.68805188,
      "learning_rate": 2.0731705110895282e-08,
      "loss": 0.70951164,
      "num_input_tokens_seen": 171557385,
      "step": 7946,
      "time_per_iteration": 2.7446048259735107
    },
    {
      "auxiliary_loss_clip": 0.01153019,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.05088007,
      "balance_loss_mlp": 1.02044535,
      "epoch": 0.9555702519088559,
      "flos": 23513517400320.0,
      "grad_norm": 2.1810661253492243,
      "language_loss": 0.87044561,
      "learning_rate": 2.0619985977360587e-08,
      "loss": 0.89225799,
      "num_input_tokens_seen": 171575705,
      "step": 7947,
      "time_per_iteration": 2.7508299350738525
    },
    {
      "auxiliary_loss_clip": 0.01155111,
      "auxiliary_loss_mlp": 0.01031254,
      "balance_loss_clip": 1.04827416,
      "balance_loss_mlp": 1.02421474,
      "epoch": 0.955690494799495,
      "flos": 22962072827520.0,
      "grad_norm": 1.6802091498471865,
      "language_loss": 0.76767665,
      "learning_rate": 2.0508567114200237e-08,
      "loss": 0.78954029,
      "num_input_tokens_seen": 171595620,
      "step": 7948,
      "time_per_iteration": 2.6969692707061768
    },
    {
      "auxiliary_loss_clip": 0.0112791,
      "auxiliary_loss_mlp": 0.01025018,
      "balance_loss_clip": 1.04666317,
      "balance_loss_mlp": 1.01751113,
      "epoch": 0.955810737690134,
      "flos": 26031250333440.0,
      "grad_norm": 2.5432341760769552,
      "language_loss": 0.78601354,
      "learning_rate": 2.0397448538316485e-08,
      "loss": 0.8075428,
      "num_input_tokens_seen": 171616660,
      "step": 7949,
      "time_per_iteration": 2.799800157546997
    },
    {
      "auxiliary_loss_clip": 0.01137793,
      "auxiliary_loss_mlp": 0.01035873,
      "balance_loss_clip": 1.04707479,
      "balance_loss_mlp": 1.02818418,
      "epoch": 0.9559309805807732,
      "flos": 20849951249280.0,
      "grad_norm": 3.00852555792906,
      "language_loss": 0.6674695,
      "learning_rate": 2.028663026656563e-08,
      "loss": 0.68920618,
      "num_input_tokens_seen": 171635515,
      "step": 7950,
      "time_per_iteration": 2.7436535358428955
    },
    {
      "auxiliary_loss_clip": 0.01133389,
      "auxiliary_loss_mlp": 0.01025656,
      "balance_loss_clip": 1.04242325,
      "balance_loss_mlp": 1.01822972,
      "epoch": 0.9560512234714122,
      "flos": 21578219498880.0,
      "grad_norm": 2.085758941909113,
      "language_loss": 0.72094548,
      "learning_rate": 2.0176112315758885e-08,
      "loss": 0.74253589,
      "num_input_tokens_seen": 171653305,
      "step": 7951,
      "time_per_iteration": 3.754347324371338
    },
    {
      "auxiliary_loss_clip": 0.0116147,
      "auxiliary_loss_mlp": 0.01029622,
      "balance_loss_clip": 1.04914773,
      "balance_loss_mlp": 1.02207315,
      "epoch": 0.9561714663620513,
      "flos": 17450144029440.0,
      "grad_norm": 3.747846970469251,
      "language_loss": 0.69315195,
      "learning_rate": 2.0065894702661957e-08,
      "loss": 0.71506292,
      "num_input_tokens_seen": 171669980,
      "step": 7952,
      "time_per_iteration": 2.636845588684082
    },
    {
      "auxiliary_loss_clip": 0.0113935,
      "auxiliary_loss_mlp": 0.01028828,
      "balance_loss_clip": 1.04388928,
      "balance_loss_mlp": 1.02064466,
      "epoch": 0.9562917092526905,
      "flos": 26098510550400.0,
      "grad_norm": 1.8894203232138154,
      "language_loss": 0.78157485,
      "learning_rate": 1.9955977443994577e-08,
      "loss": 0.80325663,
      "num_input_tokens_seen": 171689970,
      "step": 7953,
      "time_per_iteration": 2.775235414505005
    },
    {
      "auxiliary_loss_clip": 0.01149886,
      "auxiliary_loss_mlp": 0.01033366,
      "balance_loss_clip": 1.05185378,
      "balance_loss_mlp": 1.02437186,
      "epoch": 0.9564119521433295,
      "flos": 24096742531200.0,
      "grad_norm": 2.746884245660004,
      "language_loss": 0.62295508,
      "learning_rate": 1.9846360556430965e-08,
      "loss": 0.64478761,
      "num_input_tokens_seen": 171708270,
      "step": 7954,
      "time_per_iteration": 2.71130108833313
    },
    {
      "auxiliary_loss_clip": 0.01171659,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.05036569,
      "balance_loss_mlp": 1.01835859,
      "epoch": 0.9565321950339686,
      "flos": 32008903896960.0,
      "grad_norm": 2.2683615255966147,
      "language_loss": 0.61701667,
      "learning_rate": 1.973704405660004e-08,
      "loss": 0.63899422,
      "num_input_tokens_seen": 171729385,
      "step": 7955,
      "time_per_iteration": 2.755420207977295
    },
    {
      "auxiliary_loss_clip": 0.01147852,
      "auxiliary_loss_mlp": 0.01025968,
      "balance_loss_clip": 1.04621911,
      "balance_loss_mlp": 1.01893473,
      "epoch": 0.9566524379246077,
      "flos": 23588642695680.0,
      "grad_norm": 1.5493206506250774,
      "language_loss": 0.77974331,
      "learning_rate": 1.9628027961085203e-08,
      "loss": 0.8014816,
      "num_input_tokens_seen": 171752615,
      "step": 7956,
      "time_per_iteration": 2.7151176929473877
    },
    {
      "auxiliary_loss_clip": 0.01092893,
      "auxiliary_loss_mlp": 0.01026041,
      "balance_loss_clip": 1.03970051,
      "balance_loss_mlp": 1.01854873,
      "epoch": 0.9567726808152468,
      "flos": 38067716240640.0,
      "grad_norm": 1.9724582194284344,
      "language_loss": 0.84225315,
      "learning_rate": 1.9519312286423894e-08,
      "loss": 0.86344254,
      "num_input_tokens_seen": 171775810,
      "step": 7957,
      "time_per_iteration": 2.969426393508911
    },
    {
      "auxiliary_loss_clip": 0.01131259,
      "auxiliary_loss_mlp": 0.00761605,
      "balance_loss_clip": 1.05129051,
      "balance_loss_mlp": 1.00046813,
      "epoch": 0.9568929237058859,
      "flos": 22744059229440.0,
      "grad_norm": 1.8056956661248975,
      "language_loss": 0.77887875,
      "learning_rate": 1.9410897049108255e-08,
      "loss": 0.7978074,
      "num_input_tokens_seen": 171795090,
      "step": 7958,
      "time_per_iteration": 2.7118685245513916
    },
    {
      "auxiliary_loss_clip": 0.01164634,
      "auxiliary_loss_mlp": 0.01028972,
      "balance_loss_clip": 1.05036592,
      "balance_loss_mlp": 1.02076423,
      "epoch": 0.957013166596525,
      "flos": 23841633162240.0,
      "grad_norm": 1.9042550015531239,
      "language_loss": 0.91178912,
      "learning_rate": 1.9302782265584905e-08,
      "loss": 0.93372518,
      "num_input_tokens_seen": 171815755,
      "step": 7959,
      "time_per_iteration": 2.7105555534362793
    },
    {
      "auxiliary_loss_clip": 0.01156594,
      "auxiliary_loss_mlp": 0.01028703,
      "balance_loss_clip": 1.05026054,
      "balance_loss_mlp": 1.02078795,
      "epoch": 0.9571334094871641,
      "flos": 17639286071040.0,
      "grad_norm": 2.658434806122567,
      "language_loss": 0.86593682,
      "learning_rate": 1.9194967952254282e-08,
      "loss": 0.88778973,
      "num_input_tokens_seen": 171834330,
      "step": 7960,
      "time_per_iteration": 3.6724817752838135
    },
    {
      "auxiliary_loss_clip": 0.01136542,
      "auxiliary_loss_mlp": 0.00761827,
      "balance_loss_clip": 1.04858589,
      "balance_loss_mlp": 1.0004679,
      "epoch": 0.9572536523778031,
      "flos": 15369623441280.0,
      "grad_norm": 2.4399936842856356,
      "language_loss": 0.80720979,
      "learning_rate": 1.9087454125472635e-08,
      "loss": 0.82619345,
      "num_input_tokens_seen": 171848805,
      "step": 7961,
      "time_per_iteration": 3.6970956325531006
    },
    {
      "auxiliary_loss_clip": 0.01173441,
      "auxiliary_loss_mlp": 0.01027429,
      "balance_loss_clip": 1.05111158,
      "balance_loss_mlp": 1.0193882,
      "epoch": 0.9573738952684423,
      "flos": 24969838417920.0,
      "grad_norm": 3.5123992418845025,
      "language_loss": 0.78452069,
      "learning_rate": 1.8980240801548696e-08,
      "loss": 0.80652934,
      "num_input_tokens_seen": 171867995,
      "step": 7962,
      "time_per_iteration": 2.7131738662719727
    },
    {
      "auxiliary_loss_clip": 0.01112968,
      "auxiliary_loss_mlp": 0.01027707,
      "balance_loss_clip": 1.04765379,
      "balance_loss_mlp": 1.02057576,
      "epoch": 0.9574941381590814,
      "flos": 25769461034880.0,
      "grad_norm": 1.8428061341064035,
      "language_loss": 0.7407608,
      "learning_rate": 1.8873327996747458e-08,
      "loss": 0.76216757,
      "num_input_tokens_seen": 171886495,
      "step": 7963,
      "time_per_iteration": 2.8269245624542236
    },
    {
      "auxiliary_loss_clip": 0.01152102,
      "auxiliary_loss_mlp": 0.01033205,
      "balance_loss_clip": 1.04644918,
      "balance_loss_mlp": 1.02578449,
      "epoch": 0.9576143810497204,
      "flos": 32307178435200.0,
      "grad_norm": 1.8903083359984683,
      "language_loss": 0.66149664,
      "learning_rate": 1.8766715727287053e-08,
      "loss": 0.68334973,
      "num_input_tokens_seen": 171908200,
      "step": 7964,
      "time_per_iteration": 2.7738397121429443
    },
    {
      "auxiliary_loss_clip": 0.01131119,
      "auxiliary_loss_mlp": 0.01025658,
      "balance_loss_clip": 1.04411685,
      "balance_loss_mlp": 1.01711118,
      "epoch": 0.9577346239403596,
      "flos": 27745733376000.0,
      "grad_norm": 2.2440847206485324,
      "language_loss": 0.79494953,
      "learning_rate": 1.8660404009340546e-08,
      "loss": 0.81651735,
      "num_input_tokens_seen": 171928650,
      "step": 7965,
      "time_per_iteration": 2.772570848464966
    },
    {
      "auxiliary_loss_clip": 0.01031989,
      "auxiliary_loss_mlp": 0.01001034,
      "balance_loss_clip": 1.01010442,
      "balance_loss_mlp": 1.00002086,
      "epoch": 0.9578548668309986,
      "flos": 57468313710720.0,
      "grad_norm": 0.876073601504908,
      "language_loss": 0.59462792,
      "learning_rate": 1.8554392859035485e-08,
      "loss": 0.61495817,
      "num_input_tokens_seen": 171986400,
      "step": 7966,
      "time_per_iteration": 3.2674636840820312
    },
    {
      "auxiliary_loss_clip": 0.01133651,
      "auxiliary_loss_mlp": 0.01030529,
      "balance_loss_clip": 1.04508555,
      "balance_loss_mlp": 1.0225718,
      "epoch": 0.9579751097216377,
      "flos": 19756040503680.0,
      "grad_norm": 1.760375513652797,
      "language_loss": 0.7908864,
      "learning_rate": 1.8448682292453444e-08,
      "loss": 0.81252819,
      "num_input_tokens_seen": 172005475,
      "step": 7967,
      "time_per_iteration": 3.7628793716430664
    },
    {
      "auxiliary_loss_clip": 0.01117518,
      "auxiliary_loss_mlp": 0.01029317,
      "balance_loss_clip": 1.04813433,
      "balance_loss_mlp": 1.02185702,
      "epoch": 0.9580953526122769,
      "flos": 18041270152320.0,
      "grad_norm": 2.1601146832547427,
      "language_loss": 0.65921372,
      "learning_rate": 1.8343272325631154e-08,
      "loss": 0.68068206,
      "num_input_tokens_seen": 172024420,
      "step": 7968,
      "time_per_iteration": 2.7851130962371826
    },
    {
      "auxiliary_loss_clip": 0.011272,
      "auxiliary_loss_mlp": 0.01038071,
      "balance_loss_clip": 1.04373419,
      "balance_loss_mlp": 1.02998531,
      "epoch": 0.9582155955029159,
      "flos": 24270154416000.0,
      "grad_norm": 2.3914173202719553,
      "language_loss": 0.78293598,
      "learning_rate": 1.8238162974558492e-08,
      "loss": 0.80458868,
      "num_input_tokens_seen": 172038350,
      "step": 7969,
      "time_per_iteration": 2.6632847785949707
    },
    {
      "auxiliary_loss_clip": 0.01141114,
      "auxiliary_loss_mlp": 0.01031437,
      "balance_loss_clip": 1.04933357,
      "balance_loss_mlp": 1.02382207,
      "epoch": 0.958335838393555,
      "flos": 22783309816320.0,
      "grad_norm": 2.069257739896494,
      "language_loss": 0.74874139,
      "learning_rate": 1.8133354255181144e-08,
      "loss": 0.77046692,
      "num_input_tokens_seen": 172058665,
      "step": 7970,
      "time_per_iteration": 2.710623025894165
    },
    {
      "auxiliary_loss_clip": 0.01142139,
      "auxiliary_loss_mlp": 0.01027185,
      "balance_loss_clip": 1.04972434,
      "balance_loss_mlp": 1.01957893,
      "epoch": 0.958456081284194,
      "flos": 16911484698240.0,
      "grad_norm": 1.9424368847095623,
      "language_loss": 0.74587613,
      "learning_rate": 1.802884618339795e-08,
      "loss": 0.76756942,
      "num_input_tokens_seen": 172077470,
      "step": 7971,
      "time_per_iteration": 2.6757824420928955
    },
    {
      "auxiliary_loss_clip": 0.01163263,
      "auxiliary_loss_mlp": 0.01026973,
      "balance_loss_clip": 1.05096936,
      "balance_loss_mlp": 1.01827633,
      "epoch": 0.9585763241748332,
      "flos": 19974951941760.0,
      "grad_norm": 2.1039369446135945,
      "language_loss": 0.81403965,
      "learning_rate": 1.7924638775062894e-08,
      "loss": 0.83594203,
      "num_input_tokens_seen": 172096590,
      "step": 7972,
      "time_per_iteration": 2.6553266048431396
    },
    {
      "auxiliary_loss_clip": 0.01120269,
      "auxiliary_loss_mlp": 0.01026568,
      "balance_loss_clip": 1.0432539,
      "balance_loss_mlp": 1.01910806,
      "epoch": 0.9586965670654722,
      "flos": 21395649646080.0,
      "grad_norm": 3.8584709639370525,
      "language_loss": 0.81720823,
      "learning_rate": 1.7820732045984444e-08,
      "loss": 0.83867657,
      "num_input_tokens_seen": 172116735,
      "step": 7973,
      "time_per_iteration": 2.7146406173706055
    },
    {
      "auxiliary_loss_clip": 0.01163838,
      "auxiliary_loss_mlp": 0.01033285,
      "balance_loss_clip": 1.04999888,
      "balance_loss_mlp": 1.02514279,
      "epoch": 0.9588168099561113,
      "flos": 21435115714560.0,
      "grad_norm": 1.9472488965690815,
      "language_loss": 0.73919272,
      "learning_rate": 1.7717126011924655e-08,
      "loss": 0.76116395,
      "num_input_tokens_seen": 172138320,
      "step": 7974,
      "time_per_iteration": 2.715442180633545
    },
    {
      "auxiliary_loss_clip": 0.01136483,
      "auxiliary_loss_mlp": 0.01028802,
      "balance_loss_clip": 1.04801989,
      "balance_loss_mlp": 1.02107167,
      "epoch": 0.9589370528467505,
      "flos": 11763761852160.0,
      "grad_norm": 5.229811959866665,
      "language_loss": 0.77175903,
      "learning_rate": 1.7613820688600957e-08,
      "loss": 0.79341185,
      "num_input_tokens_seen": 172154225,
      "step": 7975,
      "time_per_iteration": 2.642277479171753
    },
    {
      "auxiliary_loss_clip": 0.0112822,
      "auxiliary_loss_mlp": 0.01030913,
      "balance_loss_clip": 1.04682553,
      "balance_loss_mlp": 1.02329564,
      "epoch": 0.9590572957373895,
      "flos": 23441516588160.0,
      "grad_norm": 2.375932309311876,
      "language_loss": 0.7888695,
      "learning_rate": 1.7510816091684588e-08,
      "loss": 0.81046075,
      "num_input_tokens_seen": 172174150,
      "step": 7976,
      "time_per_iteration": 2.81553316116333
    },
    {
      "auxiliary_loss_clip": 0.01177013,
      "auxiliary_loss_mlp": 0.01025261,
      "balance_loss_clip": 1.05225646,
      "balance_loss_mlp": 1.01714885,
      "epoch": 0.9591775386280286,
      "flos": 22528272274560.0,
      "grad_norm": 3.2215795841207093,
      "language_loss": 0.78621954,
      "learning_rate": 1.740811223680083e-08,
      "loss": 0.8082422,
      "num_input_tokens_seen": 172191005,
      "step": 7977,
      "time_per_iteration": 3.568770170211792
    },
    {
      "auxiliary_loss_clip": 0.01157996,
      "auxiliary_loss_mlp": 0.0103105,
      "balance_loss_clip": 1.04754162,
      "balance_loss_mlp": 1.02306914,
      "epoch": 0.9592977815186677,
      "flos": 18186959715840.0,
      "grad_norm": 2.6457695215911174,
      "language_loss": 0.74479258,
      "learning_rate": 1.7305709139530334e-08,
      "loss": 0.76668304,
      "num_input_tokens_seen": 172209785,
      "step": 7978,
      "time_per_iteration": 2.6499791145324707
    },
    {
      "auxiliary_loss_clip": 0.01097041,
      "auxiliary_loss_mlp": 0.01031631,
      "balance_loss_clip": 1.04044032,
      "balance_loss_mlp": 1.02368891,
      "epoch": 0.9594180244093068,
      "flos": 16537797555840.0,
      "grad_norm": 5.051181066749932,
      "language_loss": 0.74663341,
      "learning_rate": 1.7203606815407334e-08,
      "loss": 0.76792014,
      "num_input_tokens_seen": 172224380,
      "step": 7979,
      "time_per_iteration": 2.8079311847686768
    },
    {
      "auxiliary_loss_clip": 0.011553,
      "auxiliary_loss_mlp": 0.0103671,
      "balance_loss_clip": 1.04966307,
      "balance_loss_mlp": 1.02798438,
      "epoch": 0.9595382672999458,
      "flos": 20554334317440.0,
      "grad_norm": 1.834149166737044,
      "language_loss": 0.79505575,
      "learning_rate": 1.7101805279920557e-08,
      "loss": 0.81697583,
      "num_input_tokens_seen": 172242540,
      "step": 7980,
      "time_per_iteration": 2.6004998683929443
    },
    {
      "auxiliary_loss_clip": 0.01143077,
      "auxiliary_loss_mlp": 0.01034809,
      "balance_loss_clip": 1.05021095,
      "balance_loss_mlp": 1.02658987,
      "epoch": 0.959658510190585,
      "flos": 22638266697600.0,
      "grad_norm": 2.168128735229961,
      "language_loss": 0.81478459,
      "learning_rate": 1.7000304548513643e-08,
      "loss": 0.83656347,
      "num_input_tokens_seen": 172262645,
      "step": 7981,
      "time_per_iteration": 2.734168767929077
    },
    {
      "auxiliary_loss_clip": 0.01114896,
      "auxiliary_loss_mlp": 0.01027702,
      "balance_loss_clip": 1.04567385,
      "balance_loss_mlp": 1.01908326,
      "epoch": 0.9597787530812241,
      "flos": 19135252725120.0,
      "grad_norm": 2.5484742713515876,
      "language_loss": 0.82744086,
      "learning_rate": 1.6899104636583394e-08,
      "loss": 0.84886688,
      "num_input_tokens_seen": 172280695,
      "step": 7982,
      "time_per_iteration": 2.7218539714813232
    },
    {
      "auxiliary_loss_clip": 0.01037088,
      "auxiliary_loss_mlp": 0.00751873,
      "balance_loss_clip": 1.00868154,
      "balance_loss_mlp": 1.00036836,
      "epoch": 0.9598989959718631,
      "flos": 60098124055680.0,
      "grad_norm": 0.777263545516287,
      "language_loss": 0.61813879,
      "learning_rate": 1.6798205559482638e-08,
      "loss": 0.63602829,
      "num_input_tokens_seen": 172343075,
      "step": 7983,
      "time_per_iteration": 3.4421067237854004
    },
    {
      "auxiliary_loss_clip": 0.01146344,
      "auxiliary_loss_mlp": 0.01026425,
      "balance_loss_clip": 1.04724026,
      "balance_loss_mlp": 1.01852405,
      "epoch": 0.9600192388625023,
      "flos": 20886795624960.0,
      "grad_norm": 2.457302718597518,
      "language_loss": 0.76602221,
      "learning_rate": 1.669760733251713e-08,
      "loss": 0.78774989,
      "num_input_tokens_seen": 172361950,
      "step": 7984,
      "time_per_iteration": 2.729644775390625
    },
    {
      "auxiliary_loss_clip": 0.01158007,
      "auxiliary_loss_mlp": 0.01025966,
      "balance_loss_clip": 1.05054808,
      "balance_loss_mlp": 1.01854515,
      "epoch": 0.9601394817531413,
      "flos": 20445740524800.0,
      "grad_norm": 1.7741185925391474,
      "language_loss": 0.82403719,
      "learning_rate": 1.659730997094755e-08,
      "loss": 0.84587693,
      "num_input_tokens_seen": 172380440,
      "step": 7985,
      "time_per_iteration": 3.6675615310668945
    },
    {
      "auxiliary_loss_clip": 0.01141867,
      "auxiliary_loss_mlp": 0.01023228,
      "balance_loss_clip": 1.04611397,
      "balance_loss_mlp": 1.01533031,
      "epoch": 0.9602597246437804,
      "flos": 21507152440320.0,
      "grad_norm": 2.141853074360302,
      "language_loss": 0.62457609,
      "learning_rate": 1.6497313489989283e-08,
      "loss": 0.646227,
      "num_input_tokens_seen": 172400265,
      "step": 7986,
      "time_per_iteration": 2.7321579456329346
    },
    {
      "auxiliary_loss_clip": 0.01143152,
      "auxiliary_loss_mlp": 0.01027327,
      "balance_loss_clip": 1.04700148,
      "balance_loss_mlp": 1.01865494,
      "epoch": 0.9603799675344196,
      "flos": 29935099152000.0,
      "grad_norm": 2.098331566782705,
      "language_loss": 0.69774014,
      "learning_rate": 1.639761790481131e-08,
      "loss": 0.71944487,
      "num_input_tokens_seen": 172421145,
      "step": 7987,
      "time_per_iteration": 3.705937385559082
    },
    {
      "auxiliary_loss_clip": 0.0114487,
      "auxiliary_loss_mlp": 0.01033676,
      "balance_loss_clip": 1.04996514,
      "balance_loss_mlp": 1.02658629,
      "epoch": 0.9605002104250586,
      "flos": 28001525103360.0,
      "grad_norm": 3.9722443756121923,
      "language_loss": 0.79211879,
      "learning_rate": 1.6298223230537754e-08,
      "loss": 0.81390423,
      "num_input_tokens_seen": 172438945,
      "step": 7988,
      "time_per_iteration": 2.7241201400756836
    },
    {
      "auxiliary_loss_clip": 0.01116681,
      "auxiliary_loss_mlp": 0.01035988,
      "balance_loss_clip": 1.03891182,
      "balance_loss_mlp": 1.02754188,
      "epoch": 0.9606204533156977,
      "flos": 35590490870400.0,
      "grad_norm": 2.095128867371984,
      "language_loss": 0.69808501,
      "learning_rate": 1.619912948224611e-08,
      "loss": 0.71961176,
      "num_input_tokens_seen": 172460150,
      "step": 7989,
      "time_per_iteration": 2.840883493423462
    },
    {
      "auxiliary_loss_clip": 0.01142276,
      "auxiliary_loss_mlp": 0.01030225,
      "balance_loss_clip": 1.04933476,
      "balance_loss_mlp": 1.02192235,
      "epoch": 0.9607406962063368,
      "flos": 26574614346240.0,
      "grad_norm": 2.6214066149095685,
      "language_loss": 0.60851413,
      "learning_rate": 1.6100336674969682e-08,
      "loss": 0.63023913,
      "num_input_tokens_seen": 172478990,
      "step": 7990,
      "time_per_iteration": 2.70310640335083
    },
    {
      "auxiliary_loss_clip": 0.01125278,
      "auxiliary_loss_mlp": 0.01030013,
      "balance_loss_clip": 1.04541802,
      "balance_loss_mlp": 1.02217817,
      "epoch": 0.9608609390969759,
      "flos": 25331781813120.0,
      "grad_norm": 1.957852030679701,
      "language_loss": 0.76845092,
      "learning_rate": 1.600184482369449e-08,
      "loss": 0.79000384,
      "num_input_tokens_seen": 172498905,
      "step": 7991,
      "time_per_iteration": 2.7542364597320557
    },
    {
      "auxiliary_loss_clip": 0.0116261,
      "auxiliary_loss_mlp": 0.01031358,
      "balance_loss_clip": 1.04852223,
      "balance_loss_mlp": 1.02254295,
      "epoch": 0.960981181987615,
      "flos": 21069114082560.0,
      "grad_norm": 3.023988341188608,
      "language_loss": 0.88903767,
      "learning_rate": 1.5903653943362126e-08,
      "loss": 0.91097736,
      "num_input_tokens_seen": 172517900,
      "step": 7992,
      "time_per_iteration": 2.667965888977051
    },
    {
      "auxiliary_loss_clip": 0.01160232,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.04849434,
      "balance_loss_mlp": 1.02052116,
      "epoch": 0.9611014248782541,
      "flos": 17823256554240.0,
      "grad_norm": 2.0827364328960964,
      "language_loss": 0.77058268,
      "learning_rate": 1.580576404886802e-08,
      "loss": 0.79246771,
      "num_input_tokens_seen": 172536430,
      "step": 7993,
      "time_per_iteration": 3.6806490421295166
    },
    {
      "auxiliary_loss_clip": 0.01129307,
      "auxiliary_loss_mlp": 0.01030823,
      "balance_loss_clip": 1.04224765,
      "balance_loss_mlp": 1.02339947,
      "epoch": 0.9612216677688932,
      "flos": 19354631040000.0,
      "grad_norm": 2.111907314572102,
      "language_loss": 0.79985106,
      "learning_rate": 1.570817515506162e-08,
      "loss": 0.82145238,
      "num_input_tokens_seen": 172555120,
      "step": 7994,
      "time_per_iteration": 2.73506236076355
    },
    {
      "auxiliary_loss_clip": 0.01157203,
      "auxiliary_loss_mlp": 0.01031032,
      "balance_loss_clip": 1.04849613,
      "balance_loss_mlp": 1.0235877,
      "epoch": 0.9613419106595322,
      "flos": 15808739207040.0,
      "grad_norm": 2.43996357458914,
      "language_loss": 0.8134799,
      "learning_rate": 1.561088727674753e-08,
      "loss": 0.83536232,
      "num_input_tokens_seen": 172569330,
      "step": 7995,
      "time_per_iteration": 2.647810935974121
    },
    {
      "auxiliary_loss_clip": 0.01141974,
      "auxiliary_loss_mlp": 0.01027995,
      "balance_loss_clip": 1.04886413,
      "balance_loss_mlp": 1.01953721,
      "epoch": 0.9614621535501714,
      "flos": 25702488126720.0,
      "grad_norm": 3.7194329587829267,
      "language_loss": 0.71492046,
      "learning_rate": 1.551390042868417e-08,
      "loss": 0.73662019,
      "num_input_tokens_seen": 172591100,
      "step": 7996,
      "time_per_iteration": 2.7514753341674805
    },
    {
      "auxiliary_loss_clip": 0.01137689,
      "auxiliary_loss_mlp": 0.00761798,
      "balance_loss_clip": 1.04657233,
      "balance_loss_mlp": 1.00051641,
      "epoch": 0.9615823964408104,
      "flos": 17819054663040.0,
      "grad_norm": 2.1211194902097406,
      "language_loss": 0.71257573,
      "learning_rate": 1.5417214625584207e-08,
      "loss": 0.73157066,
      "num_input_tokens_seen": 172608755,
      "step": 7997,
      "time_per_iteration": 2.688380718231201
    },
    {
      "auxiliary_loss_clip": 0.01119729,
      "auxiliary_loss_mlp": 0.01026187,
      "balance_loss_clip": 1.04050863,
      "balance_loss_mlp": 1.01895142,
      "epoch": 0.9617026393314495,
      "flos": 20190020624640.0,
      "grad_norm": 1.5913398425511387,
      "language_loss": 0.85171163,
      "learning_rate": 1.5320829882114806e-08,
      "loss": 0.87317079,
      "num_input_tokens_seen": 172626830,
      "step": 7998,
      "time_per_iteration": 2.773566961288452
    },
    {
      "auxiliary_loss_clip": 0.01134481,
      "auxiliary_loss_mlp": 0.01026601,
      "balance_loss_clip": 1.04575574,
      "balance_loss_mlp": 1.01889443,
      "epoch": 0.9618228822220887,
      "flos": 20267013427200.0,
      "grad_norm": 2.035007398749175,
      "language_loss": 0.78944063,
      "learning_rate": 1.5224746212897378e-08,
      "loss": 0.81105137,
      "num_input_tokens_seen": 172646125,
      "step": 7999,
      "time_per_iteration": 2.673811197280884
    },
    {
      "auxiliary_loss_clip": 0.01098572,
      "auxiliary_loss_mlp": 0.01025574,
      "balance_loss_clip": 1.03825998,
      "balance_loss_mlp": 1.01802802,
      "epoch": 0.9619431251127277,
      "flos": 21031300039680.0,
      "grad_norm": 1.8901308407416244,
      "language_loss": 0.77411604,
      "learning_rate": 1.512896363250804e-08,
      "loss": 0.79535747,
      "num_input_tokens_seen": 172666235,
      "step": 8000,
      "time_per_iteration": 2.869938611984253
    },
    {
      "auxiliary_loss_clip": 0.01078167,
      "auxiliary_loss_mlp": 0.01028094,
      "balance_loss_clip": 1.03808248,
      "balance_loss_mlp": 1.02102447,
      "epoch": 0.9620633680033668,
      "flos": 22382654538240.0,
      "grad_norm": 1.9045901372612728,
      "language_loss": 0.75730443,
      "learning_rate": 1.503348215547673e-08,
      "loss": 0.77836704,
      "num_input_tokens_seen": 172687325,
      "step": 8001,
      "time_per_iteration": 2.8087189197540283
    },
    {
      "auxiliary_loss_clip": 0.01127528,
      "auxiliary_loss_mlp": 0.01026496,
      "balance_loss_clip": 1.04300165,
      "balance_loss_mlp": 1.01878953,
      "epoch": 0.962183610894006,
      "flos": 18471730740480.0,
      "grad_norm": 1.931555444008266,
      "language_loss": 0.81219673,
      "learning_rate": 1.4938301796288078e-08,
      "loss": 0.83373702,
      "num_input_tokens_seen": 172703895,
      "step": 8002,
      "time_per_iteration": 2.739724636077881
    },
    {
      "auxiliary_loss_clip": 0.01127805,
      "auxiliary_loss_mlp": 0.01026842,
      "balance_loss_clip": 1.04199255,
      "balance_loss_mlp": 1.01871204,
      "epoch": 0.962303853784645,
      "flos": 18435245500800.0,
      "grad_norm": 5.046631746766942,
      "language_loss": 0.81865883,
      "learning_rate": 1.4843422569380537e-08,
      "loss": 0.84020531,
      "num_input_tokens_seen": 172720650,
      "step": 8003,
      "time_per_iteration": 3.7036356925964355
    },
    {
      "auxiliary_loss_clip": 0.01127991,
      "auxiliary_loss_mlp": 0.01022584,
      "balance_loss_clip": 1.04427612,
      "balance_loss_mlp": 1.01519299,
      "epoch": 0.9624240966752841,
      "flos": 26391074826240.0,
      "grad_norm": 1.9150359424782128,
      "language_loss": 0.83235264,
      "learning_rate": 1.4748844489147483e-08,
      "loss": 0.85385835,
      "num_input_tokens_seen": 172737640,
      "step": 8004,
      "time_per_iteration": 2.7520804405212402
    },
    {
      "auxiliary_loss_clip": 0.01111516,
      "auxiliary_loss_mlp": 0.01028074,
      "balance_loss_clip": 1.0409317,
      "balance_loss_mlp": 1.02076316,
      "epoch": 0.9625443395659231,
      "flos": 14647675985280.0,
      "grad_norm": 1.970005276588587,
      "language_loss": 0.7108264,
      "learning_rate": 1.4654567569936326e-08,
      "loss": 0.73222232,
      "num_input_tokens_seen": 172755215,
      "step": 8005,
      "time_per_iteration": 2.813217878341675
    },
    {
      "auxiliary_loss_clip": 0.01136266,
      "auxiliary_loss_mlp": 0.01031669,
      "balance_loss_clip": 1.04897678,
      "balance_loss_mlp": 1.02349389,
      "epoch": 0.9626645824565623,
      "flos": 18367626147840.0,
      "grad_norm": 2.127518693511871,
      "language_loss": 0.83130062,
      "learning_rate": 1.456059182604874e-08,
      "loss": 0.85298002,
      "num_input_tokens_seen": 172774020,
      "step": 8006,
      "time_per_iteration": 2.7915072441101074
    },
    {
      "auxiliary_loss_clip": 0.01118825,
      "auxiliary_loss_mlp": 0.01032821,
      "balance_loss_clip": 1.0431267,
      "balance_loss_mlp": 1.02432775,
      "epoch": 0.9627848253472013,
      "flos": 16580424021120.0,
      "grad_norm": 2.214319720065825,
      "language_loss": 0.76968658,
      "learning_rate": 1.4466917271740653e-08,
      "loss": 0.79120308,
      "num_input_tokens_seen": 172792220,
      "step": 8007,
      "time_per_iteration": 2.832508087158203
    },
    {
      "auxiliary_loss_clip": 0.01128027,
      "auxiliary_loss_mlp": 0.01024628,
      "balance_loss_clip": 1.04416454,
      "balance_loss_mlp": 1.01546645,
      "epoch": 0.9629050682378404,
      "flos": 20886867452160.0,
      "grad_norm": 2.128793953435502,
      "language_loss": 0.68317813,
      "learning_rate": 1.4373543921222697e-08,
      "loss": 0.7047047,
      "num_input_tokens_seen": 172811805,
      "step": 8008,
      "time_per_iteration": 2.750016450881958
    },
    {
      "auxiliary_loss_clip": 0.01107986,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.03999913,
      "balance_loss_mlp": 1.01866734,
      "epoch": 0.9630253111284796,
      "flos": 17019252478080.0,
      "grad_norm": 2.008852921110829,
      "language_loss": 0.78208339,
      "learning_rate": 1.428047178865932e-08,
      "loss": 0.80342758,
      "num_input_tokens_seen": 172828595,
      "step": 8009,
      "time_per_iteration": 2.761258602142334
    },
    {
      "auxiliary_loss_clip": 0.0110654,
      "auxiliary_loss_mlp": 0.01030982,
      "balance_loss_clip": 1.04276145,
      "balance_loss_mlp": 1.02318549,
      "epoch": 0.9631455540191186,
      "flos": 20338942412160.0,
      "grad_norm": 1.8217767277856503,
      "language_loss": 0.74574888,
      "learning_rate": 1.4187700888169451e-08,
      "loss": 0.76712406,
      "num_input_tokens_seen": 172847770,
      "step": 8010,
      "time_per_iteration": 2.846109390258789
    },
    {
      "auxiliary_loss_clip": 0.010405,
      "auxiliary_loss_mlp": 0.01000757,
      "balance_loss_clip": 1.00819576,
      "balance_loss_mlp": 0.99980325,
      "epoch": 0.9632657969097577,
      "flos": 65956700033280.0,
      "grad_norm": 0.7870183259762188,
      "language_loss": 0.56970668,
      "learning_rate": 1.40952312338265e-08,
      "loss": 0.59011924,
      "num_input_tokens_seen": 172912415,
      "step": 8011,
      "time_per_iteration": 3.316316604614258
    },
    {
      "auxiliary_loss_clip": 0.01157633,
      "auxiliary_loss_mlp": 0.01026654,
      "balance_loss_clip": 1.04841685,
      "balance_loss_mlp": 1.01923347,
      "epoch": 0.9633860398003968,
      "flos": 44419523823360.0,
      "grad_norm": 2.3484952194415833,
      "language_loss": 0.68466121,
      "learning_rate": 1.4003062839657909e-08,
      "loss": 0.70650411,
      "num_input_tokens_seen": 172934895,
      "step": 8012,
      "time_per_iteration": 3.792357921600342
    },
    {
      "auxiliary_loss_clip": 0.01143029,
      "auxiliary_loss_mlp": 0.01029773,
      "balance_loss_clip": 1.04700649,
      "balance_loss_mlp": 1.02197933,
      "epoch": 0.9635062826910359,
      "flos": 24827704300800.0,
      "grad_norm": 1.5716166178703734,
      "language_loss": 0.8001796,
      "learning_rate": 1.391119571964583e-08,
      "loss": 0.82190764,
      "num_input_tokens_seen": 172955835,
      "step": 8013,
      "time_per_iteration": 3.680504322052002
    },
    {
      "auxiliary_loss_clip": 0.01081521,
      "auxiliary_loss_mlp": 0.01027253,
      "balance_loss_clip": 1.04307556,
      "balance_loss_mlp": 1.01990056,
      "epoch": 0.9636265255816749,
      "flos": 15961360095360.0,
      "grad_norm": 2.0532868326498517,
      "language_loss": 0.73423374,
      "learning_rate": 1.3819629887726225e-08,
      "loss": 0.75532144,
      "num_input_tokens_seen": 172973925,
      "step": 8014,
      "time_per_iteration": 2.8443210124969482
    },
    {
      "auxiliary_loss_clip": 0.01160698,
      "auxiliary_loss_mlp": 0.01025142,
      "balance_loss_clip": 1.05231571,
      "balance_loss_mlp": 1.01736379,
      "epoch": 0.9637467684723141,
      "flos": 22601781457920.0,
      "grad_norm": 1.9347826016037437,
      "language_loss": 0.7619977,
      "learning_rate": 1.3728365357789317e-08,
      "loss": 0.78385603,
      "num_input_tokens_seen": 172993290,
      "step": 8015,
      "time_per_iteration": 2.740107774734497
    },
    {
      "auxiliary_loss_clip": 0.01112674,
      "auxiliary_loss_mlp": 0.01028025,
      "balance_loss_clip": 1.0492357,
      "balance_loss_mlp": 1.02039528,
      "epoch": 0.9638670113629532,
      "flos": 17565812801280.0,
      "grad_norm": 3.757305386882832,
      "language_loss": 0.77042842,
      "learning_rate": 1.3637402143680254e-08,
      "loss": 0.79183543,
      "num_input_tokens_seen": 173008190,
      "step": 8016,
      "time_per_iteration": 2.790313482284546
    },
    {
      "auxiliary_loss_clip": 0.01041104,
      "auxiliary_loss_mlp": 0.01000568,
      "balance_loss_clip": 1.00852752,
      "balance_loss_mlp": 0.99972802,
      "epoch": 0.9639872542535922,
      "flos": 55072139379840.0,
      "grad_norm": 0.7459750257580262,
      "language_loss": 0.54936433,
      "learning_rate": 1.3546740259197998e-08,
      "loss": 0.56978106,
      "num_input_tokens_seen": 173061000,
      "step": 8017,
      "time_per_iteration": 3.2183687686920166
    },
    {
      "auxiliary_loss_clip": 0.0115193,
      "auxiliary_loss_mlp": 0.00762433,
      "balance_loss_clip": 1.05019522,
      "balance_loss_mlp": 1.00050902,
      "epoch": 0.9641074971442314,
      "flos": 24134484746880.0,
      "grad_norm": 2.5720304001580256,
      "language_loss": 0.69817919,
      "learning_rate": 1.3456379718095989e-08,
      "loss": 0.71732283,
      "num_input_tokens_seen": 173081415,
      "step": 8018,
      "time_per_iteration": 2.7818431854248047
    },
    {
      "auxiliary_loss_clip": 0.01051649,
      "auxiliary_loss_mlp": 0.01000177,
      "balance_loss_clip": 1.0071764,
      "balance_loss_mlp": 0.99927086,
      "epoch": 0.9642277400348704,
      "flos": 66747416077440.0,
      "grad_norm": 0.8519579603667068,
      "language_loss": 0.61948466,
      "learning_rate": 1.3366320534081487e-08,
      "loss": 0.64000285,
      "num_input_tokens_seen": 173144095,
      "step": 8019,
      "time_per_iteration": 4.235757350921631
    },
    {
      "auxiliary_loss_clip": 0.01093396,
      "auxiliary_loss_mlp": 0.01027926,
      "balance_loss_clip": 1.0442152,
      "balance_loss_mlp": 1.01985872,
      "epoch": 0.9643479829255095,
      "flos": 30920272450560.0,
      "grad_norm": 2.234841721648028,
      "language_loss": 0.76126623,
      "learning_rate": 1.3276562720816675e-08,
      "loss": 0.78247941,
      "num_input_tokens_seen": 173165605,
      "step": 8020,
      "time_per_iteration": 2.8984601497650146
    },
    {
      "auxiliary_loss_clip": 0.01155175,
      "auxiliary_loss_mlp": 0.01026679,
      "balance_loss_clip": 1.04790568,
      "balance_loss_mlp": 1.01810169,
      "epoch": 0.9644682258161487,
      "flos": 20048245643520.0,
      "grad_norm": 12.44540419059593,
      "language_loss": 0.82117724,
      "learning_rate": 1.3187106291917549e-08,
      "loss": 0.84299582,
      "num_input_tokens_seen": 173182595,
      "step": 8021,
      "time_per_iteration": 2.6880245208740234
    },
    {
      "auxiliary_loss_clip": 0.01112375,
      "auxiliary_loss_mlp": 0.01027911,
      "balance_loss_clip": 1.04231489,
      "balance_loss_mlp": 1.02034402,
      "epoch": 0.9645884687067877,
      "flos": 21178713456000.0,
      "grad_norm": 1.6579695944283692,
      "language_loss": 0.70485067,
      "learning_rate": 1.309795126095503e-08,
      "loss": 0.72625351,
      "num_input_tokens_seen": 173200895,
      "step": 8022,
      "time_per_iteration": 2.727755069732666
    },
    {
      "auxiliary_loss_clip": 0.01122105,
      "auxiliary_loss_mlp": 0.01030464,
      "balance_loss_clip": 1.04508173,
      "balance_loss_mlp": 1.02285218,
      "epoch": 0.9647087115974268,
      "flos": 18945967029120.0,
      "grad_norm": 2.210516635877813,
      "language_loss": 0.80593729,
      "learning_rate": 1.3009097641453192e-08,
      "loss": 0.82746303,
      "num_input_tokens_seen": 173218745,
      "step": 8023,
      "time_per_iteration": 2.7674365043640137
    },
    {
      "auxiliary_loss_clip": 0.01160397,
      "auxiliary_loss_mlp": 0.01023149,
      "balance_loss_clip": 1.04619241,
      "balance_loss_mlp": 1.01544225,
      "epoch": 0.9648289544880659,
      "flos": 16545088016640.0,
      "grad_norm": 1.8977937893902646,
      "language_loss": 0.76638019,
      "learning_rate": 1.2920545446891474e-08,
      "loss": 0.78821564,
      "num_input_tokens_seen": 173235465,
      "step": 8024,
      "time_per_iteration": 2.6479506492614746
    },
    {
      "auxiliary_loss_clip": 0.01104034,
      "auxiliary_loss_mlp": 0.01026662,
      "balance_loss_clip": 1.0400213,
      "balance_loss_mlp": 1.01849627,
      "epoch": 0.964949197378705,
      "flos": 24057527857920.0,
      "grad_norm": 1.8611643918141973,
      "language_loss": 0.71201968,
      "learning_rate": 1.2832294690703127e-08,
      "loss": 0.73332667,
      "num_input_tokens_seen": 173254440,
      "step": 8025,
      "time_per_iteration": 2.8409552574157715
    },
    {
      "auxiliary_loss_clip": 0.01094617,
      "auxiliary_loss_mlp": 0.01027593,
      "balance_loss_clip": 1.0395422,
      "balance_loss_mlp": 1.01929665,
      "epoch": 0.965069440269344,
      "flos": 23365565280000.0,
      "grad_norm": 3.3035923435817636,
      "language_loss": 0.7763871,
      "learning_rate": 1.2744345386275668e-08,
      "loss": 0.79760915,
      "num_input_tokens_seen": 173273980,
      "step": 8026,
      "time_per_iteration": 2.873460054397583
    },
    {
      "auxiliary_loss_clip": 0.01102796,
      "auxiliary_loss_mlp": 0.01031302,
      "balance_loss_clip": 1.04181194,
      "balance_loss_mlp": 1.02295101,
      "epoch": 0.9651896831599832,
      "flos": 25374875155200.0,
      "grad_norm": 1.6453525895948726,
      "language_loss": 0.78871012,
      "learning_rate": 1.265669754695109e-08,
      "loss": 0.81005108,
      "num_input_tokens_seen": 173293550,
      "step": 8027,
      "time_per_iteration": 2.7917933464050293
    },
    {
      "auxiliary_loss_clip": 0.01140417,
      "auxiliary_loss_mlp": 0.01028678,
      "balance_loss_clip": 1.04390621,
      "balance_loss_mlp": 1.02053607,
      "epoch": 0.9653099260506223,
      "flos": 22272875596800.0,
      "grad_norm": 2.148310897439928,
      "language_loss": 0.82121265,
      "learning_rate": 1.2569351186025201e-08,
      "loss": 0.84290361,
      "num_input_tokens_seen": 173312005,
      "step": 8028,
      "time_per_iteration": 2.6977477073669434
    },
    {
      "auxiliary_loss_clip": 0.01111882,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.04058516,
      "balance_loss_mlp": 1.0177027,
      "epoch": 0.9654301689412613,
      "flos": 26760847386240.0,
      "grad_norm": 1.5042509573660099,
      "language_loss": 0.75455129,
      "learning_rate": 1.2482306316748737e-08,
      "loss": 0.77591783,
      "num_input_tokens_seen": 173332450,
      "step": 8029,
      "time_per_iteration": 3.7208309173583984
    },
    {
      "auxiliary_loss_clip": 0.01100241,
      "auxiliary_loss_mlp": 0.0103169,
      "balance_loss_clip": 1.04161775,
      "balance_loss_mlp": 1.02358961,
      "epoch": 0.9655504118319005,
      "flos": 17412689122560.0,
      "grad_norm": 2.160328744556498,
      "language_loss": 0.78507155,
      "learning_rate": 1.2395562952326021e-08,
      "loss": 0.80639088,
      "num_input_tokens_seen": 173349610,
      "step": 8030,
      "time_per_iteration": 2.779772996902466
    },
    {
      "auxiliary_loss_clip": 0.01123635,
      "auxiliary_loss_mlp": 0.01032295,
      "balance_loss_clip": 1.04922473,
      "balance_loss_mlp": 1.0235033,
      "epoch": 0.9656706547225395,
      "flos": 22126970551680.0,
      "grad_norm": 2.3743093574096643,
      "language_loss": 0.81265569,
      "learning_rate": 1.2309121105916309e-08,
      "loss": 0.83421499,
      "num_input_tokens_seen": 173367900,
      "step": 8031,
      "time_per_iteration": 2.7823338508605957
    },
    {
      "auxiliary_loss_clip": 0.01129588,
      "auxiliary_loss_mlp": 0.01030203,
      "balance_loss_clip": 1.04731309,
      "balance_loss_mlp": 1.02203691,
      "epoch": 0.9657908976131786,
      "flos": 37049289926400.0,
      "grad_norm": 6.247921943091785,
      "language_loss": 0.69328415,
      "learning_rate": 1.222298079063222e-08,
      "loss": 0.71488214,
      "num_input_tokens_seen": 173389040,
      "step": 8032,
      "time_per_iteration": 2.8419229984283447
    },
    {
      "auxiliary_loss_clip": 0.01119587,
      "auxiliary_loss_mlp": 0.00761378,
      "balance_loss_clip": 1.04790902,
      "balance_loss_mlp": 1.00046957,
      "epoch": 0.9659111405038178,
      "flos": 24389809597440.0,
      "grad_norm": 2.396167889785006,
      "language_loss": 0.72656846,
      "learning_rate": 1.2137142019541524e-08,
      "loss": 0.74537814,
      "num_input_tokens_seen": 173407595,
      "step": 8033,
      "time_per_iteration": 2.8348729610443115
    },
    {
      "auxiliary_loss_clip": 0.01149678,
      "auxiliary_loss_mlp": 0.01028355,
      "balance_loss_clip": 1.04780054,
      "balance_loss_mlp": 1.02068388,
      "epoch": 0.9660313833944568,
      "flos": 25009412227200.0,
      "grad_norm": 1.9566727329680322,
      "language_loss": 0.734411,
      "learning_rate": 1.2051604805666027e-08,
      "loss": 0.75619137,
      "num_input_tokens_seen": 173424720,
      "step": 8034,
      "time_per_iteration": 2.719196081161499
    },
    {
      "auxiliary_loss_clip": 0.01141908,
      "auxiliary_loss_mlp": 0.01026447,
      "balance_loss_clip": 1.04857659,
      "balance_loss_mlp": 1.01871061,
      "epoch": 0.9661516262850959,
      "flos": 11801575895040.0,
      "grad_norm": 2.2451803068144573,
      "language_loss": 0.78549767,
      "learning_rate": 1.196636916198135e-08,
      "loss": 0.80718118,
      "num_input_tokens_seen": 173442260,
      "step": 8035,
      "time_per_iteration": 2.730232000350952
    },
    {
      "auxiliary_loss_clip": 0.01142127,
      "auxiliary_loss_mlp": 0.01029103,
      "balance_loss_clip": 1.04932249,
      "balance_loss_mlp": 1.02130973,
      "epoch": 0.9662718691757349,
      "flos": 20047778766720.0,
      "grad_norm": 1.9159022748501693,
      "language_loss": 0.76821983,
      "learning_rate": 1.1881435101418036e-08,
      "loss": 0.78993213,
      "num_input_tokens_seen": 173461675,
      "step": 8036,
      "time_per_iteration": 2.692239999771118
    },
    {
      "auxiliary_loss_clip": 0.01051948,
      "auxiliary_loss_mlp": 0.01002636,
      "balance_loss_clip": 1.00851583,
      "balance_loss_mlp": 1.00174165,
      "epoch": 0.9663921120663741,
      "flos": 68027703517440.0,
      "grad_norm": 0.7242514807925614,
      "language_loss": 0.65468287,
      "learning_rate": 1.1796802636860003e-08,
      "loss": 0.67522871,
      "num_input_tokens_seen": 173530205,
      "step": 8037,
      "time_per_iteration": 4.198557138442993
    },
    {
      "auxiliary_loss_clip": 0.01138693,
      "auxiliary_loss_mlp": 0.01034731,
      "balance_loss_clip": 1.04694939,
      "balance_loss_mlp": 1.02697015,
      "epoch": 0.9665123549570132,
      "flos": 26322916769280.0,
      "grad_norm": 1.9571483478703777,
      "language_loss": 0.73908198,
      "learning_rate": 1.1712471781146316e-08,
      "loss": 0.76081622,
      "num_input_tokens_seen": 173549540,
      "step": 8038,
      "time_per_iteration": 2.8198790550231934
    },
    {
      "auxiliary_loss_clip": 0.01142257,
      "auxiliary_loss_mlp": 0.01028408,
      "balance_loss_clip": 1.04737473,
      "balance_loss_mlp": 1.02049232,
      "epoch": 0.9666325978476522,
      "flos": 43941121557120.0,
      "grad_norm": 2.780209703933532,
      "language_loss": 0.67032933,
      "learning_rate": 1.1628442547069628e-08,
      "loss": 0.69203591,
      "num_input_tokens_seen": 173571740,
      "step": 8039,
      "time_per_iteration": 3.832345962524414
    },
    {
      "auxiliary_loss_clip": 0.01118661,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.04790986,
      "balance_loss_mlp": 1.02164817,
      "epoch": 0.9667528407382914,
      "flos": 21543422198400.0,
      "grad_norm": 2.2466620545711984,
      "language_loss": 0.773229,
      "learning_rate": 1.1544714947377521e-08,
      "loss": 0.79471177,
      "num_input_tokens_seen": 173589425,
      "step": 8040,
      "time_per_iteration": 2.8139772415161133
    },
    {
      "auxiliary_loss_clip": 0.01160326,
      "auxiliary_loss_mlp": 0.01029957,
      "balance_loss_clip": 1.04896319,
      "balance_loss_mlp": 1.02211356,
      "epoch": 0.9668730836289304,
      "flos": 23878585278720.0,
      "grad_norm": 2.176605478219129,
      "language_loss": 0.70599109,
      "learning_rate": 1.1461288994770945e-08,
      "loss": 0.72789389,
      "num_input_tokens_seen": 173608500,
      "step": 8041,
      "time_per_iteration": 2.6542084217071533
    },
    {
      "auxiliary_loss_clip": 0.01084386,
      "auxiliary_loss_mlp": 0.01028569,
      "balance_loss_clip": 1.040874,
      "balance_loss_mlp": 1.02032542,
      "epoch": 0.9669933265195695,
      "flos": 28293011971200.0,
      "grad_norm": 1.9718816052796635,
      "language_loss": 0.77554262,
      "learning_rate": 1.1378164701906002e-08,
      "loss": 0.79667211,
      "num_input_tokens_seen": 173630265,
      "step": 8042,
      "time_per_iteration": 3.0465290546417236
    },
    {
      "auxiliary_loss_clip": 0.0113442,
      "auxiliary_loss_mlp": 0.01020992,
      "balance_loss_clip": 1.04538,
      "balance_loss_mlp": 1.01292777,
      "epoch": 0.9671135694102087,
      "flos": 22454763091200.0,
      "grad_norm": 18.357958549788457,
      "language_loss": 0.6662724,
      "learning_rate": 1.1295342081392156e-08,
      "loss": 0.68782651,
      "num_input_tokens_seen": 173649625,
      "step": 8043,
      "time_per_iteration": 2.7394659519195557
    },
    {
      "auxiliary_loss_clip": 0.01114561,
      "auxiliary_loss_mlp": 0.0076167,
      "balance_loss_clip": 1.04077792,
      "balance_loss_mlp": 1.00044608,
      "epoch": 0.9672338123008477,
      "flos": 20155941596160.0,
      "grad_norm": 1.994885696486569,
      "language_loss": 0.69841194,
      "learning_rate": 1.1212821145793804e-08,
      "loss": 0.71717423,
      "num_input_tokens_seen": 173669240,
      "step": 8044,
      "time_per_iteration": 2.778524160385132
    },
    {
      "auxiliary_loss_clip": 0.01093815,
      "auxiliary_loss_mlp": 0.01031864,
      "balance_loss_clip": 1.04345179,
      "balance_loss_mlp": 1.02437449,
      "epoch": 0.9673540551914868,
      "flos": 16977487939200.0,
      "grad_norm": 1.9734844721951017,
      "language_loss": 0.79000771,
      "learning_rate": 1.1130601907629156e-08,
      "loss": 0.81126451,
      "num_input_tokens_seen": 173686970,
      "step": 8045,
      "time_per_iteration": 2.817418336868286
    },
    {
      "auxiliary_loss_clip": 0.01042881,
      "auxiliary_loss_mlp": 0.01001532,
      "balance_loss_clip": 1.0074954,
      "balance_loss_mlp": 1.00063801,
      "epoch": 0.9674742980821259,
      "flos": 61892903952000.0,
      "grad_norm": 0.8210969051186056,
      "language_loss": 0.64675927,
      "learning_rate": 1.1048684379370899e-08,
      "loss": 0.66720337,
      "num_input_tokens_seen": 173747655,
      "step": 8046,
      "time_per_iteration": 4.149870157241821
    },
    {
      "auxiliary_loss_clip": 0.01144771,
      "auxiliary_loss_mlp": 0.01027171,
      "balance_loss_clip": 1.04891133,
      "balance_loss_mlp": 1.02011943,
      "epoch": 0.967594540972765,
      "flos": 18697824898560.0,
      "grad_norm": 2.4906490333974323,
      "language_loss": 0.74551564,
      "learning_rate": 1.0967068573445759e-08,
      "loss": 0.76723504,
      "num_input_tokens_seen": 173765140,
      "step": 8047,
      "time_per_iteration": 2.7458720207214355
    },
    {
      "auxiliary_loss_clip": 0.01131932,
      "auxiliary_loss_mlp": 0.01027313,
      "balance_loss_clip": 1.04613447,
      "balance_loss_mlp": 1.01951408,
      "epoch": 0.967714783863404,
      "flos": 20777411733120.0,
      "grad_norm": 2.4427128565469025,
      "language_loss": 0.65083551,
      "learning_rate": 1.0885754502234945e-08,
      "loss": 0.67242795,
      "num_input_tokens_seen": 173784800,
      "step": 8048,
      "time_per_iteration": 2.7811267375946045
    },
    {
      "auxiliary_loss_clip": 0.01155989,
      "auxiliary_loss_mlp": 0.01030768,
      "balance_loss_clip": 1.04866004,
      "balance_loss_mlp": 1.02350783,
      "epoch": 0.9678350267540432,
      "flos": 23185473465600.0,
      "grad_norm": 1.8462729557115827,
      "language_loss": 0.78566569,
      "learning_rate": 1.08047421780737e-08,
      "loss": 0.80753326,
      "num_input_tokens_seen": 173803990,
      "step": 8049,
      "time_per_iteration": 2.7116634845733643
    },
    {
      "auxiliary_loss_clip": 0.01112904,
      "auxiliary_loss_mlp": 0.0102371,
      "balance_loss_clip": 1.04848063,
      "balance_loss_mlp": 1.01644421,
      "epoch": 0.9679552696446823,
      "flos": 21726063878400.0,
      "grad_norm": 2.2080679535335315,
      "language_loss": 0.73650599,
      "learning_rate": 1.0724031613251305e-08,
      "loss": 0.7578721,
      "num_input_tokens_seen": 173821890,
      "step": 8050,
      "time_per_iteration": 2.826763868331909
    },
    {
      "auxiliary_loss_clip": 0.0114887,
      "auxiliary_loss_mlp": 0.01028067,
      "balance_loss_clip": 1.04674911,
      "balance_loss_mlp": 1.02007377,
      "epoch": 0.9680755125353213,
      "flos": 26869046129280.0,
      "grad_norm": 2.120462215087544,
      "language_loss": 0.6635462,
      "learning_rate": 1.0643622820011744e-08,
      "loss": 0.68531561,
      "num_input_tokens_seen": 173842945,
      "step": 8051,
      "time_per_iteration": 2.7888338565826416
    },
    {
      "auxiliary_loss_clip": 0.01175822,
      "auxiliary_loss_mlp": 0.01034071,
      "balance_loss_clip": 1.05111623,
      "balance_loss_mlp": 1.02560759,
      "epoch": 0.9681957554259605,
      "flos": 28325008010880.0,
      "grad_norm": 5.056712419018554,
      "language_loss": 0.68372303,
      "learning_rate": 1.0563515810552814e-08,
      "loss": 0.70582199,
      "num_input_tokens_seen": 173859915,
      "step": 8052,
      "time_per_iteration": 2.647719383239746
    },
    {
      "auxiliary_loss_clip": 0.01151976,
      "auxiliary_loss_mlp": 0.01030523,
      "balance_loss_clip": 1.05266142,
      "balance_loss_mlp": 1.02265549,
      "epoch": 0.9683159983165995,
      "flos": 20557674282240.0,
      "grad_norm": 1.5811937115126562,
      "language_loss": 0.73491293,
      "learning_rate": 1.0483710597026795e-08,
      "loss": 0.75673789,
      "num_input_tokens_seen": 173879775,
      "step": 8053,
      "time_per_iteration": 2.7338809967041016
    },
    {
      "auxiliary_loss_clip": 0.0114412,
      "auxiliary_loss_mlp": 0.01024777,
      "balance_loss_clip": 1.05011141,
      "balance_loss_mlp": 1.01751757,
      "epoch": 0.9684362412072386,
      "flos": 24207958016640.0,
      "grad_norm": 2.0012593728784758,
      "language_loss": 0.74165177,
      "learning_rate": 1.0404207191540227e-08,
      "loss": 0.76334071,
      "num_input_tokens_seen": 173900230,
      "step": 8054,
      "time_per_iteration": 2.7580506801605225
    },
    {
      "auxiliary_loss_clip": 0.01129501,
      "auxiliary_loss_mlp": 0.01027747,
      "balance_loss_clip": 1.04369891,
      "balance_loss_mlp": 1.02052283,
      "epoch": 0.9685564840978778,
      "flos": 22346241125760.0,
      "grad_norm": 2.573433652036074,
      "language_loss": 0.74841768,
      "learning_rate": 1.0325005606153236e-08,
      "loss": 0.76999021,
      "num_input_tokens_seen": 173919690,
      "step": 8055,
      "time_per_iteration": 3.7131543159484863
    },
    {
      "auxiliary_loss_clip": 0.01111537,
      "auxiliary_loss_mlp": 0.01030964,
      "balance_loss_clip": 1.04551566,
      "balance_loss_mlp": 1.02341795,
      "epoch": 0.9686767269885168,
      "flos": 14386389477120.0,
      "grad_norm": 2.5936692128432464,
      "language_loss": 0.79680467,
      "learning_rate": 1.0246105852881104e-08,
      "loss": 0.81822962,
      "num_input_tokens_seen": 173934790,
      "step": 8056,
      "time_per_iteration": 2.6801507472991943
    },
    {
      "auxiliary_loss_clip": 0.01141174,
      "auxiliary_loss_mlp": 0.01027998,
      "balance_loss_clip": 1.04670632,
      "balance_loss_mlp": 1.01946259,
      "epoch": 0.9687969698791559,
      "flos": 21287630471040.0,
      "grad_norm": 6.406747270374578,
      "language_loss": 0.79242218,
      "learning_rate": 1.0167507943692476e-08,
      "loss": 0.81411386,
      "num_input_tokens_seen": 173953875,
      "step": 8057,
      "time_per_iteration": 2.728736162185669
    },
    {
      "auxiliary_loss_clip": 0.01115485,
      "auxiliary_loss_mlp": 0.01031752,
      "balance_loss_clip": 1.04260409,
      "balance_loss_mlp": 1.0234611,
      "epoch": 0.968917212769795,
      "flos": 19828328624640.0,
      "grad_norm": 2.548740616695647,
      "language_loss": 0.71517318,
      "learning_rate": 1.008921189051093e-08,
      "loss": 0.73664558,
      "num_input_tokens_seen": 173971220,
      "step": 8058,
      "time_per_iteration": 2.7513034343719482
    },
    {
      "auxiliary_loss_clip": 0.01149524,
      "auxiliary_loss_mlp": 0.01033994,
      "balance_loss_clip": 1.04844403,
      "balance_loss_mlp": 1.02625155,
      "epoch": 0.9690374556604341,
      "flos": 21681749473920.0,
      "grad_norm": 16.730906239489553,
      "language_loss": 0.77256727,
      "learning_rate": 1.0011217705213848e-08,
      "loss": 0.79440248,
      "num_input_tokens_seen": 173989095,
      "step": 8059,
      "time_per_iteration": 2.7058136463165283
    },
    {
      "auxiliary_loss_clip": 0.01115118,
      "auxiliary_loss_mlp": 0.01029098,
      "balance_loss_clip": 1.04794168,
      "balance_loss_mlp": 1.02186203,
      "epoch": 0.9691576985510731,
      "flos": 32635437851520.0,
      "grad_norm": 2.368666044698305,
      "language_loss": 0.74592412,
      "learning_rate": 9.933525399632658e-09,
      "loss": 0.76736629,
      "num_input_tokens_seen": 174007330,
      "step": 8060,
      "time_per_iteration": 2.762254476547241
    },
    {
      "auxiliary_loss_clip": 0.01121419,
      "auxiliary_loss_mlp": 0.01026797,
      "balance_loss_clip": 1.04293644,
      "balance_loss_mlp": 1.01817799,
      "epoch": 0.9692779414417123,
      "flos": 35663174040960.0,
      "grad_norm": 1.8901452609188356,
      "language_loss": 0.65360498,
      "learning_rate": 9.856134985553488e-09,
      "loss": 0.67508709,
      "num_input_tokens_seen": 174027055,
      "step": 8061,
      "time_per_iteration": 2.8621606826782227
    },
    {
      "auxiliary_loss_clip": 0.01148789,
      "auxiliary_loss_mlp": 0.01028512,
      "balance_loss_clip": 1.05163562,
      "balance_loss_mlp": 1.02094245,
      "epoch": 0.9693981843323514,
      "flos": 28366952117760.0,
      "grad_norm": 1.6066010918095426,
      "language_loss": 0.73987031,
      "learning_rate": 9.77904647471628e-09,
      "loss": 0.76164335,
      "num_input_tokens_seen": 174050235,
      "step": 8062,
      "time_per_iteration": 2.757634401321411
    },
    {
      "auxiliary_loss_clip": 0.0116163,
      "auxiliary_loss_mlp": 0.00761392,
      "balance_loss_clip": 1.04834449,
      "balance_loss_mlp": 1.00045347,
      "epoch": 0.9695184272229904,
      "flos": 23622865378560.0,
      "grad_norm": 1.5357203971692293,
      "language_loss": 0.74151957,
      "learning_rate": 9.702259878815454e-09,
      "loss": 0.76074982,
      "num_input_tokens_seen": 174070560,
      "step": 8063,
      "time_per_iteration": 3.54995059967041
    },
    {
      "auxiliary_loss_clip": 0.01146395,
      "auxiliary_loss_mlp": 0.01026212,
      "balance_loss_clip": 1.04958618,
      "balance_loss_mlp": 1.01797485,
      "epoch": 0.9696386701136296,
      "flos": 23294677789440.0,
      "grad_norm": 2.1950813663958897,
      "language_loss": 0.74280393,
      "learning_rate": 9.625775209499254e-09,
      "loss": 0.76453006,
      "num_input_tokens_seen": 174090565,
      "step": 8064,
      "time_per_iteration": 2.7442665100097656
    },
    {
      "auxiliary_loss_clip": 0.01130527,
      "auxiliary_loss_mlp": 0.01026936,
      "balance_loss_clip": 1.04873252,
      "balance_loss_mlp": 1.01878202,
      "epoch": 0.9697589130042686,
      "flos": 15121876360320.0,
      "grad_norm": 5.050330177180802,
      "language_loss": 0.74474013,
      "learning_rate": 9.549592478370172e-09,
      "loss": 0.76631474,
      "num_input_tokens_seen": 174108745,
      "step": 8065,
      "time_per_iteration": 3.688302516937256
    },
    {
      "auxiliary_loss_clip": 0.0114136,
      "auxiliary_loss_mlp": 0.01030508,
      "balance_loss_clip": 1.04684496,
      "balance_loss_mlp": 1.02345657,
      "epoch": 0.9698791558949077,
      "flos": 18879532824960.0,
      "grad_norm": 1.5764710239189343,
      "language_loss": 0.79426956,
      "learning_rate": 9.473711696985632e-09,
      "loss": 0.8159883,
      "num_input_tokens_seen": 174128075,
      "step": 8066,
      "time_per_iteration": 2.674586057662964
    },
    {
      "auxiliary_loss_clip": 0.01169303,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.04859376,
      "balance_loss_mlp": 1.02118099,
      "epoch": 0.9699993987855468,
      "flos": 17931455297280.0,
      "grad_norm": 1.9482248793213954,
      "language_loss": 0.76196289,
      "learning_rate": 9.398132876856201e-09,
      "loss": 0.78394723,
      "num_input_tokens_seen": 174147040,
      "step": 8067,
      "time_per_iteration": 2.665045976638794
    },
    {
      "auxiliary_loss_clip": 0.01025969,
      "auxiliary_loss_mlp": 0.01003067,
      "balance_loss_clip": 1.00749421,
      "balance_loss_mlp": 1.00213766,
      "epoch": 0.9701196416761859,
      "flos": 67182186297600.0,
      "grad_norm": 0.7770528449818389,
      "language_loss": 0.60678691,
      "learning_rate": 9.322856029447379e-09,
      "loss": 0.62707728,
      "num_input_tokens_seen": 174208225,
      "step": 8068,
      "time_per_iteration": 3.211669921875
    },
    {
      "auxiliary_loss_clip": 0.01160302,
      "auxiliary_loss_mlp": 0.01028688,
      "balance_loss_clip": 1.05080783,
      "balance_loss_mlp": 1.02180636,
      "epoch": 0.970239884566825,
      "flos": 24277804012800.0,
      "grad_norm": 1.9562394671422199,
      "language_loss": 0.80690062,
      "learning_rate": 9.247881166178695e-09,
      "loss": 0.82879055,
      "num_input_tokens_seen": 174226935,
      "step": 8069,
      "time_per_iteration": 2.7368133068084717
    },
    {
      "auxiliary_loss_clip": 0.01162954,
      "auxiliary_loss_mlp": 0.00761785,
      "balance_loss_clip": 1.05039573,
      "balance_loss_mlp": 1.00051117,
      "epoch": 0.970360127457464,
      "flos": 25301689194240.0,
      "grad_norm": 2.6046416444320606,
      "language_loss": 0.7722621,
      "learning_rate": 9.173208298423274e-09,
      "loss": 0.79150951,
      "num_input_tokens_seen": 174248140,
      "step": 8070,
      "time_per_iteration": 2.717078685760498
    },
    {
      "auxiliary_loss_clip": 0.01149676,
      "auxiliary_loss_mlp": 0.01026346,
      "balance_loss_clip": 1.04580903,
      "balance_loss_mlp": 1.0177927,
      "epoch": 0.9704803703481032,
      "flos": 29572473398400.0,
      "grad_norm": 1.6592267950255215,
      "language_loss": 0.76211858,
      "learning_rate": 9.09883743750961e-09,
      "loss": 0.7838788,
      "num_input_tokens_seen": 174271030,
      "step": 8071,
      "time_per_iteration": 3.667346954345703
    },
    {
      "auxiliary_loss_clip": 0.01106278,
      "auxiliary_loss_mlp": 0.01028297,
      "balance_loss_clip": 1.04246402,
      "balance_loss_mlp": 1.02013755,
      "epoch": 0.9706006132387422,
      "flos": 17380046638080.0,
      "grad_norm": 1.7686108455271954,
      "language_loss": 0.83985156,
      "learning_rate": 9.024768594719124e-09,
      "loss": 0.86119735,
      "num_input_tokens_seen": 174289410,
      "step": 8072,
      "time_per_iteration": 2.799591541290283
    },
    {
      "auxiliary_loss_clip": 0.01172719,
      "auxiliary_loss_mlp": 0.01026641,
      "balance_loss_clip": 1.04964757,
      "balance_loss_mlp": 1.01924407,
      "epoch": 0.9707208561293813,
      "flos": 18186421011840.0,
      "grad_norm": 5.631991046514578,
      "language_loss": 0.72705626,
      "learning_rate": 8.95100178128816e-09,
      "loss": 0.74904984,
      "num_input_tokens_seen": 174308550,
      "step": 8073,
      "time_per_iteration": 2.5901858806610107
    },
    {
      "auxiliary_loss_clip": 0.01140753,
      "auxiliary_loss_mlp": 0.01024441,
      "balance_loss_clip": 1.04551589,
      "balance_loss_mlp": 1.01634073,
      "epoch": 0.9708410990200205,
      "flos": 31248388212480.0,
      "grad_norm": 1.9115498407795457,
      "language_loss": 0.70243967,
      "learning_rate": 8.877537008407321e-09,
      "loss": 0.72409159,
      "num_input_tokens_seen": 174328600,
      "step": 8074,
      "time_per_iteration": 2.854787826538086
    },
    {
      "auxiliary_loss_clip": 0.01160357,
      "auxiliary_loss_mlp": 0.01031463,
      "balance_loss_clip": 1.05112565,
      "balance_loss_mlp": 1.02374411,
      "epoch": 0.9709613419106595,
      "flos": 30554450386560.0,
      "grad_norm": 1.6655281745803396,
      "language_loss": 0.6861099,
      "learning_rate": 8.804374287221028e-09,
      "loss": 0.70802814,
      "num_input_tokens_seen": 174349835,
      "step": 8075,
      "time_per_iteration": 2.7243998050689697
    },
    {
      "auxiliary_loss_clip": 0.01161523,
      "auxiliary_loss_mlp": 0.01025539,
      "balance_loss_clip": 1.04983568,
      "balance_loss_mlp": 1.01806486,
      "epoch": 0.9710815848012986,
      "flos": 23730166281600.0,
      "grad_norm": 4.465224824650002,
      "language_loss": 0.84687114,
      "learning_rate": 8.731513628827958e-09,
      "loss": 0.86874175,
      "num_input_tokens_seen": 174369200,
      "step": 8076,
      "time_per_iteration": 2.6951355934143066
    },
    {
      "auxiliary_loss_clip": 0.01105317,
      "auxiliary_loss_mlp": 0.01026826,
      "balance_loss_clip": 1.04203463,
      "balance_loss_mlp": 1.01887429,
      "epoch": 0.9712018276919377,
      "flos": 23761875012480.0,
      "grad_norm": 2.0990853602411574,
      "language_loss": 0.83030546,
      "learning_rate": 8.658955044280825e-09,
      "loss": 0.85162687,
      "num_input_tokens_seen": 174388125,
      "step": 8077,
      "time_per_iteration": 2.756051778793335
    },
    {
      "auxiliary_loss_clip": 0.01092169,
      "auxiliary_loss_mlp": 0.00761853,
      "balance_loss_clip": 1.03937626,
      "balance_loss_mlp": 1.00051045,
      "epoch": 0.9713220705825768,
      "flos": 23330983461120.0,
      "grad_norm": 2.3536965116252255,
      "language_loss": 0.77636081,
      "learning_rate": 8.586698544587268e-09,
      "loss": 0.79490101,
      "num_input_tokens_seen": 174409735,
      "step": 8078,
      "time_per_iteration": 2.8283379077911377
    },
    {
      "auxiliary_loss_clip": 0.01178612,
      "auxiliary_loss_mlp": 0.01030832,
      "balance_loss_clip": 1.05533874,
      "balance_loss_mlp": 1.0229578,
      "epoch": 0.9714423134732159,
      "flos": 22200946611840.0,
      "grad_norm": 2.365454548972056,
      "language_loss": 0.7432282,
      "learning_rate": 8.514744140707853e-09,
      "loss": 0.76532257,
      "num_input_tokens_seen": 174428875,
      "step": 8079,
      "time_per_iteration": 2.5572476387023926
    },
    {
      "auxiliary_loss_clip": 0.01122073,
      "auxiliary_loss_mlp": 0.010248,
      "balance_loss_clip": 1.04454505,
      "balance_loss_mlp": 1.01768017,
      "epoch": 0.971562556363855,
      "flos": 20229917656320.0,
      "grad_norm": 1.6085268519224372,
      "language_loss": 0.76522547,
      "learning_rate": 8.443091843558515e-09,
      "loss": 0.78669417,
      "num_input_tokens_seen": 174447960,
      "step": 8080,
      "time_per_iteration": 3.6649768352508545
    },
    {
      "auxiliary_loss_clip": 0.01157068,
      "auxiliary_loss_mlp": 0.01024545,
      "balance_loss_clip": 1.05117738,
      "balance_loss_mlp": 1.01694536,
      "epoch": 0.9716827992544941,
      "flos": 24970197553920.0,
      "grad_norm": 2.2551698347542044,
      "language_loss": 0.64906919,
      "learning_rate": 8.37174166400878e-09,
      "loss": 0.67088532,
      "num_input_tokens_seen": 174463535,
      "step": 8081,
      "time_per_iteration": 2.7066714763641357
    },
    {
      "auxiliary_loss_clip": 0.01153185,
      "auxiliary_loss_mlp": 0.01036095,
      "balance_loss_clip": 1.04931128,
      "balance_loss_mlp": 1.0286324,
      "epoch": 0.9718030421451331,
      "flos": 24681476033280.0,
      "grad_norm": 2.5289713733239263,
      "language_loss": 0.8514657,
      "learning_rate": 8.300693612881992e-09,
      "loss": 0.87335849,
      "num_input_tokens_seen": 174483600,
      "step": 8082,
      "time_per_iteration": 2.8261306285858154
    },
    {
      "auxiliary_loss_clip": 0.010835,
      "auxiliary_loss_mlp": 0.01026034,
      "balance_loss_clip": 1.03846335,
      "balance_loss_mlp": 1.01793921,
      "epoch": 0.9719232850357723,
      "flos": 22090700793600.0,
      "grad_norm": 2.018868109321885,
      "language_loss": 0.81612742,
      "learning_rate": 8.22994770095664e-09,
      "loss": 0.83722275,
      "num_input_tokens_seen": 174502175,
      "step": 8083,
      "time_per_iteration": 2.8615567684173584
    },
    {
      "auxiliary_loss_clip": 0.01151457,
      "auxiliary_loss_mlp": 0.01031188,
      "balance_loss_clip": 1.0485692,
      "balance_loss_mlp": 1.02308154,
      "epoch": 0.9720435279264114,
      "flos": 23656908493440.0,
      "grad_norm": 2.878524336330306,
      "language_loss": 0.75634098,
      "learning_rate": 8.159503938964585e-09,
      "loss": 0.77816737,
      "num_input_tokens_seen": 174519495,
      "step": 8084,
      "time_per_iteration": 2.649777889251709
    },
    {
      "auxiliary_loss_clip": 0.01133933,
      "auxiliary_loss_mlp": 0.01030393,
      "balance_loss_clip": 1.04600453,
      "balance_loss_mlp": 1.02306175,
      "epoch": 0.9721637708170504,
      "flos": 28365910623360.0,
      "grad_norm": 2.3157321974635527,
      "language_loss": 0.70587003,
      "learning_rate": 8.089362337592164e-09,
      "loss": 0.72751331,
      "num_input_tokens_seen": 174543120,
      "step": 8085,
      "time_per_iteration": 2.7947030067443848
    },
    {
      "auxiliary_loss_clip": 0.01124234,
      "auxiliary_loss_mlp": 0.01028014,
      "balance_loss_clip": 1.04241741,
      "balance_loss_mlp": 1.01991987,
      "epoch": 0.9722840137076896,
      "flos": 29130807767040.0,
      "grad_norm": 1.5235600622315044,
      "language_loss": 0.72491336,
      "learning_rate": 8.019522907479536e-09,
      "loss": 0.74643582,
      "num_input_tokens_seen": 174563480,
      "step": 8086,
      "time_per_iteration": 2.7430245876312256
    },
    {
      "auxiliary_loss_clip": 0.01107271,
      "auxiliary_loss_mlp": 0.00761625,
      "balance_loss_clip": 1.04803717,
      "balance_loss_mlp": 1.00053501,
      "epoch": 0.9724042565983286,
      "flos": 19243954258560.0,
      "grad_norm": 2.780561812297408,
      "language_loss": 0.77207553,
      "learning_rate": 7.949985659221558e-09,
      "loss": 0.79076445,
      "num_input_tokens_seen": 174580745,
      "step": 8087,
      "time_per_iteration": 2.809819459915161
    },
    {
      "auxiliary_loss_clip": 0.01117289,
      "auxiliary_loss_mlp": 0.01028196,
      "balance_loss_clip": 1.04276586,
      "balance_loss_mlp": 1.02090931,
      "epoch": 0.9725244994889677,
      "flos": 23039676161280.0,
      "grad_norm": 2.6066453999419634,
      "language_loss": 0.79219234,
      "learning_rate": 7.880750603366904e-09,
      "loss": 0.81364721,
      "num_input_tokens_seen": 174599615,
      "step": 8088,
      "time_per_iteration": 2.8015501499176025
    },
    {
      "auxiliary_loss_clip": 0.01134302,
      "auxiliary_loss_mlp": 0.01025281,
      "balance_loss_clip": 1.04411495,
      "balance_loss_mlp": 1.01663828,
      "epoch": 0.9726447423796069,
      "flos": 23367468700800.0,
      "grad_norm": 2.525980403970561,
      "language_loss": 0.79659355,
      "learning_rate": 7.811817750418282e-09,
      "loss": 0.81818938,
      "num_input_tokens_seen": 174618375,
      "step": 8089,
      "time_per_iteration": 3.572124481201172
    },
    {
      "auxiliary_loss_clip": 0.01130233,
      "auxiliary_loss_mlp": 0.00762859,
      "balance_loss_clip": 1.0431062,
      "balance_loss_mlp": 1.00050616,
      "epoch": 0.9727649852702459,
      "flos": 26541648639360.0,
      "grad_norm": 1.5469842098050772,
      "language_loss": 0.80085492,
      "learning_rate": 7.743187110833105e-09,
      "loss": 0.81978589,
      "num_input_tokens_seen": 174641135,
      "step": 8090,
      "time_per_iteration": 2.78442645072937
    },
    {
      "auxiliary_loss_clip": 0.01110356,
      "auxiliary_loss_mlp": 0.01029367,
      "balance_loss_clip": 1.04260671,
      "balance_loss_mlp": 1.02220297,
      "epoch": 0.972885228160885,
      "flos": 20522338277760.0,
      "grad_norm": 1.8604002632961145,
      "language_loss": 0.80676204,
      "learning_rate": 7.674858695022602e-09,
      "loss": 0.82815921,
      "num_input_tokens_seen": 174659490,
      "step": 8091,
      "time_per_iteration": 3.6979589462280273
    },
    {
      "auxiliary_loss_clip": 0.01060069,
      "auxiliary_loss_mlp": 0.01031882,
      "balance_loss_clip": 1.03957653,
      "balance_loss_mlp": 1.02358472,
      "epoch": 0.9730054710515241,
      "flos": 17566064196480.0,
      "grad_norm": 2.6326499476991874,
      "language_loss": 0.7617023,
      "learning_rate": 7.606832513351591e-09,
      "loss": 0.7826218,
      "num_input_tokens_seen": 174677440,
      "step": 8092,
      "time_per_iteration": 2.996721029281616
    },
    {
      "auxiliary_loss_clip": 0.01041266,
      "auxiliary_loss_mlp": 0.00752014,
      "balance_loss_clip": 1.0092926,
      "balance_loss_mlp": 1.00032902,
      "epoch": 0.9731257139421632,
      "flos": 68972010117120.0,
      "grad_norm": 0.8430448862884031,
      "language_loss": 0.63829976,
      "learning_rate": 7.539108576140264e-09,
      "loss": 0.65623254,
      "num_input_tokens_seen": 174741550,
      "step": 8093,
      "time_per_iteration": 3.480529308319092
    },
    {
      "auxiliary_loss_clip": 0.01116084,
      "auxiliary_loss_mlp": 0.01026525,
      "balance_loss_clip": 1.04716623,
      "balance_loss_mlp": 1.01906824,
      "epoch": 0.9732459568328022,
      "flos": 18478841633280.0,
      "grad_norm": 2.489675229108108,
      "language_loss": 0.70597804,
      "learning_rate": 7.471686893661732e-09,
      "loss": 0.72740412,
      "num_input_tokens_seen": 174759845,
      "step": 8094,
      "time_per_iteration": 2.72318172454834
    },
    {
      "auxiliary_loss_clip": 0.01131812,
      "auxiliary_loss_mlp": 0.010253,
      "balance_loss_clip": 1.04527044,
      "balance_loss_mlp": 1.01821351,
      "epoch": 0.9733661997234414,
      "flos": 20883886623360.0,
      "grad_norm": 1.7655691502372475,
      "language_loss": 0.64688718,
      "learning_rate": 7.4045674761442636e-09,
      "loss": 0.66845834,
      "num_input_tokens_seen": 174777175,
      "step": 8095,
      "time_per_iteration": 2.758394718170166
    },
    {
      "auxiliary_loss_clip": 0.01160738,
      "auxiliary_loss_mlp": 0.01035451,
      "balance_loss_clip": 1.0492475,
      "balance_loss_mlp": 1.02750313,
      "epoch": 0.9734864426140805,
      "flos": 23766795175680.0,
      "grad_norm": 2.07567790324828,
      "language_loss": 0.74482489,
      "learning_rate": 7.337750333769488e-09,
      "loss": 0.76678681,
      "num_input_tokens_seen": 174796980,
      "step": 8096,
      "time_per_iteration": 3.6682732105255127
    },
    {
      "auxiliary_loss_clip": 0.01123233,
      "auxiliary_loss_mlp": 0.01025798,
      "balance_loss_clip": 1.0460937,
      "balance_loss_mlp": 1.0167861,
      "epoch": 0.9736066855047195,
      "flos": 35042422176000.0,
      "grad_norm": 2.169820854057791,
      "language_loss": 0.72748733,
      "learning_rate": 7.2712354766737425e-09,
      "loss": 0.74897766,
      "num_input_tokens_seen": 174817310,
      "step": 8097,
      "time_per_iteration": 2.8059465885162354
    },
    {
      "auxiliary_loss_clip": 0.01161842,
      "auxiliary_loss_mlp": 0.01026258,
      "balance_loss_clip": 1.05055833,
      "balance_loss_mlp": 1.01848578,
      "epoch": 0.9737269283953586,
      "flos": 20410620001920.0,
      "grad_norm": 1.6133711415747451,
      "language_loss": 0.81090689,
      "learning_rate": 7.2050229149469565e-09,
      "loss": 0.83278787,
      "num_input_tokens_seen": 174837320,
      "step": 8098,
      "time_per_iteration": 2.7546136379241943
    },
    {
      "auxiliary_loss_clip": 0.01130682,
      "auxiliary_loss_mlp": 0.01028062,
      "balance_loss_clip": 1.04500747,
      "balance_loss_mlp": 1.02087426,
      "epoch": 0.9738471712859977,
      "flos": 28911680847360.0,
      "grad_norm": 1.7307469904127275,
      "language_loss": 0.63963675,
      "learning_rate": 7.139112658633984e-09,
      "loss": 0.66122419,
      "num_input_tokens_seen": 174857470,
      "step": 8099,
      "time_per_iteration": 2.6902623176574707
    },
    {
      "auxiliary_loss_clip": 0.01108334,
      "auxiliary_loss_mlp": 0.01030643,
      "balance_loss_clip": 1.04328954,
      "balance_loss_mlp": 1.02231002,
      "epoch": 0.9739674141766368,
      "flos": 27782326356480.0,
      "grad_norm": 1.8797576820580253,
      "language_loss": 0.70729256,
      "learning_rate": 7.073504717733048e-09,
      "loss": 0.72868228,
      "num_input_tokens_seen": 174877035,
      "step": 8100,
      "time_per_iteration": 2.7904579639434814
    },
    {
      "auxiliary_loss_clip": 0.01026116,
      "auxiliary_loss_mlp": 0.01001034,
      "balance_loss_clip": 1.00712204,
      "balance_loss_mlp": 1.00017011,
      "epoch": 0.9740876570672758,
      "flos": 68863057188480.0,
      "grad_norm": 0.7359446870690496,
      "language_loss": 0.57183522,
      "learning_rate": 7.008199102196855e-09,
      "loss": 0.59210676,
      "num_input_tokens_seen": 174938460,
      "step": 8101,
      "time_per_iteration": 3.219398260116577
    },
    {
      "auxiliary_loss_clip": 0.01031158,
      "auxiliary_loss_mlp": 0.01000696,
      "balance_loss_clip": 1.00934625,
      "balance_loss_mlp": 0.99976635,
      "epoch": 0.974207899957915,
      "flos": 58236622646400.0,
      "grad_norm": 0.7936150019062816,
      "language_loss": 0.58899105,
      "learning_rate": 6.9431958219321464e-09,
      "loss": 0.60930967,
      "num_input_tokens_seen": 174994625,
      "step": 8102,
      "time_per_iteration": 3.218508005142212
    },
    {
      "auxiliary_loss_clip": 0.01094284,
      "auxiliary_loss_mlp": 0.01031837,
      "balance_loss_clip": 1.03968382,
      "balance_loss_mlp": 1.0232358,
      "epoch": 0.9743281428485541,
      "flos": 22600057605120.0,
      "grad_norm": 1.716733527034041,
      "language_loss": 0.77865869,
      "learning_rate": 6.878494886800146e-09,
      "loss": 0.7999199,
      "num_input_tokens_seen": 175015400,
      "step": 8103,
      "time_per_iteration": 2.844261884689331
    },
    {
      "auxiliary_loss_clip": 0.01139103,
      "auxiliary_loss_mlp": 0.01025041,
      "balance_loss_clip": 1.04638076,
      "balance_loss_mlp": 1.01767945,
      "epoch": 0.9744483857391931,
      "flos": 20008815488640.0,
      "grad_norm": 1.728815715151365,
      "language_loss": 0.76719272,
      "learning_rate": 6.814096306615669e-09,
      "loss": 0.78883421,
      "num_input_tokens_seen": 175033540,
      "step": 8104,
      "time_per_iteration": 2.686253547668457
    },
    {
      "auxiliary_loss_clip": 0.01109875,
      "auxiliary_loss_mlp": 0.01027441,
      "balance_loss_clip": 1.04001021,
      "balance_loss_mlp": 1.01880443,
      "epoch": 0.9745686286298323,
      "flos": 17675268520320.0,
      "grad_norm": 3.14696694870003,
      "language_loss": 0.66061586,
      "learning_rate": 6.750000091148011e-09,
      "loss": 0.68198907,
      "num_input_tokens_seen": 175050835,
      "step": 8105,
      "time_per_iteration": 2.7972943782806396
    },
    {
      "auxiliary_loss_clip": 0.01136638,
      "auxiliary_loss_mlp": 0.01025078,
      "balance_loss_clip": 1.04718173,
      "balance_loss_mlp": 1.01723695,
      "epoch": 0.9746888715204713,
      "flos": 29460252332160.0,
      "grad_norm": 3.462427121629555,
      "language_loss": 0.72976291,
      "learning_rate": 6.686206250120729e-09,
      "loss": 0.75138015,
      "num_input_tokens_seen": 175072330,
      "step": 8106,
      "time_per_iteration": 3.701171636581421
    },
    {
      "auxiliary_loss_clip": 0.01129252,
      "auxiliary_loss_mlp": 0.01022747,
      "balance_loss_clip": 1.04704177,
      "balance_loss_mlp": 1.01565075,
      "epoch": 0.9748091144111104,
      "flos": 18479308510080.0,
      "grad_norm": 1.961449542201095,
      "language_loss": 0.74479705,
      "learning_rate": 6.622714793210749e-09,
      "loss": 0.76631707,
      "num_input_tokens_seen": 175091250,
      "step": 8107,
      "time_per_iteration": 2.6866846084594727
    },
    {
      "auxiliary_loss_clip": 0.01172628,
      "auxiliary_loss_mlp": 0.0102741,
      "balance_loss_clip": 1.05030894,
      "balance_loss_mlp": 1.01965845,
      "epoch": 0.9749293573017496,
      "flos": 20665154753280.0,
      "grad_norm": 2.115101247927632,
      "language_loss": 0.78532827,
      "learning_rate": 6.559525730050364e-09,
      "loss": 0.8073287,
      "num_input_tokens_seen": 175111350,
      "step": 8108,
      "time_per_iteration": 2.7075939178466797
    },
    {
      "auxiliary_loss_clip": 0.01146974,
      "auxiliary_loss_mlp": 0.01029978,
      "balance_loss_clip": 1.04650664,
      "balance_loss_mlp": 1.02146053,
      "epoch": 0.9750496001923886,
      "flos": 18478590238080.0,
      "grad_norm": 1.8426264035810835,
      "language_loss": 0.75972867,
      "learning_rate": 6.496639070224574e-09,
      "loss": 0.78149819,
      "num_input_tokens_seen": 175129835,
      "step": 8109,
      "time_per_iteration": 2.72115421295166
    },
    {
      "auxiliary_loss_clip": 0.0116054,
      "auxiliary_loss_mlp": 0.01030918,
      "balance_loss_clip": 1.0489893,
      "balance_loss_mlp": 1.02359295,
      "epoch": 0.9751698430830277,
      "flos": 19572967860480.0,
      "grad_norm": 2.196097354842335,
      "language_loss": 0.83960056,
      "learning_rate": 6.4340548232739714e-09,
      "loss": 0.86151516,
      "num_input_tokens_seen": 175146035,
      "step": 8110,
      "time_per_iteration": 2.7022476196289062
    },
    {
      "auxiliary_loss_clip": 0.0116316,
      "auxiliary_loss_mlp": 0.01028229,
      "balance_loss_clip": 1.0511992,
      "balance_loss_mlp": 1.02059317,
      "epoch": 0.9752900859736668,
      "flos": 23550325862400.0,
      "grad_norm": 1.8644073887005048,
      "language_loss": 0.7930578,
      "learning_rate": 6.371772998692071e-09,
      "loss": 0.81497169,
      "num_input_tokens_seen": 175165290,
      "step": 8111,
      "time_per_iteration": 2.6404366493225098
    },
    {
      "auxiliary_loss_clip": 0.01122979,
      "auxiliary_loss_mlp": 0.01024386,
      "balance_loss_clip": 1.0453521,
      "balance_loss_mlp": 1.01685214,
      "epoch": 0.9754103288643059,
      "flos": 20303211358080.0,
      "grad_norm": 4.403963014053926,
      "language_loss": 0.6537711,
      "learning_rate": 6.309793605927094e-09,
      "loss": 0.67524481,
      "num_input_tokens_seen": 175183610,
      "step": 8112,
      "time_per_iteration": 2.7634365558624268
    },
    {
      "auxiliary_loss_clip": 0.01142388,
      "auxiliary_loss_mlp": 0.0102664,
      "balance_loss_clip": 1.04790998,
      "balance_loss_mlp": 1.01935589,
      "epoch": 0.975530571754945,
      "flos": 19350680544000.0,
      "grad_norm": 1.7893460711408815,
      "language_loss": 0.80166781,
      "learning_rate": 6.248116654381297e-09,
      "loss": 0.82335818,
      "num_input_tokens_seen": 175202080,
      "step": 8113,
      "time_per_iteration": 2.6839256286621094
    },
    {
      "auxiliary_loss_clip": 0.01158289,
      "auxiliary_loss_mlp": 0.01029039,
      "balance_loss_clip": 1.04810882,
      "balance_loss_mlp": 1.02184772,
      "epoch": 0.9756508146455841,
      "flos": 23583399310080.0,
      "grad_norm": 2.0183706525860394,
      "language_loss": 0.7268095,
      "learning_rate": 6.186742153410751e-09,
      "loss": 0.74868274,
      "num_input_tokens_seen": 175221575,
      "step": 8114,
      "time_per_iteration": 4.086115121841431
    },
    {
      "auxiliary_loss_clip": 0.01132072,
      "auxiliary_loss_mlp": 0.01028578,
      "balance_loss_clip": 1.04761887,
      "balance_loss_mlp": 1.01957774,
      "epoch": 0.9757710575362232,
      "flos": 22966921163520.0,
      "grad_norm": 1.991085439223713,
      "language_loss": 0.87578535,
      "learning_rate": 6.125670112326453e-09,
      "loss": 0.89739186,
      "num_input_tokens_seen": 175240835,
      "step": 8115,
      "time_per_iteration": 2.8277909755706787
    },
    {
      "auxiliary_loss_clip": 0.01139373,
      "auxiliary_loss_mlp": 0.01028384,
      "balance_loss_clip": 1.04747736,
      "balance_loss_mlp": 1.02061212,
      "epoch": 0.9758913004268622,
      "flos": 27966009530880.0,
      "grad_norm": 1.76859699365259,
      "language_loss": 0.70914853,
      "learning_rate": 6.064900540392548e-09,
      "loss": 0.73082614,
      "num_input_tokens_seen": 175262930,
      "step": 8116,
      "time_per_iteration": 3.748957395553589
    },
    {
      "auxiliary_loss_clip": 0.01125818,
      "auxiliary_loss_mlp": 0.0101827,
      "balance_loss_clip": 1.04414415,
      "balance_loss_mlp": 1.01145983,
      "epoch": 0.9760115433175014,
      "flos": 22200156512640.0,
      "grad_norm": 2.472973839601492,
      "language_loss": 0.79155034,
      "learning_rate": 6.0044334468278835e-09,
      "loss": 0.81299126,
      "num_input_tokens_seen": 175282275,
      "step": 8117,
      "time_per_iteration": 2.6974782943725586
    },
    {
      "auxiliary_loss_clip": 0.01162389,
      "auxiliary_loss_mlp": 0.01033925,
      "balance_loss_clip": 1.04956377,
      "balance_loss_mlp": 1.02626848,
      "epoch": 0.9761317862081405,
      "flos": 26250736389120.0,
      "grad_norm": 2.0905497546419696,
      "language_loss": 0.71834826,
      "learning_rate": 5.944268840805345e-09,
      "loss": 0.74031138,
      "num_input_tokens_seen": 175303020,
      "step": 8118,
      "time_per_iteration": 2.716292381286621
    },
    {
      "auxiliary_loss_clip": 0.01151151,
      "auxiliary_loss_mlp": 0.01025213,
      "balance_loss_clip": 1.04780138,
      "balance_loss_mlp": 1.01754749,
      "epoch": 0.9762520290987795,
      "flos": 26575440359040.0,
      "grad_norm": 2.2971789469824255,
      "language_loss": 0.64327031,
      "learning_rate": 5.88440673145163e-09,
      "loss": 0.66503394,
      "num_input_tokens_seen": 175324070,
      "step": 8119,
      "time_per_iteration": 2.6833243370056152
    },
    {
      "auxiliary_loss_clip": 0.01108403,
      "auxiliary_loss_mlp": 0.01027099,
      "balance_loss_clip": 1.04567218,
      "balance_loss_mlp": 1.0191772,
      "epoch": 0.9763722719894187,
      "flos": 18005036307840.0,
      "grad_norm": 2.4716876452402827,
      "language_loss": 0.82773042,
      "learning_rate": 5.824847127848142e-09,
      "loss": 0.84908539,
      "num_input_tokens_seen": 175342595,
      "step": 8120,
      "time_per_iteration": 2.8104562759399414
    },
    {
      "auxiliary_loss_clip": 0.01094178,
      "auxiliary_loss_mlp": 0.01024375,
      "balance_loss_clip": 1.0369463,
      "balance_loss_mlp": 1.01670384,
      "epoch": 0.9764925148800577,
      "flos": 22455660931200.0,
      "grad_norm": 1.9239144906181365,
      "language_loss": 0.78955466,
      "learning_rate": 5.765590039029433e-09,
      "loss": 0.81074023,
      "num_input_tokens_seen": 175361915,
      "step": 8121,
      "time_per_iteration": 2.7669122219085693
    },
    {
      "auxiliary_loss_clip": 0.0113697,
      "auxiliary_loss_mlp": 0.01026791,
      "balance_loss_clip": 1.0478909,
      "balance_loss_mlp": 1.01940012,
      "epoch": 0.9766127577706968,
      "flos": 36757084786560.0,
      "grad_norm": 1.740065292163286,
      "language_loss": 0.71123457,
      "learning_rate": 5.706635473985422e-09,
      "loss": 0.73287213,
      "num_input_tokens_seen": 175385785,
      "step": 8122,
      "time_per_iteration": 3.7557716369628906
    },
    {
      "auxiliary_loss_clip": 0.01120113,
      "auxiliary_loss_mlp": 0.01022153,
      "balance_loss_clip": 1.04385531,
      "balance_loss_mlp": 1.01474118,
      "epoch": 0.976733000661336,
      "flos": 22309971367680.0,
      "grad_norm": 1.9607270747054344,
      "language_loss": 0.85109556,
      "learning_rate": 5.6479834416591764e-09,
      "loss": 0.87251824,
      "num_input_tokens_seen": 175405145,
      "step": 8123,
      "time_per_iteration": 2.7891552448272705
    },
    {
      "auxiliary_loss_clip": 0.01119062,
      "auxiliary_loss_mlp": 0.00762793,
      "balance_loss_clip": 1.04363036,
      "balance_loss_mlp": 1.0004952,
      "epoch": 0.976853243551975,
      "flos": 25810938264960.0,
      "grad_norm": 1.986827584319839,
      "language_loss": 0.68551338,
      "learning_rate": 5.589633950947803e-09,
      "loss": 0.70433187,
      "num_input_tokens_seen": 175422645,
      "step": 8124,
      "time_per_iteration": 2.765655517578125
    },
    {
      "auxiliary_loss_clip": 0.01166786,
      "auxiliary_loss_mlp": 0.01030551,
      "balance_loss_clip": 1.05411923,
      "balance_loss_mlp": 1.02099657,
      "epoch": 0.9769734864426141,
      "flos": 21397445326080.0,
      "grad_norm": 2.294408581734529,
      "language_loss": 0.69770342,
      "learning_rate": 5.5315870107035535e-09,
      "loss": 0.71967679,
      "num_input_tokens_seen": 175440695,
      "step": 8125,
      "time_per_iteration": 2.711186647415161
    },
    {
      "auxiliary_loss_clip": 0.01112849,
      "auxiliary_loss_mlp": 0.01027493,
      "balance_loss_clip": 1.04168391,
      "balance_loss_mlp": 1.02021527,
      "epoch": 0.9770937293332532,
      "flos": 13990977584640.0,
      "grad_norm": 2.243556114545793,
      "language_loss": 0.78933036,
      "learning_rate": 5.473842629731607e-09,
      "loss": 0.8107338,
      "num_input_tokens_seen": 175459195,
      "step": 8126,
      "time_per_iteration": 2.7347824573516846
    },
    {
      "auxiliary_loss_clip": 0.01125787,
      "auxiliary_loss_mlp": 0.01028913,
      "balance_loss_clip": 1.04645085,
      "balance_loss_mlp": 1.02058601,
      "epoch": 0.9772139722238923,
      "flos": 17931994001280.0,
      "grad_norm": 2.889733279531239,
      "language_loss": 0.7803781,
      "learning_rate": 5.416400816792066e-09,
      "loss": 0.80192506,
      "num_input_tokens_seen": 175476710,
      "step": 8127,
      "time_per_iteration": 2.744939088821411
    },
    {
      "auxiliary_loss_clip": 0.01148958,
      "auxiliary_loss_mlp": 0.00761532,
      "balance_loss_clip": 1.04746437,
      "balance_loss_mlp": 1.00044107,
      "epoch": 0.9773342151145313,
      "flos": 20446171488000.0,
      "grad_norm": 3.7252311151541733,
      "language_loss": 0.78437209,
      "learning_rate": 5.359261580598407e-09,
      "loss": 0.80347699,
      "num_input_tokens_seen": 175492550,
      "step": 8128,
      "time_per_iteration": 2.676053047180176
    },
    {
      "auxiliary_loss_clip": 0.01162023,
      "auxiliary_loss_mlp": 0.01035501,
      "balance_loss_clip": 1.05038011,
      "balance_loss_mlp": 1.02610755,
      "epoch": 0.9774544580051704,
      "flos": 11837306949120.0,
      "grad_norm": 10.993853892050042,
      "language_loss": 0.7803725,
      "learning_rate": 5.302424929819027e-09,
      "loss": 0.80234772,
      "num_input_tokens_seen": 175506560,
      "step": 8129,
      "time_per_iteration": 2.63480544090271
    },
    {
      "auxiliary_loss_clip": 0.01176325,
      "auxiliary_loss_mlp": 0.00762333,
      "balance_loss_clip": 1.05224645,
      "balance_loss_mlp": 1.0004425,
      "epoch": 0.9775747008958096,
      "flos": 13479932833920.0,
      "grad_norm": 2.779339153025773,
      "language_loss": 0.72777689,
      "learning_rate": 5.24589087307592e-09,
      "loss": 0.74716353,
      "num_input_tokens_seen": 175524180,
      "step": 8130,
      "time_per_iteration": 2.577754020690918
    },
    {
      "auxiliary_loss_clip": 0.01138759,
      "auxiliary_loss_mlp": 0.01029389,
      "balance_loss_clip": 1.04634905,
      "balance_loss_mlp": 1.02000773,
      "epoch": 0.9776949437864486,
      "flos": 59532314042880.0,
      "grad_norm": 1.4691384448338758,
      "language_loss": 0.65052187,
      "learning_rate": 5.189659418944891e-09,
      "loss": 0.67220342,
      "num_input_tokens_seen": 175554355,
      "step": 8131,
      "time_per_iteration": 3.071259021759033
    },
    {
      "auxiliary_loss_clip": 0.01162516,
      "auxiliary_loss_mlp": 0.01025332,
      "balance_loss_clip": 1.05085635,
      "balance_loss_mlp": 1.01777434,
      "epoch": 0.9778151866770877,
      "flos": 21178605715200.0,
      "grad_norm": 2.1597683112522934,
      "language_loss": 0.78692615,
      "learning_rate": 5.133730575956674e-09,
      "loss": 0.80880457,
      "num_input_tokens_seen": 175574025,
      "step": 8132,
      "time_per_iteration": 3.618861436843872
    },
    {
      "auxiliary_loss_clip": 0.01174506,
      "auxiliary_loss_mlp": 0.01030514,
      "balance_loss_clip": 1.05234575,
      "balance_loss_mlp": 1.02237201,
      "epoch": 0.9779354295677268,
      "flos": 20886795624960.0,
      "grad_norm": 2.114637034569139,
      "language_loss": 0.72443032,
      "learning_rate": 5.0781043525953696e-09,
      "loss": 0.74648046,
      "num_input_tokens_seen": 175592090,
      "step": 8133,
      "time_per_iteration": 2.6383562088012695
    },
    {
      "auxiliary_loss_clip": 0.01126481,
      "auxiliary_loss_mlp": 0.0102682,
      "balance_loss_clip": 1.04608583,
      "balance_loss_mlp": 1.01942885,
      "epoch": 0.9780556724583659,
      "flos": 23440618748160.0,
      "grad_norm": 1.7567349053918375,
      "language_loss": 0.74160123,
      "learning_rate": 5.0227807572995605e-09,
      "loss": 0.76313424,
      "num_input_tokens_seen": 175614065,
      "step": 8134,
      "time_per_iteration": 2.780766487121582
    },
    {
      "auxiliary_loss_clip": 0.01133606,
      "auxiliary_loss_mlp": 0.00761979,
      "balance_loss_clip": 1.04607952,
      "balance_loss_mlp": 1.00048685,
      "epoch": 0.9781759153490049,
      "flos": 20923244951040.0,
      "grad_norm": 2.497205575736192,
      "language_loss": 0.67527312,
      "learning_rate": 4.967759798461646e-09,
      "loss": 0.69422901,
      "num_input_tokens_seen": 175632410,
      "step": 8135,
      "time_per_iteration": 2.697702407836914
    },
    {
      "auxiliary_loss_clip": 0.01084941,
      "auxiliary_loss_mlp": 0.01028314,
      "balance_loss_clip": 1.03832006,
      "balance_loss_mlp": 1.02095878,
      "epoch": 0.9782961582396441,
      "flos": 28293191539200.0,
      "grad_norm": 2.076298054486896,
      "language_loss": 0.75243473,
      "learning_rate": 4.913041484428282e-09,
      "loss": 0.7735672,
      "num_input_tokens_seen": 175652885,
      "step": 8136,
      "time_per_iteration": 2.9214446544647217
    },
    {
      "auxiliary_loss_clip": 0.01157318,
      "auxiliary_loss_mlp": 0.01029379,
      "balance_loss_clip": 1.05038822,
      "balance_loss_mlp": 1.022295,
      "epoch": 0.9784164011302832,
      "flos": 25552955808000.0,
      "grad_norm": 2.168253163025149,
      "language_loss": 0.7429378,
      "learning_rate": 4.858625823500384e-09,
      "loss": 0.76480478,
      "num_input_tokens_seen": 175670585,
      "step": 8137,
      "time_per_iteration": 2.670117139816284
    },
    {
      "auxiliary_loss_clip": 0.01132726,
      "auxiliary_loss_mlp": 0.01034219,
      "balance_loss_clip": 1.04497981,
      "balance_loss_mlp": 1.0269115,
      "epoch": 0.9785366440209222,
      "flos": 29965945956480.0,
      "grad_norm": 2.161643228978372,
      "language_loss": 0.73522818,
      "learning_rate": 4.80451282393246e-09,
      "loss": 0.75689769,
      "num_input_tokens_seen": 175690570,
      "step": 8138,
      "time_per_iteration": 2.8340771198272705
    },
    {
      "auxiliary_loss_clip": 0.01145744,
      "auxiliary_loss_mlp": 0.01030597,
      "balance_loss_clip": 1.04700971,
      "balance_loss_mlp": 1.02237761,
      "epoch": 0.9786568869115614,
      "flos": 32343591847680.0,
      "grad_norm": 2.0474854030526815,
      "language_loss": 0.673347,
      "learning_rate": 4.750702493933722e-09,
      "loss": 0.69511038,
      "num_input_tokens_seen": 175710455,
      "step": 8139,
      "time_per_iteration": 2.853801727294922
    },
    {
      "auxiliary_loss_clip": 0.01154893,
      "auxiliary_loss_mlp": 0.01034375,
      "balance_loss_clip": 1.04918921,
      "balance_loss_mlp": 1.02673423,
      "epoch": 0.9787771298022004,
      "flos": 23331414424320.0,
      "grad_norm": 1.9823563029954145,
      "language_loss": 0.85221189,
      "learning_rate": 4.697194841666974e-09,
      "loss": 0.87410462,
      "num_input_tokens_seen": 175729380,
      "step": 8140,
      "time_per_iteration": 3.6131680011749268
    },
    {
      "auxiliary_loss_clip": 0.01148575,
      "auxiliary_loss_mlp": 0.00762321,
      "balance_loss_clip": 1.04473948,
      "balance_loss_mlp": 1.00043702,
      "epoch": 0.9788973726928395,
      "flos": 21468548298240.0,
      "grad_norm": 2.0750910148207455,
      "language_loss": 0.82198215,
      "learning_rate": 4.6439898752492764e-09,
      "loss": 0.84109116,
      "num_input_tokens_seen": 175749520,
      "step": 8141,
      "time_per_iteration": 2.7333152294158936
    },
    {
      "auxiliary_loss_clip": 0.01034404,
      "auxiliary_loss_mlp": 0.01000313,
      "balance_loss_clip": 1.01543427,
      "balance_loss_mlp": 0.99943048,
      "epoch": 0.9790176155834787,
      "flos": 68897459439360.0,
      "grad_norm": 0.7473825042276206,
      "language_loss": 0.63561916,
      "learning_rate": 4.591087602751731e-09,
      "loss": 0.65596634,
      "num_input_tokens_seen": 175811380,
      "step": 8142,
      "time_per_iteration": 4.265977621078491
    },
    {
      "auxiliary_loss_clip": 0.01088728,
      "auxiliary_loss_mlp": 0.01022908,
      "balance_loss_clip": 1.0398469,
      "balance_loss_mlp": 1.01542521,
      "epoch": 0.9791378584741177,
      "flos": 21430877909760.0,
      "grad_norm": 2.7864384120597134,
      "language_loss": 0.72339219,
      "learning_rate": 4.538488032199916e-09,
      "loss": 0.74450856,
      "num_input_tokens_seen": 175829480,
      "step": 8143,
      "time_per_iteration": 2.7702836990356445
    },
    {
      "auxiliary_loss_clip": 0.01156093,
      "auxiliary_loss_mlp": 0.01030618,
      "balance_loss_clip": 1.04801786,
      "balance_loss_mlp": 1.02266037,
      "epoch": 0.9792581013647568,
      "flos": 20153032594560.0,
      "grad_norm": 2.3197224555222413,
      "language_loss": 0.69511044,
      "learning_rate": 4.486191171572784e-09,
      "loss": 0.7169776,
      "num_input_tokens_seen": 175846750,
      "step": 8144,
      "time_per_iteration": 2.6337480545043945
    },
    {
      "auxiliary_loss_clip": 0.01095908,
      "auxiliary_loss_mlp": 0.01032352,
      "balance_loss_clip": 1.04351211,
      "balance_loss_mlp": 1.02413905,
      "epoch": 0.9793783442553959,
      "flos": 23728191033600.0,
      "grad_norm": 1.9603429205781298,
      "language_loss": 0.77734488,
      "learning_rate": 4.434197028803766e-09,
      "loss": 0.79862744,
      "num_input_tokens_seen": 175865975,
      "step": 8145,
      "time_per_iteration": 2.7303836345672607
    },
    {
      "auxiliary_loss_clip": 0.01133541,
      "auxiliary_loss_mlp": 0.01033023,
      "balance_loss_clip": 1.04755449,
      "balance_loss_mlp": 1.02478552,
      "epoch": 0.979498587146035,
      "flos": 23038742407680.0,
      "grad_norm": 2.0327553383800825,
      "language_loss": 0.82285702,
      "learning_rate": 4.3825056117805514e-09,
      "loss": 0.84452271,
      "num_input_tokens_seen": 175881860,
      "step": 8146,
      "time_per_iteration": 2.781104564666748
    },
    {
      "auxiliary_loss_clip": 0.0112919,
      "auxiliary_loss_mlp": 0.01031249,
      "balance_loss_clip": 1.04530346,
      "balance_loss_mlp": 1.02286839,
      "epoch": 0.979618830036674,
      "flos": 14318841951360.0,
      "grad_norm": 2.3028872900548576,
      "language_loss": 0.79705882,
      "learning_rate": 4.331116928344425e-09,
      "loss": 0.81866318,
      "num_input_tokens_seen": 175898175,
      "step": 8147,
      "time_per_iteration": 2.6852524280548096
    },
    {
      "auxiliary_loss_clip": 0.01149052,
      "auxiliary_loss_mlp": 0.01028321,
      "balance_loss_clip": 1.04764915,
      "balance_loss_mlp": 1.02059364,
      "epoch": 0.9797390729273132,
      "flos": 16727514215040.0,
      "grad_norm": 1.9848407677509812,
      "language_loss": 0.62659562,
      "learning_rate": 4.28003098629115e-09,
      "loss": 0.64836931,
      "num_input_tokens_seen": 175914310,
      "step": 8148,
      "time_per_iteration": 3.6068055629730225
    },
    {
      "auxiliary_loss_clip": 0.01157294,
      "auxiliary_loss_mlp": 0.01026664,
      "balance_loss_clip": 1.04713535,
      "balance_loss_mlp": 1.01877189,
      "epoch": 0.9798593158179523,
      "flos": 24532661986560.0,
      "grad_norm": 2.1822532001071733,
      "language_loss": 0.78581202,
      "learning_rate": 4.229247793370305e-09,
      "loss": 0.80765152,
      "num_input_tokens_seen": 175933435,
      "step": 8149,
      "time_per_iteration": 2.711613893508911
    },
    {
      "auxiliary_loss_clip": 0.01117862,
      "auxiliary_loss_mlp": 0.01026094,
      "balance_loss_clip": 1.04475117,
      "balance_loss_mlp": 1.01808929,
      "epoch": 0.9799795587085913,
      "flos": 27308808339840.0,
      "grad_norm": 1.7996168384847941,
      "language_loss": 0.7073158,
      "learning_rate": 4.178767357285951e-09,
      "loss": 0.72875535,
      "num_input_tokens_seen": 175955065,
      "step": 8150,
      "time_per_iteration": 2.854198455810547
    },
    {
      "auxiliary_loss_clip": 0.0112193,
      "auxiliary_loss_mlp": 0.01024897,
      "balance_loss_clip": 1.04538906,
      "balance_loss_mlp": 1.01674342,
      "epoch": 0.9800998015992305,
      "flos": 26286575184000.0,
      "grad_norm": 1.9133519922374933,
      "language_loss": 0.71531284,
      "learning_rate": 4.128589685695516e-09,
      "loss": 0.73678112,
      "num_input_tokens_seen": 175975490,
      "step": 8151,
      "time_per_iteration": 2.7771494388580322
    },
    {
      "auxiliary_loss_clip": 0.0115923,
      "auxiliary_loss_mlp": 0.01029476,
      "balance_loss_clip": 1.05095673,
      "balance_loss_mlp": 1.02181101,
      "epoch": 0.9802200444898695,
      "flos": 16723635546240.0,
      "grad_norm": 4.663366368061629,
      "language_loss": 0.84927976,
      "learning_rate": 4.078714786211135e-09,
      "loss": 0.87116683,
      "num_input_tokens_seen": 175991340,
      "step": 8152,
      "time_per_iteration": 2.6626594066619873
    },
    {
      "auxiliary_loss_clip": 0.01158872,
      "auxiliary_loss_mlp": 0.0076174,
      "balance_loss_clip": 1.0501101,
      "balance_loss_mlp": 1.00053477,
      "epoch": 0.9803402873805086,
      "flos": 24900459298560.0,
      "grad_norm": 1.814189453700095,
      "language_loss": 0.7704097,
      "learning_rate": 4.029142666398977e-09,
      "loss": 0.78961581,
      "num_input_tokens_seen": 176011505,
      "step": 8153,
      "time_per_iteration": 2.7052648067474365
    },
    {
      "auxiliary_loss_clip": 0.01146962,
      "auxiliary_loss_mlp": 0.01029744,
      "balance_loss_clip": 1.05008745,
      "balance_loss_mlp": 1.02241218,
      "epoch": 0.9804605302711478,
      "flos": 22564937082240.0,
      "grad_norm": 2.0621765220994543,
      "language_loss": 0.80032969,
      "learning_rate": 3.979873333778805e-09,
      "loss": 0.82209671,
      "num_input_tokens_seen": 176029680,
      "step": 8154,
      "time_per_iteration": 2.652543306350708
    },
    {
      "auxiliary_loss_clip": 0.01122932,
      "auxiliary_loss_mlp": 0.010285,
      "balance_loss_clip": 1.04533529,
      "balance_loss_mlp": 1.02040517,
      "epoch": 0.9805807731617868,
      "flos": 38905368382080.0,
      "grad_norm": 1.9446494780390937,
      "language_loss": 0.73768353,
      "learning_rate": 3.930906795824862e-09,
      "loss": 0.75919783,
      "num_input_tokens_seen": 176050355,
      "step": 8155,
      "time_per_iteration": 2.8942248821258545
    },
    {
      "auxiliary_loss_clip": 0.01143375,
      "auxiliary_loss_mlp": 0.0076158,
      "balance_loss_clip": 1.04755092,
      "balance_loss_mlp": 1.0005008,
      "epoch": 0.9807010160524259,
      "flos": 17821999578240.0,
      "grad_norm": 2.507295443242597,
      "language_loss": 0.76829517,
      "learning_rate": 3.882243059965207e-09,
      "loss": 0.78734469,
      "num_input_tokens_seen": 176068070,
      "step": 8156,
      "time_per_iteration": 2.677896738052368
    },
    {
      "auxiliary_loss_clip": 0.01131483,
      "auxiliary_loss_mlp": 0.01030863,
      "balance_loss_clip": 1.04686356,
      "balance_loss_mlp": 1.02287543,
      "epoch": 0.980821258943065,
      "flos": 13552975140480.0,
      "grad_norm": 2.185949875175454,
      "language_loss": 0.66280687,
      "learning_rate": 3.833882133582156e-09,
      "loss": 0.6844303,
      "num_input_tokens_seen": 176083730,
      "step": 8157,
      "time_per_iteration": 3.7795562744140625
    },
    {
      "auxiliary_loss_clip": 0.01106117,
      "auxiliary_loss_mlp": 0.01025453,
      "balance_loss_clip": 1.04237115,
      "balance_loss_mlp": 1.01807666,
      "epoch": 0.9809415018337041,
      "flos": 21689794120320.0,
      "grad_norm": 1.786867897028345,
      "language_loss": 0.78281343,
      "learning_rate": 3.785824024012285e-09,
      "loss": 0.80412924,
      "num_input_tokens_seen": 176102730,
      "step": 8158,
      "time_per_iteration": 2.8128514289855957
    },
    {
      "auxiliary_loss_clip": 0.01146272,
      "auxiliary_loss_mlp": 0.01030894,
      "balance_loss_clip": 1.04824543,
      "balance_loss_mlp": 1.02344942,
      "epoch": 0.9810617447243432,
      "flos": 23294857357440.0,
      "grad_norm": 1.5984431674706898,
      "language_loss": 0.78849137,
      "learning_rate": 3.738068738545541e-09,
      "loss": 0.81026304,
      "num_input_tokens_seen": 176121815,
      "step": 8159,
      "time_per_iteration": 2.6648449897766113
    },
    {
      "auxiliary_loss_clip": 0.01160194,
      "auxiliary_loss_mlp": 0.0102673,
      "balance_loss_clip": 1.05171943,
      "balance_loss_mlp": 1.01895165,
      "epoch": 0.9811819876149822,
      "flos": 18332038748160.0,
      "grad_norm": 2.436379362537976,
      "language_loss": 0.7907362,
      "learning_rate": 3.6906162844265733e-09,
      "loss": 0.81260538,
      "num_input_tokens_seen": 176138900,
      "step": 8160,
      "time_per_iteration": 2.674351215362549
    },
    {
      "auxiliary_loss_clip": 0.0115722,
      "auxiliary_loss_mlp": 0.01024032,
      "balance_loss_clip": 1.04948378,
      "balance_loss_mlp": 1.01588976,
      "epoch": 0.9813022305056214,
      "flos": 22601961025920.0,
      "grad_norm": 1.888785450842392,
      "language_loss": 0.70921695,
      "learning_rate": 3.643466668853845e-09,
      "loss": 0.73102951,
      "num_input_tokens_seen": 176156925,
      "step": 8161,
      "time_per_iteration": 2.662444829940796
    },
    {
      "auxiliary_loss_clip": 0.01117189,
      "auxiliary_loss_mlp": 0.01029353,
      "balance_loss_clip": 1.04771948,
      "balance_loss_mlp": 1.02173829,
      "epoch": 0.9814224733962604,
      "flos": 25413335642880.0,
      "grad_norm": 3.0869130862708425,
      "language_loss": 0.75579226,
      "learning_rate": 3.59661989898008e-09,
      "loss": 0.77725768,
      "num_input_tokens_seen": 176177980,
      "step": 8162,
      "time_per_iteration": 2.821453809738159
    },
    {
      "auxiliary_loss_clip": 0.01126724,
      "auxiliary_loss_mlp": 0.01021739,
      "balance_loss_clip": 1.04507792,
      "balance_loss_mlp": 1.01454186,
      "epoch": 0.9815427162868995,
      "flos": 25007185584000.0,
      "grad_norm": 2.4002719265115005,
      "language_loss": 0.7698676,
      "learning_rate": 3.5500759819115934e-09,
      "loss": 0.79135227,
      "num_input_tokens_seen": 176198345,
      "step": 8163,
      "time_per_iteration": 2.78686261177063
    },
    {
      "auxiliary_loss_clip": 0.01119857,
      "auxiliary_loss_mlp": 0.01028506,
      "balance_loss_clip": 1.04576588,
      "balance_loss_mlp": 1.02049494,
      "epoch": 0.9816629591775387,
      "flos": 20662604887680.0,
      "grad_norm": 2.2803184675867088,
      "language_loss": 0.81412661,
      "learning_rate": 3.5038349247094034e-09,
      "loss": 0.83561027,
      "num_input_tokens_seen": 176215605,
      "step": 8164,
      "time_per_iteration": 2.6968793869018555
    },
    {
      "auxiliary_loss_clip": 0.01164346,
      "auxiliary_loss_mlp": 0.01031051,
      "balance_loss_clip": 1.04937053,
      "balance_loss_mlp": 1.02335012,
      "epoch": 0.9817832020681777,
      "flos": 17712220636800.0,
      "grad_norm": 3.344539239985238,
      "language_loss": 0.77463269,
      "learning_rate": 3.4578967343878994e-09,
      "loss": 0.79658663,
      "num_input_tokens_seen": 176231810,
      "step": 8165,
      "time_per_iteration": 3.5630509853363037
    },
    {
      "auxiliary_loss_clip": 0.01151024,
      "auxiliary_loss_mlp": 0.01025014,
      "balance_loss_clip": 1.05103886,
      "balance_loss_mlp": 1.01728606,
      "epoch": 0.9819034449588168,
      "flos": 22530032040960.0,
      "grad_norm": 2.1070377936987894,
      "language_loss": 0.81121767,
      "learning_rate": 3.4122614179161733e-09,
      "loss": 0.83297801,
      "num_input_tokens_seen": 176251770,
      "step": 8166,
      "time_per_iteration": 2.6790244579315186
    },
    {
      "auxiliary_loss_clip": 0.01124442,
      "auxiliary_loss_mlp": 0.00761464,
      "balance_loss_clip": 1.04317403,
      "balance_loss_mlp": 1.00055623,
      "epoch": 0.9820236878494559,
      "flos": 20011221699840.0,
      "grad_norm": 1.6596621913860312,
      "language_loss": 0.78175926,
      "learning_rate": 3.36692898221691e-09,
      "loss": 0.80061835,
      "num_input_tokens_seen": 176270135,
      "step": 8167,
      "time_per_iteration": 2.8481500148773193
    },
    {
      "auxiliary_loss_clip": 0.01157229,
      "auxiliary_loss_mlp": 0.01023653,
      "balance_loss_clip": 1.04979348,
      "balance_loss_mlp": 1.01644361,
      "epoch": 0.982143930740095,
      "flos": 18807316531200.0,
      "grad_norm": 1.9350982874103495,
      "language_loss": 0.73804015,
      "learning_rate": 3.3218994341668305e-09,
      "loss": 0.75984901,
      "num_input_tokens_seen": 176289065,
      "step": 8168,
      "time_per_iteration": 3.6200454235076904
    },
    {
      "auxiliary_loss_clip": 0.01094957,
      "auxiliary_loss_mlp": 0.01028892,
      "balance_loss_clip": 1.04293227,
      "balance_loss_mlp": 1.02185607,
      "epoch": 0.982264173630734,
      "flos": 26578026138240.0,
      "grad_norm": 1.5439511986776981,
      "language_loss": 0.75501943,
      "learning_rate": 3.2771727805971373e-09,
      "loss": 0.77625787,
      "num_input_tokens_seen": 176310450,
      "step": 8169,
      "time_per_iteration": 2.8446238040924072
    },
    {
      "auxiliary_loss_clip": 0.0112212,
      "auxiliary_loss_mlp": 0.01028608,
      "balance_loss_clip": 1.04509568,
      "balance_loss_mlp": 1.02127707,
      "epoch": 0.9823844165213732,
      "flos": 22014462176640.0,
      "grad_norm": 2.6197938620700585,
      "language_loss": 0.77210885,
      "learning_rate": 3.232749028292847e-09,
      "loss": 0.79361612,
      "num_input_tokens_seen": 176327415,
      "step": 8170,
      "time_per_iteration": 2.7421867847442627
    },
    {
      "auxiliary_loss_clip": 0.01158659,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.04807723,
      "balance_loss_mlp": 1.01856959,
      "epoch": 0.9825046594120123,
      "flos": 21908166854400.0,
      "grad_norm": 1.7702388152633775,
      "language_loss": 0.8841908,
      "learning_rate": 3.188628183992792e-09,
      "loss": 0.90604472,
      "num_input_tokens_seen": 176347680,
      "step": 8171,
      "time_per_iteration": 2.616847276687622
    },
    {
      "auxiliary_loss_clip": 0.01053486,
      "auxiliary_loss_mlp": 0.0099997,
      "balance_loss_clip": 1.0082432,
      "balance_loss_mlp": 0.99909383,
      "epoch": 0.9826249023026513,
      "flos": 59494610718720.0,
      "grad_norm": 0.7359843693288204,
      "language_loss": 0.62485927,
      "learning_rate": 3.1448102543902844e-09,
      "loss": 0.64539385,
      "num_input_tokens_seen": 176411595,
      "step": 8172,
      "time_per_iteration": 3.1669251918792725
    },
    {
      "auxiliary_loss_clip": 0.01143626,
      "auxiliary_loss_mlp": 0.01034123,
      "balance_loss_clip": 1.05000424,
      "balance_loss_mlp": 1.02633524,
      "epoch": 0.9827451451932905,
      "flos": 16071031296000.0,
      "grad_norm": 3.46778788767887,
      "language_loss": 0.6801728,
      "learning_rate": 3.1012952461324515e-09,
      "loss": 0.70195031,
      "num_input_tokens_seen": 176430570,
      "step": 8173,
      "time_per_iteration": 2.6961402893066406
    },
    {
      "auxiliary_loss_clip": 0.01172745,
      "auxiliary_loss_mlp": 0.01027715,
      "balance_loss_clip": 1.05263603,
      "balance_loss_mlp": 1.02005601,
      "epoch": 0.9828653880839295,
      "flos": 20262775622400.0,
      "grad_norm": 2.745886955969787,
      "language_loss": 0.74047858,
      "learning_rate": 3.0580831658204575e-09,
      "loss": 0.76248324,
      "num_input_tokens_seen": 176448150,
      "step": 8174,
      "time_per_iteration": 3.514190912246704
    },
    {
      "auxiliary_loss_clip": 0.01084998,
      "auxiliary_loss_mlp": 0.01024519,
      "balance_loss_clip": 1.03647411,
      "balance_loss_mlp": 1.01688361,
      "epoch": 0.9829856309745686,
      "flos": 21616141282560.0,
      "grad_norm": 2.1731966888033405,
      "language_loss": 0.78101623,
      "learning_rate": 3.015174020009281e-09,
      "loss": 0.80211139,
      "num_input_tokens_seen": 176467475,
      "step": 8175,
      "time_per_iteration": 2.7919504642486572
    },
    {
      "auxiliary_loss_clip": 0.01104369,
      "auxiliary_loss_mlp": 0.01029151,
      "balance_loss_clip": 1.04418957,
      "balance_loss_mlp": 1.0211637,
      "epoch": 0.9831058738652078,
      "flos": 23764209396480.0,
      "grad_norm": 1.8451516876109975,
      "language_loss": 0.75301075,
      "learning_rate": 2.9725678152086043e-09,
      "loss": 0.77434599,
      "num_input_tokens_seen": 176486045,
      "step": 8176,
      "time_per_iteration": 2.741631031036377
    },
    {
      "auxiliary_loss_clip": 0.01147542,
      "auxiliary_loss_mlp": 0.01032287,
      "balance_loss_clip": 1.04634976,
      "balance_loss_mlp": 1.02473783,
      "epoch": 0.9832261167558468,
      "flos": 11320911072000.0,
      "grad_norm": 3.2296254289514894,
      "language_loss": 0.82541573,
      "learning_rate": 2.930264557881257e-09,
      "loss": 0.8472141,
      "num_input_tokens_seen": 176501230,
      "step": 8177,
      "time_per_iteration": 2.682802438735962
    },
    {
      "auxiliary_loss_clip": 0.01061393,
      "auxiliary_loss_mlp": 0.01001115,
      "balance_loss_clip": 1.00774884,
      "balance_loss_mlp": 1.00023925,
      "epoch": 0.9833463596464859,
      "flos": 60000304343040.0,
      "grad_norm": 0.82596164425706,
      "language_loss": 0.58054388,
      "learning_rate": 2.8882642544452163e-09,
      "loss": 0.60116899,
      "num_input_tokens_seen": 176565955,
      "step": 8178,
      "time_per_iteration": 3.2462377548217773
    },
    {
      "auxiliary_loss_clip": 0.0113975,
      "auxiliary_loss_mlp": 0.01025982,
      "balance_loss_clip": 1.04649234,
      "balance_loss_mlp": 1.01863229,
      "epoch": 0.983466602537125,
      "flos": 13626699805440.0,
      "grad_norm": 2.6898842974188297,
      "language_loss": 0.74614924,
      "learning_rate": 2.8465669112716083e-09,
      "loss": 0.76780659,
      "num_input_tokens_seen": 176583480,
      "step": 8179,
      "time_per_iteration": 2.7280027866363525
    },
    {
      "auxiliary_loss_clip": 0.0115159,
      "auxiliary_loss_mlp": 0.01031486,
      "balance_loss_clip": 1.05107105,
      "balance_loss_mlp": 1.02376103,
      "epoch": 0.9835868454277641,
      "flos": 22926844563840.0,
      "grad_norm": 1.9143393671641669,
      "language_loss": 0.76247394,
      "learning_rate": 2.8051725346858177e-09,
      "loss": 0.78430474,
      "num_input_tokens_seen": 176603740,
      "step": 8180,
      "time_per_iteration": 2.7277824878692627
    },
    {
      "auxiliary_loss_clip": 0.0115976,
      "auxiliary_loss_mlp": 0.01025581,
      "balance_loss_clip": 1.04984939,
      "balance_loss_mlp": 1.01783788,
      "epoch": 0.9837070883184031,
      "flos": 27673409341440.0,
      "grad_norm": 1.9991878975945898,
      "language_loss": 0.71284175,
      "learning_rate": 2.7640811309674883e-09,
      "loss": 0.73469514,
      "num_input_tokens_seen": 176623240,
      "step": 8181,
      "time_per_iteration": 2.6825313568115234
    },
    {
      "auxiliary_loss_clip": 0.01149633,
      "auxiliary_loss_mlp": 0.0102667,
      "balance_loss_clip": 1.05183816,
      "balance_loss_mlp": 1.01904321,
      "epoch": 0.9838273312090423,
      "flos": 29241951425280.0,
      "grad_norm": 1.647223648129801,
      "language_loss": 0.80878192,
      "learning_rate": 2.7232927063498557e-09,
      "loss": 0.83054495,
      "num_input_tokens_seen": 176643615,
      "step": 8182,
      "time_per_iteration": 2.8058013916015625
    },
    {
      "auxiliary_loss_clip": 0.01151542,
      "auxiliary_loss_mlp": 0.01026555,
      "balance_loss_clip": 1.04807842,
      "balance_loss_mlp": 1.01769781,
      "epoch": 0.9839475740996814,
      "flos": 40110207304320.0,
      "grad_norm": 2.786960517191089,
      "language_loss": 0.68819213,
      "learning_rate": 2.682807267020859e-09,
      "loss": 0.70997316,
      "num_input_tokens_seen": 176666375,
      "step": 8183,
      "time_per_iteration": 3.8060195446014404
    },
    {
      "auxiliary_loss_clip": 0.01077495,
      "auxiliary_loss_mlp": 0.01024461,
      "balance_loss_clip": 1.03518414,
      "balance_loss_mlp": 1.01652157,
      "epoch": 0.9840678169903204,
      "flos": 24169389788160.0,
      "grad_norm": 1.656113085766858,
      "language_loss": 0.62561166,
      "learning_rate": 2.642624819121808e-09,
      "loss": 0.64663124,
      "num_input_tokens_seen": 176686525,
      "step": 8184,
      "time_per_iteration": 2.94246768951416
    },
    {
      "auxiliary_loss_clip": 0.01160478,
      "auxiliary_loss_mlp": 0.01027944,
      "balance_loss_clip": 1.05123162,
      "balance_loss_mlp": 1.02094316,
      "epoch": 0.9841880598809596,
      "flos": 14684484447360.0,
      "grad_norm": 1.9084100684867231,
      "language_loss": 0.62074536,
      "learning_rate": 2.6027453687487154e-09,
      "loss": 0.64262962,
      "num_input_tokens_seen": 176703615,
      "step": 8185,
      "time_per_iteration": 2.9862589836120605
    },
    {
      "auxiliary_loss_clip": 0.01160571,
      "auxiliary_loss_mlp": 0.01031619,
      "balance_loss_clip": 1.04990911,
      "balance_loss_mlp": 1.02384019,
      "epoch": 0.9843083027715986,
      "flos": 22344768668160.0,
      "grad_norm": 2.9844125840270475,
      "language_loss": 0.54366004,
      "learning_rate": 2.5631689219509643e-09,
      "loss": 0.56558192,
      "num_input_tokens_seen": 176722295,
      "step": 8186,
      "time_per_iteration": 2.638843059539795
    },
    {
      "auxiliary_loss_clip": 0.01144172,
      "auxiliary_loss_mlp": 0.01027507,
      "balance_loss_clip": 1.04715872,
      "balance_loss_mlp": 1.02045274,
      "epoch": 0.9844285456622377,
      "flos": 21800111765760.0,
      "grad_norm": 1.895230870942853,
      "language_loss": 0.8354584,
      "learning_rate": 2.523895484732197e-09,
      "loss": 0.85717523,
      "num_input_tokens_seen": 176741750,
      "step": 8187,
      "time_per_iteration": 2.695589542388916
    },
    {
      "auxiliary_loss_clip": 0.01161719,
      "auxiliary_loss_mlp": 0.01034463,
      "balance_loss_clip": 1.05019546,
      "balance_loss_mlp": 1.02636886,
      "epoch": 0.9845487885528769,
      "flos": 18035380321920.0,
      "grad_norm": 3.267074629206297,
      "language_loss": 0.74765253,
      "learning_rate": 2.4849250630505357e-09,
      "loss": 0.7696144,
      "num_input_tokens_seen": 176759995,
      "step": 8188,
      "time_per_iteration": 2.6953675746917725
    },
    {
      "auxiliary_loss_clip": 0.01147178,
      "auxiliary_loss_mlp": 0.00761749,
      "balance_loss_clip": 1.04968035,
      "balance_loss_mlp": 1.00045156,
      "epoch": 0.9846690314435159,
      "flos": 25228610974080.0,
      "grad_norm": 1.986292862989593,
      "language_loss": 0.7337665,
      "learning_rate": 2.4462576628172528e-09,
      "loss": 0.75285578,
      "num_input_tokens_seen": 176778625,
      "step": 8189,
      "time_per_iteration": 2.7559380531311035
    },
    {
      "auxiliary_loss_clip": 0.01135752,
      "auxiliary_loss_mlp": 0.0102781,
      "balance_loss_clip": 1.04653287,
      "balance_loss_mlp": 1.02023411,
      "epoch": 0.984789274334155,
      "flos": 18552171248640.0,
      "grad_norm": 2.4176143331865165,
      "language_loss": 0.73977029,
      "learning_rate": 2.407893289898766e-09,
      "loss": 0.76140594,
      "num_input_tokens_seen": 176797655,
      "step": 8190,
      "time_per_iteration": 2.6964197158813477
    },
    {
      "auxiliary_loss_clip": 0.01078871,
      "auxiliary_loss_mlp": 0.01031317,
      "balance_loss_clip": 1.03667712,
      "balance_loss_mlp": 1.02367878,
      "epoch": 0.984909517224794,
      "flos": 27345437233920.0,
      "grad_norm": 2.0099740398197485,
      "language_loss": 0.84209943,
      "learning_rate": 2.3698319501144202e-09,
      "loss": 0.86320126,
      "num_input_tokens_seen": 176818640,
      "step": 8191,
      "time_per_iteration": 3.8029282093048096
    },
    {
      "auxiliary_loss_clip": 0.01143719,
      "auxiliary_loss_mlp": 0.01033206,
      "balance_loss_clip": 1.04579616,
      "balance_loss_mlp": 1.0244019,
      "epoch": 0.9850297601154332,
      "flos": 18734058743040.0,
      "grad_norm": 1.6657461250056103,
      "language_loss": 0.73795426,
      "learning_rate": 2.3320736492382644e-09,
      "loss": 0.75972354,
      "num_input_tokens_seen": 176837475,
      "step": 8192,
      "time_per_iteration": 2.6602835655212402
    },
    {
      "auxiliary_loss_clip": 0.01150308,
      "auxiliary_loss_mlp": 0.0103048,
      "balance_loss_clip": 1.04841018,
      "balance_loss_mlp": 1.02277327,
      "epoch": 0.9851500030060723,
      "flos": 22308247514880.0,
      "grad_norm": 1.6257451821721023,
      "language_loss": 0.68109429,
      "learning_rate": 2.29461839299816e-09,
      "loss": 0.70290214,
      "num_input_tokens_seen": 176857190,
      "step": 8193,
      "time_per_iteration": 2.70255184173584
    },
    {
      "auxiliary_loss_clip": 0.01159502,
      "auxiliary_loss_mlp": 0.01029163,
      "balance_loss_clip": 1.04937363,
      "balance_loss_mlp": 1.02172482,
      "epoch": 0.9852702458967113,
      "flos": 26353691746560.0,
      "grad_norm": 1.6241736645850557,
      "language_loss": 0.80062002,
      "learning_rate": 2.257466187076229e-09,
      "loss": 0.82250667,
      "num_input_tokens_seen": 176876395,
      "step": 8194,
      "time_per_iteration": 3.6606574058532715
    },
    {
      "auxiliary_loss_clip": 0.01131185,
      "auxiliary_loss_mlp": 0.01033494,
      "balance_loss_clip": 1.05125856,
      "balance_loss_mlp": 1.02493465,
      "epoch": 0.9853904887873505,
      "flos": 20883599314560.0,
      "grad_norm": 1.687595435502474,
      "language_loss": 0.71364295,
      "learning_rate": 2.2206170371081854e-09,
      "loss": 0.73528975,
      "num_input_tokens_seen": 176894980,
      "step": 8195,
      "time_per_iteration": 2.786346673965454
    },
    {
      "auxiliary_loss_clip": 0.01093855,
      "auxiliary_loss_mlp": 0.01031442,
      "balance_loss_clip": 1.04504728,
      "balance_loss_mlp": 1.02338934,
      "epoch": 0.9855107316779895,
      "flos": 25263444188160.0,
      "grad_norm": 1.5970988031504483,
      "language_loss": 0.85023171,
      "learning_rate": 2.1840709486842247e-09,
      "loss": 0.87148476,
      "num_input_tokens_seen": 176914600,
      "step": 8196,
      "time_per_iteration": 2.805500030517578
    },
    {
      "auxiliary_loss_clip": 0.01163264,
      "auxiliary_loss_mlp": 0.01029662,
      "balance_loss_clip": 1.04928303,
      "balance_loss_mlp": 1.02147198,
      "epoch": 0.9856309745686286,
      "flos": 19062102677760.0,
      "grad_norm": 2.2215505649911407,
      "language_loss": 0.79753155,
      "learning_rate": 2.1478279273481335e-09,
      "loss": 0.81946075,
      "num_input_tokens_seen": 176933085,
      "step": 8197,
      "time_per_iteration": 2.6766700744628906
    },
    {
      "auxiliary_loss_clip": 0.01159193,
      "auxiliary_loss_mlp": 0.01030811,
      "balance_loss_clip": 1.05047357,
      "balance_loss_mlp": 1.02275896,
      "epoch": 0.9857512174592677,
      "flos": 34130758060800.0,
      "grad_norm": 2.1309006923740945,
      "language_loss": 0.80233771,
      "learning_rate": 2.1118879785981815e-09,
      "loss": 0.82423776,
      "num_input_tokens_seen": 176953225,
      "step": 8198,
      "time_per_iteration": 2.73393177986145
    },
    {
      "auxiliary_loss_clip": 0.01128997,
      "auxiliary_loss_mlp": 0.01022574,
      "balance_loss_clip": 1.04853952,
      "balance_loss_mlp": 1.01561546,
      "epoch": 0.9858714603499068,
      "flos": 25994693266560.0,
      "grad_norm": 1.9110512700838653,
      "language_loss": 0.79650789,
      "learning_rate": 2.0762511078862288e-09,
      "loss": 0.81802368,
      "num_input_tokens_seen": 176973570,
      "step": 8199,
      "time_per_iteration": 3.6372251510620117
    },
    {
      "auxiliary_loss_clip": 0.01146159,
      "auxiliary_loss_mlp": 0.01027028,
      "balance_loss_clip": 1.04472649,
      "balance_loss_mlp": 1.01934481,
      "epoch": 0.9859917032405459,
      "flos": 23696230907520.0,
      "grad_norm": 1.839463508755099,
      "language_loss": 0.65112734,
      "learning_rate": 2.0409173206186183e-09,
      "loss": 0.67285919,
      "num_input_tokens_seen": 176992810,
      "step": 8200,
      "time_per_iteration": 2.713904619216919
    },
    {
      "auxiliary_loss_clip": 0.01122663,
      "auxiliary_loss_mlp": 0.01025628,
      "balance_loss_clip": 1.0435679,
      "balance_loss_mlp": 1.01833487,
      "epoch": 0.986111946131185,
      "flos": 19938287134080.0,
      "grad_norm": 3.265251790720697,
      "language_loss": 0.87452585,
      "learning_rate": 2.0058866221550617e-09,
      "loss": 0.89600873,
      "num_input_tokens_seen": 177011050,
      "step": 8201,
      "time_per_iteration": 2.7275960445404053
    },
    {
      "auxiliary_loss_clip": 0.01059982,
      "auxiliary_loss_mlp": 0.01027932,
      "balance_loss_clip": 1.03478765,
      "balance_loss_mlp": 1.02012336,
      "epoch": 0.9862321890218241,
      "flos": 19828831415040.0,
      "grad_norm": 1.9700477955745643,
      "language_loss": 0.75772226,
      "learning_rate": 1.971159017809976e-09,
      "loss": 0.77860141,
      "num_input_tokens_seen": 177029340,
      "step": 8202,
      "time_per_iteration": 3.0726535320281982
    },
    {
      "auxiliary_loss_clip": 0.0114305,
      "auxiliary_loss_mlp": 0.01030277,
      "balance_loss_clip": 1.04981792,
      "balance_loss_mlp": 1.0223372,
      "epoch": 0.9863524319124631,
      "flos": 21652051904640.0,
      "grad_norm": 2.568742557230279,
      "language_loss": 0.77556324,
      "learning_rate": 1.93673451285159e-09,
      "loss": 0.79729652,
      "num_input_tokens_seen": 177048390,
      "step": 8203,
      "time_per_iteration": 3.1970489025115967
    },
    {
      "auxiliary_loss_clip": 0.0104151,
      "auxiliary_loss_mlp": 0.00999981,
      "balance_loss_clip": 1.00699472,
      "balance_loss_mlp": 0.99902767,
      "epoch": 0.9864726748031023,
      "flos": 52769977920000.0,
      "grad_norm": 0.7453498672942604,
      "language_loss": 0.56479198,
      "learning_rate": 1.9026131125019495e-09,
      "loss": 0.58520687,
      "num_input_tokens_seen": 177105760,
      "step": 8204,
      "time_per_iteration": 3.204962968826294
    },
    {
      "auxiliary_loss_clip": 0.01139785,
      "auxiliary_loss_mlp": 0.01027749,
      "balance_loss_clip": 1.04835689,
      "balance_loss_mlp": 1.02007496,
      "epoch": 0.9865929176937414,
      "flos": 23364631526400.0,
      "grad_norm": 1.8721775298039864,
      "language_loss": 0.86825293,
      "learning_rate": 1.8687948219371363e-09,
      "loss": 0.88992822,
      "num_input_tokens_seen": 177124985,
      "step": 8205,
      "time_per_iteration": 2.646322011947632
    },
    {
      "auxiliary_loss_clip": 0.0115291,
      "auxiliary_loss_mlp": 0.01029305,
      "balance_loss_clip": 1.04767084,
      "balance_loss_mlp": 1.02112758,
      "epoch": 0.9867131605843804,
      "flos": 21616679986560.0,
      "grad_norm": 2.417881691761291,
      "language_loss": 0.88703048,
      "learning_rate": 1.835279646287491e-09,
      "loss": 0.9088527,
      "num_input_tokens_seen": 177142995,
      "step": 8206,
      "time_per_iteration": 2.6405529975891113
    },
    {
      "auxiliary_loss_clip": 0.01127646,
      "auxiliary_loss_mlp": 0.01034468,
      "balance_loss_clip": 1.05039179,
      "balance_loss_mlp": 1.02624869,
      "epoch": 0.9868334034750196,
      "flos": 22271403139200.0,
      "grad_norm": 2.0256687884211386,
      "language_loss": 0.76489031,
      "learning_rate": 1.8020675906371685e-09,
      "loss": 0.78651142,
      "num_input_tokens_seen": 177162390,
      "step": 8207,
      "time_per_iteration": 2.7449984550476074
    },
    {
      "auxiliary_loss_clip": 0.01172534,
      "auxiliary_loss_mlp": 0.0076184,
      "balance_loss_clip": 1.05131221,
      "balance_loss_mlp": 1.00049233,
      "epoch": 0.9869536463656586,
      "flos": 25809573548160.0,
      "grad_norm": 2.1403539502537483,
      "language_loss": 0.7528162,
      "learning_rate": 1.7691586600243612e-09,
      "loss": 0.77215999,
      "num_input_tokens_seen": 177181290,
      "step": 8208,
      "time_per_iteration": 2.7139272689819336
    },
    {
      "auxiliary_loss_clip": 0.01163431,
      "auxiliary_loss_mlp": 0.01026509,
      "balance_loss_clip": 1.05027115,
      "balance_loss_mlp": 1.01929069,
      "epoch": 0.9870738892562977,
      "flos": 16398500613120.0,
      "grad_norm": 2.8648686659304405,
      "language_loss": 0.87185478,
      "learning_rate": 1.7365528594415202e-09,
      "loss": 0.89375412,
      "num_input_tokens_seen": 177195360,
      "step": 8209,
      "time_per_iteration": 3.86102032661438
    },
    {
      "auxiliary_loss_clip": 0.0114041,
      "auxiliary_loss_mlp": 0.01030077,
      "balance_loss_clip": 1.04634106,
      "balance_loss_mlp": 1.02224469,
      "epoch": 0.9871941321469369,
      "flos": 35481358373760.0,
      "grad_norm": 1.5883976091664056,
      "language_loss": 0.67728245,
      "learning_rate": 1.7042501938346888e-09,
      "loss": 0.69898736,
      "num_input_tokens_seen": 177218090,
      "step": 8210,
      "time_per_iteration": 2.8710131645202637
    },
    {
      "auxiliary_loss_clip": 0.01108654,
      "auxiliary_loss_mlp": 0.01030681,
      "balance_loss_clip": 1.04384279,
      "balance_loss_mlp": 1.02321291,
      "epoch": 0.9873143750375759,
      "flos": 21434217874560.0,
      "grad_norm": 1.9542810341933967,
      "language_loss": 0.76429468,
      "learning_rate": 1.6722506681043913e-09,
      "loss": 0.78568804,
      "num_input_tokens_seen": 177237050,
      "step": 8211,
      "time_per_iteration": 2.79115891456604
    },
    {
      "auxiliary_loss_clip": 0.01145282,
      "auxiliary_loss_mlp": 0.01025447,
      "balance_loss_clip": 1.04972863,
      "balance_loss_mlp": 1.0174545,
      "epoch": 0.987434617928215,
      "flos": 16326499800960.0,
      "grad_norm": 4.063035039441665,
      "language_loss": 0.69584632,
      "learning_rate": 1.640554287104745e-09,
      "loss": 0.71755362,
      "num_input_tokens_seen": 177255325,
      "step": 8212,
      "time_per_iteration": 2.645026445388794
    },
    {
      "auxiliary_loss_clip": 0.01110795,
      "auxiliary_loss_mlp": 0.01028882,
      "balance_loss_clip": 1.04089737,
      "balance_loss_mlp": 1.02046573,
      "epoch": 0.9875548608188541,
      "flos": 17851984456320.0,
      "grad_norm": 2.2960527238595536,
      "language_loss": 0.80322999,
      "learning_rate": 1.609161055644348e-09,
      "loss": 0.8246268,
      "num_input_tokens_seen": 177271250,
      "step": 8213,
      "time_per_iteration": 2.7235267162323
    },
    {
      "auxiliary_loss_clip": 0.01124865,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.04389715,
      "balance_loss_mlp": 1.01954842,
      "epoch": 0.9876751037094932,
      "flos": 26132876887680.0,
      "grad_norm": 2.2307557346094447,
      "language_loss": 0.68404162,
      "learning_rate": 1.5780709784849467e-09,
      "loss": 0.70556748,
      "num_input_tokens_seen": 177288270,
      "step": 8214,
      "time_per_iteration": 2.7016162872314453
    },
    {
      "auxiliary_loss_clip": 0.01138319,
      "auxiliary_loss_mlp": 0.01026498,
      "balance_loss_clip": 1.04550529,
      "balance_loss_mlp": 1.0185225,
      "epoch": 0.9877953466001322,
      "flos": 15991344973440.0,
      "grad_norm": 1.932538830193701,
      "language_loss": 0.82072252,
      "learning_rate": 1.5472840603436565e-09,
      "loss": 0.84237069,
      "num_input_tokens_seen": 177305500,
      "step": 8215,
      "time_per_iteration": 2.69634747505188
    },
    {
      "auxiliary_loss_clip": 0.01101684,
      "auxiliary_loss_mlp": 0.01026865,
      "balance_loss_clip": 1.04011464,
      "balance_loss_mlp": 1.01905727,
      "epoch": 0.9879155894907714,
      "flos": 18806777827200.0,
      "grad_norm": 1.9565947854037882,
      "language_loss": 0.78447711,
      "learning_rate": 1.5168003058900757e-09,
      "loss": 0.80576259,
      "num_input_tokens_seen": 177323500,
      "step": 8216,
      "time_per_iteration": 2.873119592666626
    },
    {
      "auxiliary_loss_clip": 0.01156328,
      "auxiliary_loss_mlp": 0.01027644,
      "balance_loss_clip": 1.04876781,
      "balance_loss_mlp": 1.02057195,
      "epoch": 0.9880358323814105,
      "flos": 22382044007040.0,
      "grad_norm": 2.2280756600497016,
      "language_loss": 0.91932058,
      "learning_rate": 1.4866197197491715e-09,
      "loss": 0.94116032,
      "num_input_tokens_seen": 177342860,
      "step": 8217,
      "time_per_iteration": 3.84879469871521
    },
    {
      "auxiliary_loss_clip": 0.0113676,
      "auxiliary_loss_mlp": 0.00762415,
      "balance_loss_clip": 1.04879951,
      "balance_loss_mlp": 1.0004313,
      "epoch": 0.9881560752720495,
      "flos": 15668831733120.0,
      "grad_norm": 7.038433213728763,
      "language_loss": 0.7851541,
      "learning_rate": 1.4567423064988371e-09,
      "loss": 0.80414587,
      "num_input_tokens_seen": 177360210,
      "step": 8218,
      "time_per_iteration": 2.7283384799957275
    },
    {
      "auxiliary_loss_clip": 0.01137412,
      "auxiliary_loss_mlp": 0.01031927,
      "balance_loss_clip": 1.0463872,
      "balance_loss_mlp": 1.02413082,
      "epoch": 0.9882763181626887,
      "flos": 21500113374720.0,
      "grad_norm": 2.181861739234128,
      "language_loss": 0.78535187,
      "learning_rate": 1.4271680706718913e-09,
      "loss": 0.80704528,
      "num_input_tokens_seen": 177377885,
      "step": 8219,
      "time_per_iteration": 3.6617605686187744
    },
    {
      "auxiliary_loss_clip": 0.01095594,
      "auxiliary_loss_mlp": 0.00762616,
      "balance_loss_clip": 1.03844428,
      "balance_loss_mlp": 1.00045419,
      "epoch": 0.9883965610533277,
      "flos": 28034598551040.0,
      "grad_norm": 1.7084635074367867,
      "language_loss": 0.82804996,
      "learning_rate": 1.3978970167543013e-09,
      "loss": 0.84663206,
      "num_input_tokens_seen": 177398065,
      "step": 8220,
      "time_per_iteration": 2.8608756065368652
    },
    {
      "auxiliary_loss_clip": 0.01164522,
      "auxiliary_loss_mlp": 0.01029105,
      "balance_loss_clip": 1.05055523,
      "balance_loss_mlp": 1.02104998,
      "epoch": 0.9885168039439668,
      "flos": 14098601710080.0,
      "grad_norm": 2.6944269171897455,
      "language_loss": 0.77914906,
      "learning_rate": 1.3689291491867372e-09,
      "loss": 0.80108535,
      "num_input_tokens_seen": 177416380,
      "step": 8221,
      "time_per_iteration": 2.682530641555786
    },
    {
      "auxiliary_loss_clip": 0.0110041,
      "auxiliary_loss_mlp": 0.0076234,
      "balance_loss_clip": 1.03858805,
      "balance_loss_mlp": 1.00048423,
      "epoch": 0.988637046834606,
      "flos": 26432013352320.0,
      "grad_norm": 1.8518468893415196,
      "language_loss": 0.73647934,
      "learning_rate": 1.3402644723636836e-09,
      "loss": 0.75510681,
      "num_input_tokens_seen": 177438410,
      "step": 8222,
      "time_per_iteration": 2.8858296871185303
    },
    {
      "auxiliary_loss_clip": 0.01136717,
      "auxiliary_loss_mlp": 0.01029586,
      "balance_loss_clip": 1.0464884,
      "balance_loss_mlp": 1.0218972,
      "epoch": 0.988757289725245,
      "flos": 25229113764480.0,
      "grad_norm": 2.605750660467446,
      "language_loss": 0.83667088,
      "learning_rate": 1.311902990633218e-09,
      "loss": 0.85833389,
      "num_input_tokens_seen": 177457375,
      "step": 8223,
      "time_per_iteration": 2.769533395767212
    },
    {
      "auxiliary_loss_clip": 0.01145809,
      "auxiliary_loss_mlp": 0.01031757,
      "balance_loss_clip": 1.04621339,
      "balance_loss_mlp": 1.02410388,
      "epoch": 0.9888775326158841,
      "flos": 26359042872960.0,
      "grad_norm": 1.5647698333554638,
      "language_loss": 0.71079409,
      "learning_rate": 1.2838447082978987e-09,
      "loss": 0.73256981,
      "num_input_tokens_seen": 177478530,
      "step": 8224,
      "time_per_iteration": 2.791344404220581
    },
    {
      "auxiliary_loss_clip": 0.01115198,
      "auxiliary_loss_mlp": 0.0102577,
      "balance_loss_clip": 1.041085,
      "balance_loss_mlp": 1.01775026,
      "epoch": 0.9889977755065231,
      "flos": 24316120846080.0,
      "grad_norm": 2.7600399221018455,
      "language_loss": 0.83287406,
      "learning_rate": 1.2560896296143208e-09,
      "loss": 0.85428369,
      "num_input_tokens_seen": 177496995,
      "step": 8225,
      "time_per_iteration": 3.7288553714752197
    },
    {
      "auxiliary_loss_clip": 0.01134172,
      "auxiliary_loss_mlp": 0.0102701,
      "balance_loss_clip": 1.0458802,
      "balance_loss_mlp": 1.01939523,
      "epoch": 0.9891180183971623,
      "flos": 18951066760320.0,
      "grad_norm": 3.722056508349943,
      "language_loss": 0.82727098,
      "learning_rate": 1.2286377587926722e-09,
      "loss": 0.84888285,
      "num_input_tokens_seen": 177513785,
      "step": 8226,
      "time_per_iteration": 2.718928337097168
    },
    {
      "auxiliary_loss_clip": 0.01121726,
      "auxiliary_loss_mlp": 0.01025675,
      "balance_loss_clip": 1.04458451,
      "balance_loss_mlp": 1.01803994,
      "epoch": 0.9892382612878013,
      "flos": 26176580760960.0,
      "grad_norm": 2.013737135170844,
      "language_loss": 0.74964988,
      "learning_rate": 1.2014890999973992e-09,
      "loss": 0.77112389,
      "num_input_tokens_seen": 177530705,
      "step": 8227,
      "time_per_iteration": 2.788259506225586
    },
    {
      "auxiliary_loss_clip": 0.01122581,
      "auxiliary_loss_mlp": 0.00761575,
      "balance_loss_clip": 1.04350209,
      "balance_loss_mlp": 1.00053239,
      "epoch": 0.9893585041784404,
      "flos": 25449605400960.0,
      "grad_norm": 1.7747755808861712,
      "language_loss": 0.7836408,
      "learning_rate": 1.1746436573472073e-09,
      "loss": 0.80248237,
      "num_input_tokens_seen": 177552440,
      "step": 8228,
      "time_per_iteration": 2.7910234928131104
    },
    {
      "auxiliary_loss_clip": 0.010382,
      "auxiliary_loss_mlp": 0.01023658,
      "balance_loss_clip": 1.0352633,
      "balance_loss_mlp": 1.01639867,
      "epoch": 0.9894787470690796,
      "flos": 20189302352640.0,
      "grad_norm": 1.9244348864791698,
      "language_loss": 0.68851101,
      "learning_rate": 1.1481014349141726e-09,
      "loss": 0.70912957,
      "num_input_tokens_seen": 177569660,
      "step": 8229,
      "time_per_iteration": 3.1805641651153564
    },
    {
      "auxiliary_loss_clip": 0.01120026,
      "auxiliary_loss_mlp": 0.01028203,
      "balance_loss_clip": 1.04344332,
      "balance_loss_mlp": 1.0193758,
      "epoch": 0.9895989899597186,
      "flos": 24644308435200.0,
      "grad_norm": 1.9256872277898902,
      "language_loss": 0.84404469,
      "learning_rate": 1.121862436724852e-09,
      "loss": 0.86552691,
      "num_input_tokens_seen": 177588500,
      "step": 8230,
      "time_per_iteration": 3.1446549892425537
    },
    {
      "auxiliary_loss_clip": 0.0114002,
      "auxiliary_loss_mlp": 0.00761712,
      "balance_loss_clip": 1.04947376,
      "balance_loss_mlp": 1.00050569,
      "epoch": 0.9897192328503577,
      "flos": 21799034357760.0,
      "grad_norm": 1.9916244293057483,
      "language_loss": 0.70655787,
      "learning_rate": 1.0959266667598388e-09,
      "loss": 0.72557521,
      "num_input_tokens_seen": 177607315,
      "step": 8231,
      "time_per_iteration": 2.7074997425079346
    },
    {
      "auxiliary_loss_clip": 0.01146687,
      "auxiliary_loss_mlp": 0.01024861,
      "balance_loss_clip": 1.05064654,
      "balance_loss_mlp": 1.01540136,
      "epoch": 0.9898394757409968,
      "flos": 21325229032320.0,
      "grad_norm": 2.376471061985066,
      "language_loss": 0.7500726,
      "learning_rate": 1.0702941289533196e-09,
      "loss": 0.77178806,
      "num_input_tokens_seen": 177625990,
      "step": 8232,
      "time_per_iteration": 2.6584689617156982
    },
    {
      "auxiliary_loss_clip": 0.01155478,
      "auxiliary_loss_mlp": 0.01026954,
      "balance_loss_clip": 1.04759359,
      "balance_loss_mlp": 1.01909506,
      "epoch": 0.9899597186316359,
      "flos": 18545024442240.0,
      "grad_norm": 2.414108439184306,
      "language_loss": 0.89266062,
      "learning_rate": 1.0449648271939615e-09,
      "loss": 0.91448492,
      "num_input_tokens_seen": 177642335,
      "step": 8233,
      "time_per_iteration": 2.7502784729003906
    },
    {
      "auxiliary_loss_clip": 0.01140844,
      "auxiliary_loss_mlp": 0.01030413,
      "balance_loss_clip": 1.04666412,
      "balance_loss_mlp": 1.02232468,
      "epoch": 0.990079961522275,
      "flos": 23766723348480.0,
      "grad_norm": 2.119492364987142,
      "language_loss": 0.73271638,
      "learning_rate": 1.0199387653240243e-09,
      "loss": 0.75442892,
      "num_input_tokens_seen": 177662025,
      "step": 8234,
      "time_per_iteration": 4.279987335205078
    },
    {
      "auxiliary_loss_clip": 0.01149674,
      "auxiliary_loss_mlp": 0.00761617,
      "balance_loss_clip": 1.04701805,
      "balance_loss_mlp": 1.00054634,
      "epoch": 0.9902002044129141,
      "flos": 16399182971520.0,
      "grad_norm": 1.8107109984356975,
      "language_loss": 0.71039242,
      "learning_rate": 9.952159471400267e-10,
      "loss": 0.7295053,
      "num_input_tokens_seen": 177679065,
      "step": 8235,
      "time_per_iteration": 2.7548341751098633
    },
    {
      "auxiliary_loss_clip": 0.01143558,
      "auxiliary_loss_mlp": 0.01027637,
      "balance_loss_clip": 1.04704881,
      "balance_loss_mlp": 1.02004635,
      "epoch": 0.9903204473035532,
      "flos": 22559657783040.0,
      "grad_norm": 2.092283584208144,
      "language_loss": 0.84331274,
      "learning_rate": 9.707963763923022e-10,
      "loss": 0.86502469,
      "num_input_tokens_seen": 177698115,
      "step": 8236,
      "time_per_iteration": 2.6692423820495605
    },
    {
      "auxiliary_loss_clip": 0.01121664,
      "auxiliary_loss_mlp": 0.01025262,
      "balance_loss_clip": 1.04598355,
      "balance_loss_mlp": 1.01766825,
      "epoch": 0.9904406901941922,
      "flos": 16144001775360.0,
      "grad_norm": 1.700838164864923,
      "language_loss": 0.78781056,
      "learning_rate": 9.466800567854427e-10,
      "loss": 0.8092798,
      "num_input_tokens_seen": 177716715,
      "step": 8237,
      "time_per_iteration": 2.748806953430176
    },
    {
      "auxiliary_loss_clip": 0.01116018,
      "auxiliary_loss_mlp": 0.01028926,
      "balance_loss_clip": 1.04391909,
      "balance_loss_mlp": 1.0205934,
      "epoch": 0.9905609330848314,
      "flos": 26651499408000.0,
      "grad_norm": 2.3733585282430183,
      "language_loss": 0.67985612,
      "learning_rate": 9.228669919778553e-10,
      "loss": 0.70130557,
      "num_input_tokens_seen": 177735640,
      "step": 8238,
      "time_per_iteration": 2.8283326625823975
    },
    {
      "auxiliary_loss_clip": 0.01130581,
      "auxiliary_loss_mlp": 0.01030738,
      "balance_loss_clip": 1.04870725,
      "balance_loss_mlp": 1.02298987,
      "epoch": 0.9906811759754705,
      "flos": 23111820627840.0,
      "grad_norm": 2.3652439213238434,
      "language_loss": 0.79753852,
      "learning_rate": 8.993571855817617e-10,
      "loss": 0.81915176,
      "num_input_tokens_seen": 177754470,
      "step": 8239,
      "time_per_iteration": 2.7191550731658936
    },
    {
      "auxiliary_loss_clip": 0.0114059,
      "auxiliary_loss_mlp": 0.01025572,
      "balance_loss_clip": 1.04484129,
      "balance_loss_mlp": 1.01754951,
      "epoch": 0.9908014188661095,
      "flos": 22090593052800.0,
      "grad_norm": 1.942044553239686,
      "language_loss": 0.75366014,
      "learning_rate": 8.761506411638642e-10,
      "loss": 0.77532178,
      "num_input_tokens_seen": 177773935,
      "step": 8240,
      "time_per_iteration": 2.762066125869751
    },
    {
      "auxiliary_loss_clip": 0.011105,
      "auxiliary_loss_mlp": 0.00762223,
      "balance_loss_clip": 1.04154277,
      "balance_loss_mlp": 1.00047469,
      "epoch": 0.9909216617567487,
      "flos": 19242948677760.0,
      "grad_norm": 3.337580931308121,
      "language_loss": 0.7404241,
      "learning_rate": 8.53247362244236e-10,
      "loss": 0.75915134,
      "num_input_tokens_seen": 177792745,
      "step": 8241,
      "time_per_iteration": 2.757411003112793
    },
    {
      "auxiliary_loss_clip": 0.01148672,
      "auxiliary_loss_mlp": 0.01026351,
      "balance_loss_clip": 1.0494287,
      "balance_loss_mlp": 1.01856411,
      "epoch": 0.9910419046473877,
      "flos": 23621213352960.0,
      "grad_norm": 2.076742650491605,
      "language_loss": 0.68554592,
      "learning_rate": 8.306473522976532e-10,
      "loss": 0.70729613,
      "num_input_tokens_seen": 177812150,
      "step": 8242,
      "time_per_iteration": 2.758143424987793
    },
    {
      "auxiliary_loss_clip": 0.01128362,
      "auxiliary_loss_mlp": 0.01026881,
      "balance_loss_clip": 1.04494858,
      "balance_loss_mlp": 1.01920354,
      "epoch": 0.9911621475380268,
      "flos": 22711380831360.0,
      "grad_norm": 2.839574497375616,
      "language_loss": 0.71721774,
      "learning_rate": 8.083506147522623e-10,
      "loss": 0.73877025,
      "num_input_tokens_seen": 177831545,
      "step": 8243,
      "time_per_iteration": 3.6725921630859375
    },
    {
      "auxiliary_loss_clip": 0.01107437,
      "auxiliary_loss_mlp": 0.01025364,
      "balance_loss_clip": 1.04538751,
      "balance_loss_mlp": 1.01813352,
      "epoch": 0.991282390428666,
      "flos": 13516956777600.0,
      "grad_norm": 2.741679462919497,
      "language_loss": 0.85421401,
      "learning_rate": 7.863571529906909e-10,
      "loss": 0.87554199,
      "num_input_tokens_seen": 177847130,
      "step": 8244,
      "time_per_iteration": 2.8023455142974854
    },
    {
      "auxiliary_loss_clip": 0.01061467,
      "auxiliary_loss_mlp": 0.00999478,
      "balance_loss_clip": 1.00781906,
      "balance_loss_mlp": 0.99857807,
      "epoch": 0.991402633319305,
      "flos": 61830492071040.0,
      "grad_norm": 0.7270371024360169,
      "language_loss": 0.59613013,
      "learning_rate": 7.646669703489372e-10,
      "loss": 0.61673957,
      "num_input_tokens_seen": 177911440,
      "step": 8245,
      "time_per_iteration": 4.1994171142578125
    },
    {
      "auxiliary_loss_clip": 0.0114026,
      "auxiliary_loss_mlp": 0.01028345,
      "balance_loss_clip": 1.04733348,
      "balance_loss_mlp": 1.02005339,
      "epoch": 0.9915228762099441,
      "flos": 18770148933120.0,
      "grad_norm": 2.5484447131892254,
      "language_loss": 0.57770079,
      "learning_rate": 7.432800701177023e-10,
      "loss": 0.59938681,
      "num_input_tokens_seen": 177929440,
      "step": 8246,
      "time_per_iteration": 2.6808555126190186
    },
    {
      "auxiliary_loss_clip": 0.01035139,
      "auxiliary_loss_mlp": 0.01002603,
      "balance_loss_clip": 1.00850439,
      "balance_loss_mlp": 1.00159001,
      "epoch": 0.9916431191005832,
      "flos": 65936660244480.0,
      "grad_norm": 0.793923050597551,
      "language_loss": 0.5770812,
      "learning_rate": 7.221964555415017e-10,
      "loss": 0.5974586,
      "num_input_tokens_seen": 177989100,
      "step": 8247,
      "time_per_iteration": 3.251948595046997
    },
    {
      "auxiliary_loss_clip": 0.01141079,
      "auxiliary_loss_mlp": 0.01022015,
      "balance_loss_clip": 1.04504645,
      "balance_loss_mlp": 1.01438606,
      "epoch": 0.9917633619912223,
      "flos": 16581573256320.0,
      "grad_norm": 1.9677412584324814,
      "language_loss": 0.75244796,
      "learning_rate": 7.01416129818222e-10,
      "loss": 0.77407885,
      "num_input_tokens_seen": 178006720,
      "step": 8248,
      "time_per_iteration": 2.72245717048645
    },
    {
      "auxiliary_loss_clip": 0.011316,
      "auxiliary_loss_mlp": 0.01028309,
      "balance_loss_clip": 1.04725051,
      "balance_loss_mlp": 1.02042353,
      "epoch": 0.9918836048818613,
      "flos": 25411108999680.0,
      "grad_norm": 10.851455514577507,
      "language_loss": 0.58849502,
      "learning_rate": 6.809390961006745e-10,
      "loss": 0.61009419,
      "num_input_tokens_seen": 178026850,
      "step": 8249,
      "time_per_iteration": 2.7571780681610107
    },
    {
      "auxiliary_loss_clip": 0.01143657,
      "auxiliary_loss_mlp": 0.01027584,
      "balance_loss_clip": 1.04975533,
      "balance_loss_mlp": 1.01944232,
      "epoch": 0.9920038477725005,
      "flos": 25046867134080.0,
      "grad_norm": 1.8581692024387693,
      "language_loss": 0.68489325,
      "learning_rate": 6.607653574948191e-10,
      "loss": 0.70660567,
      "num_input_tokens_seen": 178047630,
      "step": 8250,
      "time_per_iteration": 3.7045044898986816
    },
    {
      "auxiliary_loss_clip": 0.01169748,
      "auxiliary_loss_mlp": 0.01025688,
      "balance_loss_clip": 1.04969537,
      "balance_loss_mlp": 1.01858282,
      "epoch": 0.9921240906631396,
      "flos": 21829773421440.0,
      "grad_norm": 2.6742263067725056,
      "language_loss": 0.81909263,
      "learning_rate": 6.408949170613187e-10,
      "loss": 0.84104699,
      "num_input_tokens_seen": 178066895,
      "step": 8251,
      "time_per_iteration": 2.6755013465881348
    },
    {
      "auxiliary_loss_clip": 0.01137498,
      "auxiliary_loss_mlp": 0.00762331,
      "balance_loss_clip": 1.04714203,
      "balance_loss_mlp": 1.00049925,
      "epoch": 0.9922443335537786,
      "flos": 24864225454080.0,
      "grad_norm": 3.1718488674271015,
      "language_loss": 0.81787378,
      "learning_rate": 6.213277778144288e-10,
      "loss": 0.8368721,
      "num_input_tokens_seen": 178088540,
      "step": 8252,
      "time_per_iteration": 2.7657735347747803
    },
    {
      "auxiliary_loss_clip": 0.01128578,
      "auxiliary_loss_mlp": 0.01036489,
      "balance_loss_clip": 1.04355216,
      "balance_loss_mlp": 1.02842498,
      "epoch": 0.9923645764444178,
      "flos": 21613088626560.0,
      "grad_norm": 2.2693195264723602,
      "language_loss": 0.6670779,
      "learning_rate": 6.020639427224416e-10,
      "loss": 0.68872857,
      "num_input_tokens_seen": 178106185,
      "step": 8253,
      "time_per_iteration": 2.707207441329956
    },
    {
      "auxiliary_loss_clip": 0.01132834,
      "auxiliary_loss_mlp": 0.00761399,
      "balance_loss_clip": 1.04789495,
      "balance_loss_mlp": 1.00049007,
      "epoch": 0.9924848193350568,
      "flos": 25001798544000.0,
      "grad_norm": 2.464359637004191,
      "language_loss": 0.72944462,
      "learning_rate": 5.831034147076864e-10,
      "loss": 0.74838698,
      "num_input_tokens_seen": 178123435,
      "step": 8254,
      "time_per_iteration": 2.7436957359313965
    },
    {
      "auxiliary_loss_clip": 0.01040628,
      "auxiliary_loss_mlp": 0.0099977,
      "balance_loss_clip": 1.00710988,
      "balance_loss_mlp": 0.99888748,
      "epoch": 0.9926050622256959,
      "flos": 68912543151360.0,
      "grad_norm": 0.8649078127665573,
      "language_loss": 0.55693638,
      "learning_rate": 5.644461966463065e-10,
      "loss": 0.57734036,
      "num_input_tokens_seen": 178191045,
      "step": 8255,
      "time_per_iteration": 3.35475754737854
    },
    {
      "auxiliary_loss_clip": 0.01155002,
      "auxiliary_loss_mlp": 0.0102565,
      "balance_loss_clip": 1.05093026,
      "balance_loss_mlp": 1.0178833,
      "epoch": 0.9927253051163349,
      "flos": 20923675914240.0,
      "grad_norm": 2.8159026148295876,
      "language_loss": 0.75748432,
      "learning_rate": 5.460922913687049e-10,
      "loss": 0.7792908,
      "num_input_tokens_seen": 178210135,
      "step": 8256,
      "time_per_iteration": 2.699845314025879
    },
    {
      "auxiliary_loss_clip": 0.01160487,
      "auxiliary_loss_mlp": 0.00762664,
      "balance_loss_clip": 1.04918265,
      "balance_loss_mlp": 1.00042987,
      "epoch": 0.9928455480069741,
      "flos": 22308211601280.0,
      "grad_norm": 2.0919291679375225,
      "language_loss": 0.75623852,
      "learning_rate": 5.280417016593208e-10,
      "loss": 0.77547002,
      "num_input_tokens_seen": 178229925,
      "step": 8257,
      "time_per_iteration": 2.664154052734375
    },
    {
      "auxiliary_loss_clip": 0.01129862,
      "auxiliary_loss_mlp": 0.01030453,
      "balance_loss_clip": 1.04972696,
      "balance_loss_mlp": 1.02318168,
      "epoch": 0.9929657908976132,
      "flos": 17383889393280.0,
      "grad_norm": 1.8944248604243212,
      "language_loss": 0.74732882,
      "learning_rate": 5.102944302559642e-10,
      "loss": 0.76893198,
      "num_input_tokens_seen": 178247420,
      "step": 8258,
      "time_per_iteration": 2.734828472137451
    },
    {
      "auxiliary_loss_clip": 0.0114615,
      "auxiliary_loss_mlp": 0.01026526,
      "balance_loss_clip": 1.0502708,
      "balance_loss_mlp": 1.01849103,
      "epoch": 0.9930860337882522,
      "flos": 22674680110080.0,
      "grad_norm": 2.6154724721373115,
      "language_loss": 0.80046248,
      "learning_rate": 4.9285047985137e-10,
      "loss": 0.82218927,
      "num_input_tokens_seen": 178266840,
      "step": 8259,
      "time_per_iteration": 2.738621711730957
    },
    {
      "auxiliary_loss_clip": 0.0114214,
      "auxiliary_loss_mlp": 0.01040825,
      "balance_loss_clip": 1.04913259,
      "balance_loss_mlp": 1.03325582,
      "epoch": 0.9932062766788914,
      "flos": 28147789284480.0,
      "grad_norm": 1.973400379659921,
      "language_loss": 0.74605596,
      "learning_rate": 4.757098530916436e-10,
      "loss": 0.76788568,
      "num_input_tokens_seen": 178287285,
      "step": 8260,
      "time_per_iteration": 3.630077838897705
    },
    {
      "auxiliary_loss_clip": 0.01148753,
      "auxiliary_loss_mlp": 0.00762301,
      "balance_loss_clip": 1.04935503,
      "balance_loss_mlp": 1.0004586,
      "epoch": 0.9933265195695304,
      "flos": 20156659868160.0,
      "grad_norm": 4.629888988089015,
      "language_loss": 0.77456641,
      "learning_rate": 4.5887255257670563e-10,
      "loss": 0.79367697,
      "num_input_tokens_seen": 178304325,
      "step": 8261,
      "time_per_iteration": 2.6712496280670166
    },
    {
      "auxiliary_loss_clip": 0.01147886,
      "auxiliary_loss_mlp": 0.00761622,
      "balance_loss_clip": 1.04479122,
      "balance_loss_mlp": 1.00046575,
      "epoch": 0.9934467624601695,
      "flos": 21362037494400.0,
      "grad_norm": 2.8864204844170622,
      "language_loss": 0.77122504,
      "learning_rate": 4.4233858086117906e-10,
      "loss": 0.7903201,
      "num_input_tokens_seen": 178322850,
      "step": 8262,
      "time_per_iteration": 2.6940159797668457
    },
    {
      "auxiliary_loss_clip": 0.01125428,
      "auxiliary_loss_mlp": 0.00761543,
      "balance_loss_clip": 1.04709554,
      "balance_loss_mlp": 1.00045085,
      "epoch": 0.9935670053508087,
      "flos": 19756040503680.0,
      "grad_norm": 2.1722979551606154,
      "language_loss": 0.68150294,
      "learning_rate": 4.261079404528356e-10,
      "loss": 0.7003727,
      "num_input_tokens_seen": 178342330,
      "step": 8263,
      "time_per_iteration": 2.806974411010742
    },
    {
      "auxiliary_loss_clip": 0.01135558,
      "auxiliary_loss_mlp": 0.01030767,
      "balance_loss_clip": 1.04902434,
      "balance_loss_mlp": 1.02267826,
      "epoch": 0.9936872482414477,
      "flos": 21978838863360.0,
      "grad_norm": 1.91995933102815,
      "language_loss": 0.6945731,
      "learning_rate": 4.1018063381437205e-10,
      "loss": 0.71623641,
      "num_input_tokens_seen": 178362715,
      "step": 8264,
      "time_per_iteration": 2.730330228805542
    },
    {
      "auxiliary_loss_clip": 0.01044714,
      "auxiliary_loss_mlp": 0.01000684,
      "balance_loss_clip": 1.00792301,
      "balance_loss_mlp": 0.99973601,
      "epoch": 0.9938074911320868,
      "flos": 69810667839360.0,
      "grad_norm": 0.8660945860054986,
      "language_loss": 0.61013985,
      "learning_rate": 3.9455666336141167e-10,
      "loss": 0.6305939,
      "num_input_tokens_seen": 178426495,
      "step": 8265,
      "time_per_iteration": 3.340799570083618
    },
    {
      "auxiliary_loss_clip": 0.01146546,
      "auxiliary_loss_mlp": 0.00761943,
      "balance_loss_clip": 1.04638052,
      "balance_loss_mlp": 1.00049663,
      "epoch": 0.9939277340227259,
      "flos": 15084170058240.0,
      "grad_norm": 2.5952857340552438,
      "language_loss": 0.83445436,
      "learning_rate": 3.7923603146450267e-10,
      "loss": 0.85353923,
      "num_input_tokens_seen": 178442555,
      "step": 8266,
      "time_per_iteration": 2.6309051513671875
    },
    {
      "auxiliary_loss_clip": 0.01152348,
      "auxiliary_loss_mlp": 0.01024916,
      "balance_loss_clip": 1.04661965,
      "balance_loss_mlp": 1.0165118,
      "epoch": 0.994047976913365,
      "flos": 17712364291200.0,
      "grad_norm": 2.8773075942424224,
      "language_loss": 0.80812418,
      "learning_rate": 3.642187404473418e-10,
      "loss": 0.82989687,
      "num_input_tokens_seen": 178460715,
      "step": 8267,
      "time_per_iteration": 2.6796696186065674
    },
    {
      "auxiliary_loss_clip": 0.0114535,
      "auxiliary_loss_mlp": 0.01025619,
      "balance_loss_clip": 1.04572725,
      "balance_loss_mlp": 1.01781702,
      "epoch": 0.994168219804004,
      "flos": 19171558396800.0,
      "grad_norm": 3.2323629152829167,
      "language_loss": 0.86302656,
      "learning_rate": 3.495047925885508e-10,
      "loss": 0.88473624,
      "num_input_tokens_seen": 178479050,
      "step": 8268,
      "time_per_iteration": 3.7096152305603027
    },
    {
      "auxiliary_loss_clip": 0.01153025,
      "auxiliary_loss_mlp": 0.01028553,
      "balance_loss_clip": 1.05171657,
      "balance_loss_mlp": 1.02041101,
      "epoch": 0.9942884626946432,
      "flos": 17851589406720.0,
      "grad_norm": 2.369320842250262,
      "language_loss": 0.82911104,
      "learning_rate": 3.350941901199e-10,
      "loss": 0.85092682,
      "num_input_tokens_seen": 178495970,
      "step": 8269,
      "time_per_iteration": 2.6369855403900146
    },
    {
      "auxiliary_loss_clip": 0.01108979,
      "auxiliary_loss_mlp": 0.0076204,
      "balance_loss_clip": 1.04323578,
      "balance_loss_mlp": 1.00048256,
      "epoch": 0.9944087055852823,
      "flos": 18796578364800.0,
      "grad_norm": 4.182995950981083,
      "language_loss": 0.83834004,
      "learning_rate": 3.2098693522764066e-10,
      "loss": 0.85705018,
      "num_input_tokens_seen": 178509170,
      "step": 8270,
      "time_per_iteration": 3.702590227127075
    },
    {
      "auxiliary_loss_clip": 0.0114564,
      "auxiliary_loss_mlp": 0.01024796,
      "balance_loss_clip": 1.04589295,
      "balance_loss_mlp": 1.01664186,
      "epoch": 0.9945289484759213,
      "flos": 20996969616000.0,
      "grad_norm": 1.8288151747781423,
      "language_loss": 0.81162822,
      "learning_rate": 3.071830300516165e-10,
      "loss": 0.83333254,
      "num_input_tokens_seen": 178527000,
      "step": 8271,
      "time_per_iteration": 2.625684976577759
    },
    {
      "auxiliary_loss_clip": 0.01161962,
      "auxiliary_loss_mlp": 0.01027686,
      "balance_loss_clip": 1.04971743,
      "balance_loss_mlp": 1.01894164,
      "epoch": 0.9946491913665605,
      "flos": 14756952136320.0,
      "grad_norm": 2.170943140779574,
      "language_loss": 0.71077681,
      "learning_rate": 2.9368247668615234e-10,
      "loss": 0.73267329,
      "num_input_tokens_seen": 178545590,
      "step": 8272,
      "time_per_iteration": 2.6745641231536865
    },
    {
      "auxiliary_loss_clip": 0.01123153,
      "auxiliary_loss_mlp": 0.01030262,
      "balance_loss_clip": 1.05000925,
      "balance_loss_mlp": 1.02157187,
      "epoch": 0.9947694342571995,
      "flos": 12669931186560.0,
      "grad_norm": 5.601844385162072,
      "language_loss": 0.61571974,
      "learning_rate": 2.804852771789434e-10,
      "loss": 0.63725388,
      "num_input_tokens_seen": 178558890,
      "step": 8273,
      "time_per_iteration": 2.7334213256835938
    },
    {
      "auxiliary_loss_clip": 0.01158528,
      "auxiliary_loss_mlp": 0.01022277,
      "balance_loss_clip": 1.04732013,
      "balance_loss_mlp": 1.01498115,
      "epoch": 0.9948896771478386,
      "flos": 18843442634880.0,
      "grad_norm": 2.107258699335167,
      "language_loss": 0.56058288,
      "learning_rate": 2.675914335321661e-10,
      "loss": 0.5823909,
      "num_input_tokens_seen": 178577645,
      "step": 8274,
      "time_per_iteration": 2.639145851135254
    },
    {
      "auxiliary_loss_clip": 0.0114207,
      "auxiliary_loss_mlp": 0.01034874,
      "balance_loss_clip": 1.04675293,
      "balance_loss_mlp": 1.02664268,
      "epoch": 0.9950099200384778,
      "flos": 24900207903360.0,
      "grad_norm": 2.385188059762813,
      "language_loss": 0.79246885,
      "learning_rate": 2.550009477018111e-10,
      "loss": 0.81423831,
      "num_input_tokens_seen": 178596415,
      "step": 8275,
      "time_per_iteration": 2.684877395629883
    },
    {
      "auxiliary_loss_clip": 0.011358,
      "auxiliary_loss_mlp": 0.01029974,
      "balance_loss_clip": 1.04787779,
      "balance_loss_mlp": 1.02186751,
      "epoch": 0.9951301629291168,
      "flos": 23733613987200.0,
      "grad_norm": 2.2004824020094764,
      "language_loss": 0.63395017,
      "learning_rate": 2.4271382159790634e-10,
      "loss": 0.65560794,
      "num_input_tokens_seen": 178613845,
      "step": 8276,
      "time_per_iteration": 3.7891712188720703
    },
    {
      "auxiliary_loss_clip": 0.01148213,
      "auxiliary_loss_mlp": 0.01033906,
      "balance_loss_clip": 1.04995739,
      "balance_loss_mlp": 1.02606821,
      "epoch": 0.9952504058197559,
      "flos": 22236893147520.0,
      "grad_norm": 1.9149203354273063,
      "language_loss": 0.85772216,
      "learning_rate": 2.3073005708429406e-10,
      "loss": 0.87954336,
      "num_input_tokens_seen": 178633490,
      "step": 8277,
      "time_per_iteration": 2.7309725284576416
    },
    {
      "auxiliary_loss_clip": 0.01128683,
      "auxiliary_loss_mlp": 0.01030568,
      "balance_loss_clip": 1.04907596,
      "balance_loss_mlp": 1.02341259,
      "epoch": 0.995370648710395,
      "flos": 21211032718080.0,
      "grad_norm": 3.541302191071049,
      "language_loss": 0.72150356,
      "learning_rate": 2.190496559788535e-10,
      "loss": 0.74309605,
      "num_input_tokens_seen": 178651775,
      "step": 8278,
      "time_per_iteration": 2.7113821506500244
    },
    {
      "auxiliary_loss_clip": 0.01162577,
      "auxiliary_loss_mlp": 0.01035061,
      "balance_loss_clip": 1.04947805,
      "balance_loss_mlp": 1.02715158,
      "epoch": 0.9954908916010341,
      "flos": 14866731077760.0,
      "grad_norm": 2.9275778001986814,
      "language_loss": 0.76686585,
      "learning_rate": 2.0767262005372265e-10,
      "loss": 0.78884214,
      "num_input_tokens_seen": 178669290,
      "step": 8279,
      "time_per_iteration": 2.737229347229004
    },
    {
      "auxiliary_loss_clip": 0.01119328,
      "auxiliary_loss_mlp": 0.01038011,
      "balance_loss_clip": 1.04278398,
      "balance_loss_mlp": 1.02986312,
      "epoch": 0.9956111344916732,
      "flos": 19208259118080.0,
      "grad_norm": 2.2831098152650324,
      "language_loss": 0.75376487,
      "learning_rate": 1.965989510346322e-10,
      "loss": 0.77533829,
      "num_input_tokens_seen": 178688410,
      "step": 8280,
      "time_per_iteration": 2.8156986236572266
    },
    {
      "auxiliary_loss_clip": 0.01147988,
      "auxiliary_loss_mlp": 0.01028607,
      "balance_loss_clip": 1.04612827,
      "balance_loss_mlp": 1.02018499,
      "epoch": 0.9957313773823123,
      "flos": 20047060494720.0,
      "grad_norm": 2.531179092577327,
      "language_loss": 0.71519005,
      "learning_rate": 1.8582865060134955e-10,
      "loss": 0.73695594,
      "num_input_tokens_seen": 178706600,
      "step": 8281,
      "time_per_iteration": 2.610072374343872
    },
    {
      "auxiliary_loss_clip": 0.01027309,
      "auxiliary_loss_mlp": 0.01000015,
      "balance_loss_clip": 1.0087688,
      "balance_loss_mlp": 0.99903107,
      "epoch": 0.9958516202729514,
      "flos": 57483253768320.0,
      "grad_norm": 0.7795807519012806,
      "language_loss": 0.55685174,
      "learning_rate": 1.7536172038790098e-10,
      "loss": 0.57712501,
      "num_input_tokens_seen": 178766910,
      "step": 8282,
      "time_per_iteration": 3.4357728958129883
    },
    {
      "auxiliary_loss_clip": 0.01171248,
      "auxiliary_loss_mlp": 0.01026323,
      "balance_loss_clip": 1.05056918,
      "balance_loss_mlp": 1.01892006,
      "epoch": 0.9959718631635904,
      "flos": 27782900974080.0,
      "grad_norm": 2.311550701967263,
      "language_loss": 0.69464618,
      "learning_rate": 1.651981619819054e-10,
      "loss": 0.71662188,
      "num_input_tokens_seen": 178784060,
      "step": 8283,
      "time_per_iteration": 2.593916177749634
    },
    {
      "auxiliary_loss_clip": 0.01145236,
      "auxiliary_loss_mlp": 0.01027834,
      "balance_loss_clip": 1.04698467,
      "balance_loss_mlp": 1.01990628,
      "epoch": 0.9960921060542296,
      "flos": 24024095274240.0,
      "grad_norm": 2.7271326751617724,
      "language_loss": 0.71279681,
      "learning_rate": 1.5533797692546257e-10,
      "loss": 0.73452753,
      "num_input_tokens_seen": 178802795,
      "step": 8284,
      "time_per_iteration": 2.69525408744812
    },
    {
      "auxiliary_loss_clip": 0.01121999,
      "auxiliary_loss_mlp": 0.01031964,
      "balance_loss_clip": 1.04444087,
      "balance_loss_mlp": 1.02337468,
      "epoch": 0.9962123489448687,
      "flos": 18697393935360.0,
      "grad_norm": 2.2593915521421124,
      "language_loss": 0.84364647,
      "learning_rate": 1.4578116671404296e-10,
      "loss": 0.8651861,
      "num_input_tokens_seen": 178821075,
      "step": 8285,
      "time_per_iteration": 2.7116377353668213
    },
    {
      "auxiliary_loss_clip": 0.01126521,
      "auxiliary_loss_mlp": 0.01029958,
      "balance_loss_clip": 1.04622054,
      "balance_loss_mlp": 1.02251947,
      "epoch": 0.9963325918355077,
      "flos": 20010754823040.0,
      "grad_norm": 2.1916745000744036,
      "language_loss": 0.71552825,
      "learning_rate": 1.3652773279759777e-10,
      "loss": 0.73709297,
      "num_input_tokens_seen": 178837725,
      "step": 8286,
      "time_per_iteration": 3.659885883331299
    },
    {
      "auxiliary_loss_clip": 0.01160522,
      "auxiliary_loss_mlp": 0.01028652,
      "balance_loss_clip": 1.05137539,
      "balance_loss_mlp": 1.01982439,
      "epoch": 0.9964528347261468,
      "flos": 33108488991360.0,
      "grad_norm": 1.702476677430473,
      "language_loss": 0.63097072,
      "learning_rate": 1.2757767657989305e-10,
      "loss": 0.65286243,
      "num_input_tokens_seen": 178861515,
      "step": 8287,
      "time_per_iteration": 2.812451124191284
    },
    {
      "auxiliary_loss_clip": 0.01139705,
      "auxiliary_loss_mlp": 0.01027614,
      "balance_loss_clip": 1.04600358,
      "balance_loss_mlp": 1.02034259,
      "epoch": 0.9965730776167859,
      "flos": 23109342589440.0,
      "grad_norm": 1.8808577723999615,
      "language_loss": 0.87393963,
      "learning_rate": 1.1893099941850948e-10,
      "loss": 0.89561284,
      "num_input_tokens_seen": 178880410,
      "step": 8288,
      "time_per_iteration": 2.6598575115203857
    },
    {
      "auxiliary_loss_clip": 0.01118497,
      "auxiliary_loss_mlp": 0.01028592,
      "balance_loss_clip": 1.0412612,
      "balance_loss_mlp": 1.02038455,
      "epoch": 0.996693320507425,
      "flos": 22965843755520.0,
      "grad_norm": 1.9336541775957778,
      "language_loss": 0.77646613,
      "learning_rate": 1.105877026252866e-10,
      "loss": 0.79793704,
      "num_input_tokens_seen": 178898740,
      "step": 8289,
      "time_per_iteration": 2.7661550045013428
    },
    {
      "auxiliary_loss_clip": 0.01149713,
      "auxiliary_loss_mlp": 0.01025099,
      "balance_loss_clip": 1.0478971,
      "balance_loss_mlp": 1.01695681,
      "epoch": 0.996813563398064,
      "flos": 13222740476160.0,
      "grad_norm": 2.130130425020943,
      "language_loss": 0.72176397,
      "learning_rate": 1.0254778746565663e-10,
      "loss": 0.74351209,
      "num_input_tokens_seen": 178914015,
      "step": 8290,
      "time_per_iteration": 2.6203200817108154
    },
    {
      "auxiliary_loss_clip": 0.01127676,
      "auxiliary_loss_mlp": 0.01033548,
      "balance_loss_clip": 1.0466063,
      "balance_loss_mlp": 1.02604377,
      "epoch": 0.9969338062887032,
      "flos": 14647855553280.0,
      "grad_norm": 4.221489991374289,
      "language_loss": 0.73856711,
      "learning_rate": 9.481125515953259e-11,
      "loss": 0.76017934,
      "num_input_tokens_seen": 178932075,
      "step": 8291,
      "time_per_iteration": 2.722611904144287
    },
    {
      "auxiliary_loss_clip": 0.0114817,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.04798412,
      "balance_loss_mlp": 1.01846886,
      "epoch": 0.9970540491793423,
      "flos": 25735741142400.0,
      "grad_norm": 2.047811793553886,
      "language_loss": 0.80168343,
      "learning_rate": 8.737810688064228e-11,
      "loss": 0.82342613,
      "num_input_tokens_seen": 178951910,
      "step": 8292,
      "time_per_iteration": 2.7614550590515137
    },
    {
      "auxiliary_loss_clip": 0.01139459,
      "auxiliary_loss_mlp": 0.0103282,
      "balance_loss_clip": 1.04599369,
      "balance_loss_mlp": 1.02465391,
      "epoch": 0.9971742920699813,
      "flos": 21470236237440.0,
      "grad_norm": 2.1834400488734365,
      "language_loss": 0.79190761,
      "learning_rate": 8.024834375608414e-11,
      "loss": 0.8136304,
      "num_input_tokens_seen": 178970500,
      "step": 8293,
      "time_per_iteration": 2.7641305923461914
    },
    {
      "auxiliary_loss_clip": 0.01029199,
      "auxiliary_loss_mlp": 0.01001416,
      "balance_loss_clip": 1.00966525,
      "balance_loss_mlp": 1.00048637,
      "epoch": 0.9972945349606205,
      "flos": 72211223629440.0,
      "grad_norm": 0.8228503745783473,
      "language_loss": 0.62776768,
      "learning_rate": 7.342196686788149e-11,
      "loss": 0.64807385,
      "num_input_tokens_seen": 179023665,
      "step": 8294,
      "time_per_iteration": 3.96468186378479
    },
    {
      "auxiliary_loss_clip": 0.01121689,
      "auxiliary_loss_mlp": 0.01030355,
      "balance_loss_clip": 1.04726553,
      "balance_loss_mlp": 1.02252305,
      "epoch": 0.9974147778512595,
      "flos": 19678293515520.0,
      "grad_norm": 2.2327325230747244,
      "language_loss": 0.68904126,
      "learning_rate": 6.689897725142834e-11,
      "loss": 0.71056169,
      "num_input_tokens_seen": 179043140,
      "step": 8295,
      "time_per_iteration": 2.7368531227111816
    },
    {
      "auxiliary_loss_clip": 0.01148822,
      "auxiliary_loss_mlp": 0.0102786,
      "balance_loss_clip": 1.04786897,
      "balance_loss_mlp": 1.01998639,
      "epoch": 0.9975350207418986,
      "flos": 15960821391360.0,
      "grad_norm": 2.391548510433361,
      "language_loss": 0.88793182,
      "learning_rate": 6.067937589615545e-11,
      "loss": 0.90969867,
      "num_input_tokens_seen": 179061215,
      "step": 8296,
      "time_per_iteration": 3.6293728351593018
    },
    {
      "auxiliary_loss_clip": 0.01001598,
      "auxiliary_loss_mlp": 0.0100289,
      "balance_loss_clip": 1.00949121,
      "balance_loss_mlp": 1.00196636,
      "epoch": 0.9976552636325378,
      "flos": 59961879768960.0,
      "grad_norm": 0.7421360858374128,
      "language_loss": 0.5765444,
      "learning_rate": 5.476316374575241e-11,
      "loss": 0.59658927,
      "num_input_tokens_seen": 179124700,
      "step": 8297,
      "time_per_iteration": 3.456106662750244
    },
    {
      "auxiliary_loss_clip": 0.01141948,
      "auxiliary_loss_mlp": 0.01028942,
      "balance_loss_clip": 1.04567754,
      "balance_loss_mlp": 1.02006078,
      "epoch": 0.9977755065231768,
      "flos": 22487872452480.0,
      "grad_norm": 2.1683710363941837,
      "language_loss": 0.72299898,
      "learning_rate": 4.9150341697723476e-11,
      "loss": 0.74470794,
      "num_input_tokens_seen": 179144590,
      "step": 8298,
      "time_per_iteration": 3.167578935623169
    },
    {
      "auxiliary_loss_clip": 0.01108335,
      "auxiliary_loss_mlp": 0.01035874,
      "balance_loss_clip": 1.04531646,
      "balance_loss_mlp": 1.02751184,
      "epoch": 0.9978957494138159,
      "flos": 26030280666240.0,
      "grad_norm": 1.694149505602092,
      "language_loss": 0.66766632,
      "learning_rate": 4.384091060338768e-11,
      "loss": 0.68910849,
      "num_input_tokens_seen": 179165060,
      "step": 8299,
      "time_per_iteration": 2.882141351699829
    },
    {
      "auxiliary_loss_clip": 0.0113063,
      "auxiliary_loss_mlp": 0.01029936,
      "balance_loss_clip": 1.04746115,
      "balance_loss_mlp": 1.02210975,
      "epoch": 0.998015992304455,
      "flos": 22637835734400.0,
      "grad_norm": 5.170694532630881,
      "language_loss": 0.74078679,
      "learning_rate": 3.883487126810081e-11,
      "loss": 0.76239252,
      "num_input_tokens_seen": 179184320,
      "step": 8300,
      "time_per_iteration": 2.698993682861328
    },
    {
      "auxiliary_loss_clip": 0.01105134,
      "auxiliary_loss_mlp": 0.01027428,
      "balance_loss_clip": 1.04391038,
      "balance_loss_mlp": 1.01984692,
      "epoch": 0.9981362351950941,
      "flos": 18223444955520.0,
      "grad_norm": 2.134718926710036,
      "language_loss": 0.79382217,
      "learning_rate": 3.41322244516995e-11,
      "loss": 0.81514776,
      "num_input_tokens_seen": 179202265,
      "step": 8301,
      "time_per_iteration": 2.7960033416748047
    },
    {
      "auxiliary_loss_clip": 0.01121124,
      "auxiliary_loss_mlp": 0.0102418,
      "balance_loss_clip": 1.04381669,
      "balance_loss_mlp": 1.01641917,
      "epoch": 0.9982564780857331,
      "flos": 33474095573760.0,
      "grad_norm": 1.7483437625502414,
      "language_loss": 0.63028359,
      "learning_rate": 2.9732970866946925e-11,
      "loss": 0.65173662,
      "num_input_tokens_seen": 179222145,
      "step": 8302,
      "time_per_iteration": 4.03597092628479
    },
    {
      "auxiliary_loss_clip": 0.01150462,
      "auxiliary_loss_mlp": 0.01031258,
      "balance_loss_clip": 1.0503155,
      "balance_loss_mlp": 1.02252626,
      "epoch": 0.9983767209763723,
      "flos": 15523465392000.0,
      "grad_norm": 2.582761971494784,
      "language_loss": 0.78368199,
      "learning_rate": 2.563711118175327e-11,
      "loss": 0.8054992,
      "num_input_tokens_seen": 179239030,
      "step": 8303,
      "time_per_iteration": 2.7375741004943848
    },
    {
      "auxiliary_loss_clip": 0.01145367,
      "auxiliary_loss_mlp": 0.0102328,
      "balance_loss_clip": 1.04823196,
      "balance_loss_mlp": 1.01644087,
      "epoch": 0.9984969638670114,
      "flos": 19974377324160.0,
      "grad_norm": 2.8150003331210938,
      "language_loss": 0.83840704,
      "learning_rate": 2.184464601717728e-11,
      "loss": 0.86009347,
      "num_input_tokens_seen": 179257345,
      "step": 8304,
      "time_per_iteration": 2.7317240238189697
    },
    {
      "auxiliary_loss_clip": 0.0111835,
      "auxiliary_loss_mlp": 0.010349,
      "balance_loss_clip": 1.04880273,
      "balance_loss_mlp": 1.02640676,
      "epoch": 0.9986172067576504,
      "flos": 20375750874240.0,
      "grad_norm": 2.525039661267795,
      "language_loss": 0.77779084,
      "learning_rate": 1.8355575948758585e-11,
      "loss": 0.79932332,
      "num_input_tokens_seen": 179275330,
      "step": 8305,
      "time_per_iteration": 2.771968126296997
    },
    {
      "auxiliary_loss_clip": 0.01160579,
      "auxiliary_loss_mlp": 0.01029334,
      "balance_loss_clip": 1.05061221,
      "balance_loss_mlp": 1.02165151,
      "epoch": 0.9987374496482896,
      "flos": 23727903724800.0,
      "grad_norm": 4.524946434332663,
      "language_loss": 0.73850071,
      "learning_rate": 1.5169901505407424e-11,
      "loss": 0.76039982,
      "num_input_tokens_seen": 179292395,
      "step": 8306,
      "time_per_iteration": 2.6960787773132324
    },
    {
      "auxiliary_loss_clip": 0.01135627,
      "auxiliary_loss_mlp": 0.01025973,
      "balance_loss_clip": 1.04597652,
      "balance_loss_mlp": 1.01827765,
      "epoch": 0.9988576925389286,
      "flos": 25044029959680.0,
      "grad_norm": 2.2663645372205217,
      "language_loss": 0.73986679,
      "learning_rate": 1.228762317073695e-11,
      "loss": 0.76148278,
      "num_input_tokens_seen": 179311225,
      "step": 8307,
      "time_per_iteration": 2.7591724395751953
    },
    {
      "auxiliary_loss_clip": 0.01142725,
      "auxiliary_loss_mlp": 0.01028544,
      "balance_loss_clip": 1.0491178,
      "balance_loss_mlp": 1.02082527,
      "epoch": 0.9989779354295677,
      "flos": 31285627637760.0,
      "grad_norm": 2.5235119649075752,
      "language_loss": 0.79130542,
      "learning_rate": 9.70874138195299e-12,
      "loss": 0.81301814,
      "num_input_tokens_seen": 179333135,
      "step": 8308,
      "time_per_iteration": 2.7712042331695557
    },
    {
      "auxiliary_loss_clip": 0.01141634,
      "auxiliary_loss_mlp": 0.01031973,
      "balance_loss_clip": 1.04540944,
      "balance_loss_mlp": 1.02383733,
      "epoch": 0.9990981783202069,
      "flos": 19573398823680.0,
      "grad_norm": 1.93606674993389,
      "language_loss": 0.7453323,
      "learning_rate": 7.433256530076093e-12,
      "loss": 0.76706839,
      "num_input_tokens_seen": 179353090,
      "step": 8309,
      "time_per_iteration": 2.7855212688446045
    },
    {
      "auxiliary_loss_clip": 0.01124334,
      "auxiliary_loss_mlp": 0.01024948,
      "balance_loss_clip": 1.04772747,
      "balance_loss_mlp": 1.01784611,
      "epoch": 0.9992184212108459,
      "flos": 17199667514880.0,
      "grad_norm": 8.816228836025601,
      "language_loss": 0.75812781,
      "learning_rate": 5.46116896038562e-12,
      "loss": 0.77962059,
      "num_input_tokens_seen": 179367500,
      "step": 8310,
      "time_per_iteration": 2.7161262035369873
    },
    {
      "auxiliary_loss_clip": 0.01139421,
      "auxiliary_loss_mlp": 0.01030107,
      "balance_loss_clip": 1.04478288,
      "balance_loss_mlp": 1.02240062,
      "epoch": 0.999338664101485,
      "flos": 46497853681920.0,
      "grad_norm": 1.9633798284117085,
      "language_loss": 0.62061119,
      "learning_rate": 3.792478972197699e-12,
      "loss": 0.64230645,
      "num_input_tokens_seen": 179388085,
      "step": 8311,
      "time_per_iteration": 2.940264940261841
    },
    {
      "auxiliary_loss_clip": 0.01099515,
      "auxiliary_loss_mlp": 0.00761995,
      "balance_loss_clip": 1.04189146,
      "balance_loss_mlp": 1.00049114,
      "epoch": 0.9994589069921241,
      "flos": 15158253859200.0,
      "grad_norm": 2.364933312915589,
      "language_loss": 0.7020601,
      "learning_rate": 2.4271868181990895e-12,
      "loss": 0.72067511,
      "num_input_tokens_seen": 179405250,
      "step": 8312,
      "time_per_iteration": 3.696516275405884
    },
    {
      "auxiliary_loss_clip": 0.01139956,
      "auxiliary_loss_mlp": 0.00761896,
      "balance_loss_clip": 1.04917789,
      "balance_loss_mlp": 1.00041473,
      "epoch": 0.9995791498827632,
      "flos": 12531460256640.0,
      "grad_norm": 1.9855449343801137,
      "language_loss": 0.81235588,
      "learning_rate": 1.3652927060014973e-12,
      "loss": 0.83137435,
      "num_input_tokens_seen": 179420845,
      "step": 8313,
      "time_per_iteration": 2.657944917678833
    },
    {
      "auxiliary_loss_clip": 0.0114897,
      "auxiliary_loss_mlp": 0.01027402,
      "balance_loss_clip": 1.04634094,
      "balance_loss_mlp": 1.01949286,
      "epoch": 0.9996993927734023,
      "flos": 19245175320960.0,
      "grad_norm": 2.792329342972393,
      "language_loss": 0.64255166,
      "learning_rate": 6.067967965872612e-13,
      "loss": 0.66431534,
      "num_input_tokens_seen": 179440455,
      "step": 8314,
      "time_per_iteration": 2.7076520919799805
    },
    {
      "auxiliary_loss_clip": 0.01163644,
      "auxiliary_loss_mlp": 0.01024777,
      "balance_loss_clip": 1.05096734,
      "balance_loss_mlp": 1.01693285,
      "epoch": 0.9998196356640414,
      "flos": 62952804518400.0,
      "grad_norm": 1.8812410239012418,
      "language_loss": 0.77090532,
      "learning_rate": 1.5169920497548615e-13,
      "loss": 0.79278958,
      "num_input_tokens_seen": 179465075,
      "step": 8315,
      "time_per_iteration": 2.9954769611358643
    },
    {
      "auxiliary_loss_clip": 0.01108227,
      "auxiliary_loss_mlp": 0.0101639,
      "balance_loss_clip": 1.02721989,
      "balance_loss_mlp": 1.01134562,
      "epoch": 0.9999398785546805,
      "flos": 50922375073920.0,
      "grad_norm": 1.1005675530923773,
      "language_loss": 0.54815292,
      "learning_rate": 0.0,
      "loss": 0.56939906,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "time_per_iteration": 3.2042698860168457
    },
    {
      "epoch": 0.9999398785546805,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "total_flos": 6.996749092776837e+17,
      "train_loss": 0.7906841054995969,
      "train_runtime": 25514.2775,
      "train_samples_per_second": 13.038,
      "train_steps_per_second": 0.326
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 8316,
  "num_input_tokens_seen": 179513955,
  "num_train_epochs": 1,
  "save_steps": 1664,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.996749092776837e+17,
  "train_batch_size": 5,
  "trial_name": null,
  "trial_params": null
}